FIELD OF THE INVENTION This invention is in the field of microbiology. More specifically, this invention pertains to a method for microbial production of aryl-carotenoid compounds.

BACKGROUND OF THE INVENTION Carotenoids are pigments that are ubiquitous throughout nature and synthesized by all photosynthetic organisms, and in some heterotrophic growing bacteria and fungi. Carotenoids provide color for flowers, vegetables, insects, fish and birds. Colors of carotenoids range from yellow to red with variations of brown and purple. As precursors of vitamin A, carotenoids are fundamental components in our diet and they play an important role in human health. Industrial uses of carotenoids include pharmaceuticals, food supplements, animal feed additives, and colorants in cosmetics, to mention a few.

Because animals are unable to synthesize carotenoids de novo, they must obtain them by dietary means. Thus, manipulation of carotenoid production and composition in plants or bacteria can provide new or improved sources for carotenoids.

Carotenoids come in many different forms and chemical structures.

Most naturally occurring carotenoids are hydrophobic tetraterpenoids containing a C40 methyl-branched hydrocarbon backbone derived from successive condensation of eight C5 isoprene units (IPP). In addition, novel carotenoids with longer or shorter backbones occur in some species of nonphotosynthetic bacteria. Carotenoids exhibit great variations and may be cyclic, monocyclic, or bicyclic depending on whether the ends of the hydrocarbon backbones have been cyclized to yield aliphatic or cyclic ring structures (G. Armstrong, (1999) In Comprehensive Natural Products Chemistry, Elsevier Press, volume 2, pp 321-352).

Carotenoid biosynthesis starts with the isoprenoid pathway to generate the C5 isoprene unit, isopentenyl pyrophosphate (IPP). fPP is then condensed with its isomer dimethylallyl pyrophosphate (DMAPP) to generate the C10 geranyl pyrophosphate (GPP) which is then elongated to form the C15 farnesyl pyrophosphate (FPP). FPP synthesis is common in both carotenogenic and non-carotenogenic bacteria. Additional enzymes

in the carotenoid pathway are able to then generate carotenoid pigments from the FPP precursor, segregating into two categories : (i) carotene backbone synthesis enzymes and (ii) subsequent modification enzymes.

The backbone synthesis enzymes include geranyl geranyl pyrophosphate synthase, phytoene synthase, phytoene dehydrogenase, and lycopene cyclase, etc. The modification enzymes include ketolases, hydroxylases, dehydratases, glycosylases, etc.

It is known that ß-carotene can be converted to isorenieratene by the carotene desaturase. The crû gene, encoding the carotene desaturase, has been identified in a few actinomycetes including Streptomyces, Mycobacterium and Brevibacterium (Krugel et al., Biochimica et Biophysica Acta, 1439: 57-64 (1999); Krubasik and Sandmann, Mol Gen Genet, 263: 423-432 (2000); and Viveiros et al., FEMS Microbiol Lett, 187: 95-101 (2000)). Another aryl-carotene, chlorobactene, was reported in photosynthetic green bacteria (Liaaen- Jensen et al., Acta Chem. Scand, 18: 1703-1718 (1964); Takaichi et al., Arch Microbiol, 168: 270-276 (1997) ). Recent genomic sequencing of Chlorobium tepidum identified a putative carotene desaturase gene (Eisen et al., PNAS USA, 99: 9509-9514 (2002), which might be responsible for the synthesis of the native chlorobactene and derivatives. However, function of the putative carotene desaturase gene from Chlorobium has not yet been determined. It is likely that the CrtU from actinomycetes might also act on other substrates in addition to ß-carotene to produce a variety of aryl-carotenoids, such as converting y-carotene to chlorobactene. However, previous attempts to express cffU in heterologous hosts have not been successful (Schumann et al., Mol Gen Genet, 252: 658-666 (1996) ). The inability to express the carotene desaturase in a heterologous host is unfortunate and presents a significant hurdle to the synthesis of a variety of aryl-carotenoids by genetic engineering. Furthermore, natural aryl-carotenoids are always present as mixtures of the aryl-carotenoid with their precursors or derivatives (Kohl et al., Phytochemistry, 22: 207-213 (1983); Takaichi et al., supra). Production of a pure aryl-carotenoid requires the ability to express the carotene desaturase in a heterologous host.

The problem to be solved is to express a functional carotene desaturase (crû) gene for the production of aryl-carotenoids in a heterologous host. Applicants have solved the stated problem by isolating

the crtU gene from Brevibacterium linens and expressing the gene from a plasmid in the Rhodococcus erythropolis ATCC 47072 strain.

SUMMARY OF THE INVENTION The invention provides methods for the production of aryl carotenoid compounds by the bioconversion of cyclic carotenoid substrates in the presence of a carotene desaturase. Specifically the invention provides a method for the production of aryl carotenoid compounds comprising: (a) providing a host cell which comprises a cyclic carotenoid having at least one a-ionone ring; (b) transforming the host cell of (a) with a gene encoding a carotene desaturase; and (c) growing the transformed host cell of (b) under conditions whereby an aryl carotenoid is produced.

Additionally the invention provides a method of regulating aryl carotenoid biosynthesis in an host cell comprising: (a) introducing into a host cell a carotene desaturase gene encoding an polypeptide having the amino acid sequence as set forth in SEQ ID NO : 19 under the control of suitable regulatory sequences; and (b) growing the host cell of (a) under conditions whereby the carotene desaturase gene is expressed and aryl carotenoid biosynthesis is regulated.

In a preferred embodiment the invention provides a method for the production of isorenieratene comprising: (a) providing a host cell which comprises (3-carotene ; (b) transforming the host cell of (a) with a gene encoding a carotene desaturase ; and (c) growing the transformed host cell of (b) under conditions whereby an aryl carotenoid is produced.

In another preferred embodiment the invention provides a method for the production of chlorobactene comprising: (a) providing a host cell which comprises y-carotene; (b) transforming the host cell of (a) with a gene encoding a carotene desaturase; and (c) growing the transformed host cell of (b) under conditions whereby an aryl carotenoid is produced.

BRIEF DESCRIPTION OF THE DRAWINGS AND SEQUENCE DESCRIPTIONS Figure 1 illustrates the enzymatic aromatization of carotenoids by CrtU.

Figure 2 illustrates the strategy for synthesis of aryl-carotenoids in Rhodococcus erythropolis ATCC 47072.

Figure 3 is a plasmid map of pDCQ140 with cite expression downstream of dxs gene on a pDA71 based Rhodococcus shuttle vector.

Figure 4 is a plasmid map of pDCQ143 with crtU expression downstream of Cmr gene on a pRhBR171 based Rhodococcus shuttle vector.

Figure 5 describes LC/MS analysis of carotenoids synthesized by recombinant Rhodococcus strains.

Figure 6 is a TLC analysis of carotenoids produced in Rhodococcus strains.

The invention can be more fully understood from the following detailed description and the accompanying sequence descriptions, which for a part of this application.' The following sequences comply with 37 C. F. R. 1.821-1. 825 ("Requirements for Patent Applications Containing Nucleotide Sequences and/or Amino Acid Sequence Disclosures-the Sequence Rules") and are consistent with World Intellectual Property Organization (WIPO) Standard ST. 25 (1998) and the sequence listing requirements of the EPO and PCT (Rules 5.2 and 49.5 (a-bis), and Section 208 and Annex C of the Administrative Instructions). The symbols and format used for nucleotide and amino acid sequence data comply with the rules set forth in 37 C. F. R. §1.822.

SEQ ID NOs : 1-3 are oligonucleotide primers used to amplify 16s rRNA.

SEQ ID NO : 4 is the nucleotide sequence for the 16s rRNA gene from Rhodococcus erythropolis AN 12.

SEQ ID NOs : 5-6 are oligonucleotide primers designed to amplify the cet0 gene.

SEQ ID NO : 7 is the nucleotide sequence for the Rhodococcus erythropolis AN12 crt0 gene.

SEQ ID NO : 8 is the nucleotide sequence for the Rhodococcus erythropolis ATCC 47072 crtO gene.

SEQ ID NOs : 9-10 are oligonucleotide primers designed to amplify pBR328 sequence.

SEQ ID NOs : 11-12 are oligonucleotide primers designed to amplify crtO from ATCC 47072.

SEQ ID NO : 13 is the nucleotide sequence for the pRHBR17 E. coli- Rhodococcus shuttle plasmid.

SEQ ID NO : 14 is the nucleotide sequence for the dxs gene of Rhodococcus erythropolis AN12.

SEQ ID NOs : 15-17 are oligonucleotide primers used to amplify dxs from AN 12.

SEQ ID NO : 18 is the nucleotide sequence of the crû gene from Brevibacterium linens ATCC 9175 with GenBank@ Accession number AF139916.

SEQ ID NO : 19 is the amino acid sequence of CrtU from Brevibacterium linens ATCC 9175 with GenBankd3 Accession number AF139916.

SEQ ID NOs : 20 and 21 are primers used for the amplification of the cite gene.

SEQ ID NO : 22 is the nucleotide sequence of a chloramphenicol resistance marker with the Gen Bank@ Accession number AJ 308376.

SEQ ID NOs : 23 and 24 are primers used to amplify and isolate the chloramphenicol resistance marker as defined by SEQ ID NO : 22.

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for the conversion of cyclic carotenoids having a D-ionone ring to the corresponding aryl carotenoid, via the heterologous expression of a carotene desaturase gene (cato.

The expression of crû in a heterologous host is useful for the production of aryl carotenoids individually, as well as for the regulation and production of other carotenoids in the isoprenoid biosynthetic pathway.

There is a general practical utility for microbial production of carotenoid compounds as these compounds are very difficult to make chemically (Nelis and Leenheer, Appl. Bacteriol., 70: 181-191 (1991)). Introduction of the aromatic ring (s) may possibly render the carotenoids more stable, which would be desired for certain applications such as uses for food colorants In this disclosure, a number of terms and abbreviations are used for the interpretation of the Claims and the specification.

"Open reading frame"is abbreviated ORF.

"Polymerase chain reaction"is abbreviated PCR.

As used herein, an"isolated nucleic acid fragment"is a polymer of RNA or DNA that is single-or double-stranded, optionally containing synthetic, non-natural or altered nucleotide bases. An isolated nucleic acid fragment in the form of a polymer of DNA may be comprised of one or more segments of cDNA, genomic DNA or synthetic DNA.

The terms"isoprenoid"or"terpenoid"refer to the compounds derived from the isoprenoid pathway including 10 carbon terpenoids and their derivatives, such as carotenoids and xanthophylls.

The terms"Rhodococcus erythropolis AN12"or"AN12"are used interchangeably and refer to the Rhodococcus erythropolis AN 12 strain.

The terms"Rhodococcus erythropolis ATCC 47072"or "ATCC 47072"are used interchangeably and refer to the Rhodococcus erythropolis ATCC 47072 strain.

The term"carotenoid"refers to a compound composed of a polyene backbone which is condensed from five-carbon isoprene unit. Carotenoids can be acyclic or terminated with one (monocyclic) or two (bicyclic) cyclic end groups. The term"carotenoid"may include both carotenes and xanthophylls. A"carotene"refers to a hydrocarbon carotenoid. Carotene derivatives that contain one or more oxygen atoms, in the form of hydroxy- , methoxy-, oxo-, epoxy-, carboxy-, or aldehydic functional groups, or within glycosides, glycoside esters, or sulfates, are collectively known as "xanthophylls". Carotenoids that are particularly suitable in the present invention are monocyclic and bicyclic carotenoids.

The term"carotene desaturase"refers to the group of enzymes that can desaturate and transfer methyl or other groups of the p-ionone ring of mono-or bi-cyclic carotenoids. A preferred carotene desaturase for use herein is the crtU isolated from Brevibacterium linens ATCC 9175 and having the amino acid sequence as set forth in SEQ ID NO : 19.

The term"aryl-carotenoid"refers to carotenoids with at least one aromatic end group, including isorenieratene, P-isorenieratene, chlorobactene, and derivates as shown in Figure 1.

The term"percent identity", as known in the art, is a relationship between two or more polypeptide sequences or two or more polynucleotide sequences, as determined by comparing the sequences.

In the art,"identity"also means the degree of sequence relatedness between polypeptide or polynucleotide sequences, as the case may be, as

determined by the match between strings of such sequences."Identity" and"similarity"can be readily calculated by known methods, including but not limited to those described in: Computational Molecular Biology (Lesk, A. M. , ed. ) Oxford University Press, NY (1988); Biocomputing : Informatics and Genome Proiects (Smith, D. W. , ed. ) Academic Press, NY (1993); Computer Analysis of Sequence Data, Part I (Griffin, A. M. , and Griffin, H.

G. , eds. ) Humana Press, NJ (1994); Sequence Analysis in Molecular Biology (von Heinje, G. , ed. ) Academic Press (1987); and Sequence Analysis Primer (Gribskov, M. and Devereux, J. , eds. ) Stockton Press, NY (1991). Preferred methods to determine identity are designed to give the best match between the sequences tested. Methods to determine identity and similarity are codified in publicly available computer programs.

Sequence alignments and percent identity calculations may be performed using the Megalign program of the LASERGENE bioinformatics computing suite (DNASTAR Inc., Madison, WI). Multiple alignment of the sequences was performed using the Clustal method of alignment (Higgins and Sharp (1989) CABIOS. 5: 151-153) with the default parameters (GAP PENALTY=10, GAP LENGTH PENALTY=10). Default parameters for pairwise alignments using the Clustal method were KTUPLE 1, GAP PENALTY=3, WINDOW=5 and DIAGONALS SAVED=5.

"Synthetic genes"can be assembled from oligonucleotide building blocks that are chemically synthesized using procedures known to those skilled in the art. These building blocks are ligated and annealed to form gene segments that are then enzymatically assembled to construct the entire gene."Chemically synthesized", as related to a sequence of DNA, means that the component nucleotides were assembled in vitro. Manual chemical synthesis of DNA may be accomplished using well-established procedures, or automated chemical synthesis can be performed using one of a number of commercially available machines. Accordingly, the genes can be tailored for optimal gene expression based on optimization of nucleotide sequence to reflect the codon bias of the host cell. The skilled artisan appreciates the likelihood of successful gene expression if codon usage is biased towards those codons favored by the host. Determination of preferred codons can be based on a survey of genes derived from the host cell where sequence information is available.

"Gene"refers to a nucleic acid fragment that expresses a specific protein, including regulatory sequences preceding (5'non-coding sequences) and following (3'non-coding sequences) the coding sequence.

"Native gene"refers to a gene as found in nature with its own regulatory sequences. "Chimeric gene"refers to any gene that is not a native gene, comprising regulatory and coding sequences that are not found together in nature. Accordingly, a chimeric gene may comprise regulatory sequences and coding sequences that are derived from different sources, or regulatory sequences and coding sequences derived from the same source, but arranged in a manner different than that found in nature.

"Endogenous gene"refers to a native gene in its natural location in the genome of an organism. A"foreign"gene refers to a gene not normally found in the host organism, but that is introduced into the host organism by gene transfer. Foreign genes can comprise native genes inserted into a non-native organism, or chimeric genes. A"transgene"is a gene that has been introduced into the genome by a transformation procedure.

"Coding sequence"refers to a DNA sequence that codes for a specific amino acid sequence."Suitable regulatory sequences"refer to nucleotide sequences located upstream (5'non-coding sequences), within, or downstream (3'non-coding sequences) of a coding sequence, and which influence the transcription, RNA processing or stability, or translation of the associated coding sequence. Regulator sequences may include promoters, translation leader sequences, introns, polyadenylation recognition sequences, RNA processing site, effector binding site and stem-loop structure.

"Promoter"refers to a DNA sequence capable of controlling the expression of a coding sequence or functional RNA. In general, a coding sequence is located 3'to a promoter sequence. Promoters may be derived in their entirety from a native gene, or be composed of different elements derived from different promoters found in nature, or even comprise synthetic DNA segments. It is understood by those skilled in the art that different promoters may direct the expression of a gene in different tissues or cell types, or at different stages of development, or in response to different environmental or physiological conditions. Promoters which cause a gene to be expressed in most cell types at most times are commonly referred to as"constitutive promoters". It is further recognized that since in most cases the exact boundaries of regulatory sequences have not been completely defined, DNA fragments of different lengths may have identical promoter activity.

The"3'non-coding sequences"refer to DNA sequences located downstream of a coding sequence and include polyadenylation recognition

sequences and other sequences encoding regulatory. signals capable of affecting mRNA processing or gene expression. The polyadenylation signal is usually characterized by affecting the addition of polyadenylic acid tracts to the 3'end of the mRNA precursor.

"RNA transcript"refers to the product resulting from RNA polymerase-catalyzed transcription of a DNA sequence. When the RNA transcript is a perfect complementary copy of the DNA sequence, it is referred to as the primary transcript or it may be a RNA sequence derived from post-transcriptional processing of the primary transcript and is referred to as the mature RNA. "Messenger RNA (mRNA) "refers to the RNA that is without introns and that can be translated into protein by the cell."cDNA"refers to a double-stranded DNA that is complementary to and derived from mRNA."Sense"RNA refers to RNA transcript that includes the mRNA and so can be translated into protein by the cell.

"Antisense RNA"refers to a RNA transcript that is complementary to all or part of a target primary transcript or mRNA and that blocks the expression of a target gene (US 5107065; WO 9928508). The complementarity of an antisense RNA may be with any part of the specific gene transcript, i. e. , at the 5'non-coding sequence, 3'non-coding sequence, or the coding sequence."Functional RNA"refers to antisense RNA, ribozyme RNA, or other RNA that is not translated yet has an effect on cellular processes.

The term"operably linked"refers to the association of nucleic acid sequences on a single nucleic acid fragment so that the function of one is affected by the other. For example, a promoter is operably linked with a coding sequence when it is capable of affecting the expression of that coding sequence (i. e. , that the coding sequence is under the transcriptional control of the promoter). Coding sequences can be operably linked to regulatory sequences in sense or antisense orientation.

The term"expression", as used herein, refers to the transcription and stable accumulation of sense (mRNA) or antisense RNA derived from the nucleic acid fragment of the invention. Expression may also refer to translation of mRNA into a polypeptide.

"Transformation"refers to the transfer of a nucleic acid fragment into the genome of a host organism, resulting in genetically stable inheritance. Host organisms containing the transformed nucleic acid fragments are referred to as"transgenic"or"recombinant"or"transformed" organisms.

The terms"plasmid","vector", and"cassette"refer to an extra chromosomal element often carrying genes which are not part of the central metabolism of the cell, and usually in the form of circular double- stranded DNA fragments. Such elements may be autonomously replicating sequences, genome integrating sequences, phage or nucleotide sequences, linear or circular, of a single-or double-stranded DNA or RNA, derived from any source, in which a number of nucleotide sequences have been joined or recombined into a unique construction which is capable of introducing a promoter fragment and DNA sequence for a selected gene product along with appropriate 3'untranslated sequence into a cell."Transformation cassette"refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that facilitate transformation of a particular host cell."Expression cassette"refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that allow for enhanced expression of that gene in a foreign host.

Standard recombinant DNA and molecular cloning techniques used here are well known in the art and are described by Sambrook, J. , Fritsch, E. F. and Maniatis, T., Molecular Cloning : A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989) (hereinafter"Maniatis"); and by Silhavy, T. J. , Bennan, M. L. and Enquist, L. W., Experiments with Gene Fusions, Cold Spring Harbor Laboratory Cold Press Spring Harbor, NY (1984); and by Ausubel, F. M. et al., Current Protocols in Molecular Biology, published by Greene Publishing Assoc. and Wiley-Interscience (1987).

The present invention provides a method for the in vivo biotransformation of cyclic carotenes having a p-ionone ring to the corresponding aryl carotenoid. The method proceeds by expressing a heterologous gene encoding a carotene desaturase (crtU) enzyme in a host cell that produces a suitable cyclic carotene substrate, where the p- ionone ring is aromatized to produce the corresponding aryl carotenoid.

Carotene Desaturase Activity Biosynthesis of aromatic carotenoids catalyzed by CrtU proceeds by desaturation and methyltransferation of the p-ionone ring of the cyclic carotenoids (Krugel et al., supra). CrtU, expressed in its native host, has been shown to convert p-carotene (two p-ionone rings) to aromatic groups of isorenieratene in Streptomyces griseus, Brevibacterium linens and

Mycobacterium auraum A+ (Krugel et al., supra; Krubasik and Sandmann, supra; Viveiros et al., supra).

A number of carotene desaturases are known and will be suitable in the present invention. For example, carotene desaturase has been identified in Streptomyces avermitilis (GenBanko Accession No. AB070934) Streptomyces griseus, (GenBanke Accession No. AF272737), Mycobacterium aurum, (GenBankO Accession No. AJ133724), Brevibacterium linens (GenBankE Accession No. AF139916), and Streptomyces coelicolor (GenBankS Accession No. AL109989) and AL109962, and microbial genome database gnl {TIGRE M. avium 89, Chlorobium tepidum TLS (CyanoBase CT0323), where the carotene desaturase isolated from Brevibacterium linens as described by SEQ ID NO : 19 is preferred.

The preferred substrate for carotene desaturase is a cyclic carotene having at least one p-ionone ring. Typical suitable substrates include, but are not limited to, ß-carotene ; y-carotene; a-carotene ; zeaxanthin; ß- isorenieratene (, ß-carotene) ; torulene ; 1'2'-dihydro-y-carotene ; 7,8- dihydro-y-carotene; 7'8'-dihydro-p-carotene ; 7', 8', 7, 8-tetrahydro-ß- carotene; p-zeacarotene ; echinenone; 3-OH-ß-carotene ; 1', 2'-dihydro-1'- OH-torulene ; 16'-OH-torulene ; 16'-oxo-torulene ; and 16'-carboxy-torulene.

Typical aryl carotenoids that will be produced by the aromatization of the p-ionone ring on the cyclic carotenoid will include, but are not limited to isorenieratene (, -carotene) ; P-isorenieratene (, P-carotene) ; chlorobactene (, W-carotene) ; , s-carotene ; 1, 2-didehydrochlorobactene ; 1', 2'-dihydrochlorobactene ; 7, 8-dihydro-chlorobactene ; 7'8'-dihydro- isorenieratene; 7', 8', 7,8-tetrahydro-isorenieratene ; 7'8'-dihydro- chlorobactene ; ß, #-carotene-4-one ; ß, #-carotene-3-ol ; 3-OH- isorenieratene; 3,3'-dihydroxy-isorenieratene ; 7', 8'-didehydrorenieratene ; OH-chlorobactene ; 1', 2'-dihydro-1'-OH-didehydrochlorobactene ; 16'-OH- didehydrochlorobactene ; 16'-oxo-didehydrochlorobactene ; and 16'- carboxy-didehydrochlorobactene. Aryl carotenoids of the invention may be either symmetrical (having two p-ionone rings) as with {3-carotene for example, or may be asymmetrical having only a single ß-ionone ring as with y-carotene for example.

Within the context of the present invention a preferred substrate is ß-carotene which is catalyzed to isorenieratene or y-carotene which is catalyzed to chlorobactene by the action of the crtU gene product.

Recombinant Expression-Microbial A gene encoding a carotene desaturase (crtU) has been expressed recombinantly in a heterologous microbial host. Expression of crtU in recombinant microbial hosts will be useful for the synthesis of new products heretofore not possible using the native host.

Preferred heterologous host cells for expression of the instant genes and nucleic acid fragments are microbial hosts that can be found broadly within the fungal or bacterial families and which grow over a wide range of temperature, pH values, and solvent tolerances. For example, it is contemplated that any of bacteria, yeast, and filamentous fungi will be suitable hosts for expression of the present cttu genes.

It will be appreciated by the skilled artisan that the expression of the present c*U genes may be regulated by controlling a number of well known factors. For example, large-scale microbial growth and functional gene expression may utilize a wide range of simple or complex carbohydrates, organic acids and alcohols, and saturated hydrocarbons, such as methane or carbon dioxide, in the case of photosynthetic or chemoautotrophic hosts. However, the functional genes such as crû may be regulated, repressed or depressed by specific growth conditions, which may include the form and amount of nitrogen, phosphorous, sulfur, oxygen, carbon or any trace micronutrient including small inorganic ions.

In addition, the regulation of crû genes may be achieved by the presence or absence of specific regulatory molecules that are added to the culture and are not typically considered nutrient or energy sources. Growth rate may also be an important regulatory factor in gene expression.

A variety of microbial host cells will be suitable for the heterologous expression of the present crtU. Examples of host strains include but are not limited to bacterial, fungal or yeast species such as Aspergillus, Trichoderma, Saccharomyces, Pichia, Candida, Rhodotorula, Rhodosporidium, Phaffia, Hansenula, or bacterial species such as Salmonella, Bacillus, Acinetobacter, Zymomonas, Agrobacterium, Erythrobacter Chlorobium, Chromatium, Flavobacterium, Cytophaga, Rhodobacter, Rhodococcus, Streptomyces, Brevibacterium, Corynebacteria, Mycobacterium, Deinococcus, Escherichia, En/vinia, Pantoea, Pseudomonas, Sphingomonas, Methylomonas, Methylobacter, Methylococcus, Methylosinus, Methylomicrobium, Methylocystis, Alcaligenes, Synechocystis, Synechococcus, Anabaena, Thiobacillus, Methanobacterium, Klebsiella, and Myxococcus.

Microbial expression systems and expression vectors containing regulatory sequences that direct high level expression of foreign proteins are well known to those skilled in the art. Any of these could be used to construct chimeric genes for expression of present carotene desaturase.

These chimeric genes could then be introduced into appropriate microorganisms via transformation to provide high-level expression of the enzymes.

Accordingly, it is expected that introduction of chimeric gene encoding the instant bacterial enzymes under the control of the appropriate promoters will result in the production of aryl carotenoids or in the modulation of an existing isoprenoid pathway.

Vectors or cassettes useful for the transformation of suitable host cells are well known in the art. Typically the vector or cassette contains sequences directing transcription and translation of the relevant gene, a selectable marker, and sequences allowing autonomous replication or chromosomal integration. Suitable vectors comprise a region 5'of the gene which harbors transcriptional initiation controls and a region 3'of the DNA fragment which controls transcriptional termination. It is most preferred when both control regions are derived from genes homologous to the transformed host cell, although it is to be understood that such control regions need not be derived from the genes native to the specific species chosen as a production host.

Initiation control regions or promoters, which are useful to drive expression of the instant ORF's in the desired host cell are numerous and familiar to those skilled in the art. Virtually any promoter capable of driving these genes is suitable for the present invention including but not limited to <BR> <BR> <BR> CYC1, HIS3, GAL1, GAL101 ADH1, PGK, PH05, GAPDH, ADC1, TRP1, URA3, LEU2, ENO, TPI (useful for expression in Saccharomyces) ; AOX1 (useful for expression in Pichia) ; and lac, ara, tet, trp, IPL, IPR, T7, tac, and trc (useful for expression in Escherichia coli) as well as the amy, apr, and npr promoters and various phage promoters useful for expression in Bacillus.

Termination control regions may also be derived from various genes native to the preferred hosts. Optionally, a termination site may be unnecessary, however, it is most preferred if included.

Knowledge of the sequence of the present gene will be useful in manipulating the isoprenoid or carotenoid biosynthetic pathways in any organism having such a pathway. Methods of manipulating genetic

pathways are common and well known in the art. Selected genes in a particularly pathway may be upregulated or down regulated by variety of methods. Additionally, competing pathways organism may be eliminated or sublimated by gene disruption and similar techniques.

Once a key genetic pathway has been identified and sequenced specific genes may be upregulated to increase the output of the pathway.

For example, additional copies of the targeted genes may be introduced into the host cell on multicopy plasmids such as pBR322. Alternatively, the target genes may be modified so as to be under the control of non- native promoters. Where it is desired that a pathway operate at a particular point in a cell cycle or during a fermentation run, regulated or inducible promoters may used to replace the native promoter of the target gene. Similarly, in some cases the native or endogenous promoter may be modified to increase gene expression. For example, endogenous promoters can be altered in vivo by mutation, deletion, and/or substitution (see, Kmiec, US 5565350; Zarling et al., PCT/US93/03868).

Alternatively, it may be necessary to reduce or eliminate the expression of certain genes in the target pathway or in competing pathways that may serve as competing sinks for energy or carbon.

Methods of down-regulating genes for this purpose have been explored.

Where sequence of the gene to be disrupted is known, one of the most effective methods gene down regulation is targeted gene disruption where foreign DNA is inserted into a structural gene so as to disrupt transcription.

This can be effected by the creation of genetic cassettes comprising the DNA to be inserted (often a genetic marker) flanked by sequence having a high degree of homology to a portion of the gene to be disrupted.

Introduction of the cassette into the host cell results in insertion of the foreign DNA into the structural gene via the native DNA replication mechanisms of the cell. (See for example Hamilton et al., J. Bacteriol., 171: 4617-4622 (1989), Balbas et al., Gene, 136: 211-213 (1993), Gueldener et al., Nucleic Acids Res. , 24: 2519-2524 (1996), and Smith et al., Methods Mol. Cell. Biol., 5: 270-277 (1996)).

Antisense technology is another method of down regulating genes where the sequence of the target gene is known. To accomplish this, a nucleic acid segment from the desired gene is cloned and operably linked to a promoter such that the anti-sense strand of RNA will be transcribed.

This construct is then introduced into the host cell and the antisense strand of RNA is produced. Antisense RNA inhibits gene expression by

preventing the accumulation of mRNA which encodes the protein of interest. The person skilled in the art will know that special considerations are associated with the use of antisense technologies in order to reduce expression of particular genes. For example, the proper level of expression of antisense genes may require the use of different chimeric genes utilizing different regulatory elements known to the skilled artisan.

Although targeted gene disruption and antisense technology offer effective means of down regulating genes where the sequence is known, other less specific methodologies have been developed that are not sequence based. For example, cells may be exposed to a UV radiation and then screened for the desired phenotype. Mutagenesis with chemical agents is also effective for generating mutants and commonly used substances include chemicals that affect nonreplicating DNA such as HNO2 and NH2OH, as well as agents that affect replicating DNA such as acridine dyes, notable for causing frameshift mutations. Specific methods for creating mutants using radiation or chemical agents are well documented in the art. See for example Thomas D. Brock in Biotechnology : A Textbook of Industrial Microbiology, Second Edition (1989) Sinauer Associates, Inc., Sunderland, MA. , or Deshpande, Mukund V., Appl. Biochem. Biotechnol., 36,227, (1992).

Another non-specific method of gene disruption is the use of transposable elements or transposons. Transposons are genetic elements that insert randomly in DNA but can be latter retrieved on the basis of sequence to determine where the insertion has occurred. Both in vivo and in vitro transposition methods are known. Both methods involve the use of a transposable element in combination with a transposase enzyme. When the transposable element or transposon is contacted with a nucleic acid fragment in the presence of the transposase, the transposable element will randomly insert into the nucleic acid fragment. The technique is useful for random mutageneis and for gene isolation, since the disrupted gene may be identified on the basis of the sequence of the transposable element. Kits for in vitro transposition are commercially available (see, for example, The Primer Island Transposition Kit, available from Perkin Elmer Applied Biosystems, Branchburg, NJ, based upon the yeast Ty1 element; The Genome Priming System, available from New England Biolabs, Beverly, MA; based upon the bacterial transposon Tn7; and the EZ: : TN Transposon Insertion Systems,

available from Epicentre Technologies, Madison, WI, based upon the Tn5 bacterial transposable element).

Within the context of the present invention, where there is a pre- existing isoprenoid pathway in the selected host cell, it will be useful, for example to disrupt the gene encoding the ketolase encoded by crtO. This, because the gene product of crtO competes with crtU for the same substrate, and disruption of crtO will be expected to enhance the enzymatic product of crtU (see Figure 2).

Industrial Production Where commercial production of aryl-carotenoid compounds is desired using the present cffU genes, a variety of culture methodologies may be applied. For example, large-scale production of a specific gene product overexpressed from a recombinant microbial host may be produced by both batch and continuous culture methodologies.

A classical batch culturing method is a closed system where the composition of the media is set at the beginning of the culture and not subject to artificial alterations during the culturing process. Thus, at the beginning of the culturing process the media is inoculated with the desired organism or organisms and growth or metabolic activity is permitted to occur adding nothing to the system. Typically, however, a"batch"culture is batch with respect to the addition of carbon source and attempts are often made at controlling factors such as pH and oxygen concentration. In batch systems the metabolite and biomass compositions of the system change constantly up to the time the culture is terminated. Within batch cultures cells moderate through a static lag phase to a high growth log phase and finally to a stationary phase where growth rate is diminished or halted. If untreated, cells in the stationary phase will eventually die. Cells in log phase are often responsible for the bulk of production of end product or intermediate in some systems. Stationary or post-exponential phase production can be obtained in other systems.

A variation on the standard batch system is the Fed-Batch system.

Fed-Batch culture processes are also suitable in the present invention and comprise a typical batch system with the exception that the substrate is added in increments as the culture progresses. Fed-Batch systems are useful when catabolite repression is apt to inhibit the metabolism of the cells and where it is desirable to have limited amounts of substrate in the media. Measurement of the actual substrate concentration in Fed-Batch systems is difficult and is therefore estimated on the basis of the changes

of measurable factors such as pH, dissolved oxygen and the partial pressure of waste gases such as CO2. Batch and Fed-Batch culturing methods are common and well known in the art and examples may be found in Brock (supra) and Deshpande (supra).

Commercial production of aryl-carotenoids may also be accomplished with a continuous culture. Continuous cultures are an open system where a defined culture media is added continuously to a bioreactor and an equal amount of conditioned media is removed simultaneously for processing. Continuous cultures generally maintain the cells at a constant high liquid phase density where cells are primarily in log phase growth. Alternatively, continuous culture may be practiced with immobilized cells where carbon and nutrients are continuously added, and valuable products, by-products or waste products are continuously removed from the cell mass. Cell immobilization may be performed using a wide range of solid supports composed of natural and/or synthetic materials.

Continuous or semi-continuous culture allows for the modulation of one factor or any number of factors that affect cell growth or end product concentration. For example, one method will maintain a limiting nutrient such as the carbon source or nitrogen level at a fixed rate and allow all other parameters to moderate. In other systems a number of factors affecting growth can be altered continuously while the cell concentration, measured by media turbidity, is kept constant. Continuous systems strive to maintain steady state growth conditions and thus the cell loss due to media being drawn off must be balanced against the cell growth rate in the culture. Methods of modulating nutrients and growth factors for continuous culture processes as well as techniques for maximizing the rate of product formation are well known in the art of industrial microbiology and a variety of methods are detailed by Brock, supra.

Fermentation media in the present invention must contain suitable carbon substrates. Suitable substrates may include but are not limited to monosaccharides such as glucose and fructose, oligosaccharides such as lactose or sucrose, polysaccharides such as starch or cellulose or mixtures thereof and unpurified mixtures from renewable feedstocks such as cheese whey permeate, cornsteep liquor, sugar beet molasses, and barley malt. Additionally the carbon substrate may also be one-carbon substrates such as carbon dioxide, methane or methanol for which metabolic conversion into key biochemical intermediates has been

demonstrated. In addition to one and two carbon substrates methylotrophic organisms are also known to utilize a number of other carbon containing compounds such as methylamine, glucosamine and a variety of amino acids for metabolic activity. For example, methylotrophic yeast are known to utilize the carbon from methylamine to form trehalose or glycerol (Bellion et al., Microb. Growth C1 Compd. , [Int. Symp. ], 7th (1993), 415-32. Editor (s): Murrell, J. Collin ; Kelly, Don P. Publisher : Intercept, Andover, UK). Similarly, various species of Candida will metabolize alanine or oleic acid (Sulter et al., Arch. Microbiol., 153: 485-489 (1990) ). Hence it is contemplated that the source of carbon utilized in the present invention may encompass a wide variety of carbon containing substrates and will only be limited by the choice of organism.

Recombinant Expression-Plants Plants and algae are also known to produce carotenoid compounds. The crtU genes of the instant invention may be used to create transgenic plants having the ability to express a carotene desaturase. Preferred plant hosts will be any variety that will support a high production level of this enzyme. Suitable green plants will include but are not limited to soybean, rapeseed (Brassica napus, B. campestris), pepper, sunflower (Helianthus annus), cotton (Gossypium hirsutum), corn, tobacco (Nicotiana tabacum), alfalfa (Medicago sativa), wheat (Triticum sp), barley (Hordeum vulgare), oats (Avena sativa, L), sorghum (Sorghum bicolor), rice (Oryza sativa), Arabidopsis, cruciferous vegetables (broccoli, cauliflower, cabbage, parsnips, etc.), melons, carrots, celery, parsley, tomatoes, potatoes, strawberries, peanuts, grapes, grass seed crops, sugar beets, sugar cane, beans, peas, rye, flax, hardwood trees, softwood trees, and forage grasses. Algal species include, but are not limited to, commercially significant hosts such as Spirulina, Haemotacoccus, and Dunalliela. Production of the aryl carotenoid compounds may be accomplished by first constructing chimeric genes of present invention in which the coding region are operably linked to promoters capable of directing expression of a gene in the desired tissues at the desired stage of development. For reasons of convenience, the chimeric genes may comprise promoter sequences and translation leader sequences derived from the same genes. 3'Non-coding sequences encoding transcription termination signals must also be provided. The instant chimeric genes may also comprise one or more introns in order to facilitate gene expression.

Any combination of promoter and terminator capable of inducing expression of a coding region may be used in the chimeric genetic sequence. Some suitable examples of promoters and terminators include those from opaline synthase (nos), octopine synthase (ocs) and cauliflower mosaic virus (CaMV) genes. One type of efficient plant promoter that may be used is a high-level plant promoter. Such promoters, in operable linkage with the genetic sequences of the present invention, should be capable of promoting expression of the present gene product. High-level plant promoters that may be used in this invention include the promoter of the small subunit (ss) of the ribulose-1, 5- bisphosphate carboxylase from soybean (Berry-Lowe et al., J. Molecular and App. Gen. , 1: 483-498 (1982) ), and the promoter of the chlorophyll a/b binding protein. These two promoters are known to be light-induced in plant cells (see, for example, Genetic Engineering of Plants, an Agricultural Perspective, A. Cashmore, Plenum, NY (1983), pages 29-38; Coruzzi, G. et al., The Journal of Biological Chemistry, 258: 1399 (1983), and Dunsmuir, P. et al., Journal of Molecular and Applied Genetics, 2: 285 (1983)).

Plasmid vectors comprising the instant chimeric genes can then constructed. The choice of plasmid vector depends upon the method that will be used to transform host plants. The skilled artisan is well aware of the genetic elements that must be present on the plasmid vector in order to successfully transform, select and propagate host cells containing the chimeric gene. The skilled artisan will also recognize that different independent transformation events will result in different levels and patterns of expression (Jones et al., EMBO J. , 4: 2411-2418 (1985); De Almeida et al., Mol. Gen. Genetics, 218: 78-86 (1989) ), and thus multiple events must be screened in order to obtain lines displaying the desired expression level and pattern. Such screening may be accomplished by Southern analysis of DNA blots (Southern, J. Mol. Biol.

98: 503 (1975) ), Northern analysis of mRNA expression (Kroczek, <BR> <BR> <BR> J. , Chromatogr. Biomed. Appl., 618 (1-2): 133-145 (1993) ), and Western analysis of protein expression, or phenotypic analysis.

For some applications it will be useful to direct the instant proteins to different cellular compartments. It is thus envisioned that the chimeric genes described above may be further supplemented by altering the coding sequences to encode enzymes with appropriate intracellular targeting sequences such as transit sequences (Keegstra, K., Cell,

56: 247-253 (1989)), signal sequences or sequences encoding endoplasmic reticulum localization (Chrispeels, J. J., Ann. Rev. Plant Phys.

Plant Mol. Biol., 42: 21-53 (1991) ), or nuclear localization signals (Raikhel, N. , Plant Phys., 100: 1627-1632 (1992) ) added and/or with targeting sequences that are already present removed. While the references cited give examples of each of these, the list is not exhaustive and more targeting signals of utility may be discovered in the future that are useful in the invention.

Cells comprising a carotene desaturase (crtU) enzyme maybe harvested and crude cell extracts prepared by means well known in the art.

Alternatively, the crtU enzyme may be purified by ion exchange, hydroxyapatite, ammonium sulfate, sizing gel, and PAGE gel electrophoresis chromatography using standard techniques.

Crude cell extracts or purified protein may be suspended in a buffer and contacted with a carotenoid substrate (having a ß-ionone ring) dissolved in a suitable solvent such as acetone. These reactants may be mixed and incubated for suitable time and the reaction stopped by solvent extraction of the carotenoid products.

Description of the Preferred Embodiments The present crû gene and its expression product, CrtU, a carotene desaturase, is useful for the creation of recombinant organisms that have the ability to produce aryl-carotenoid compounds. Nucleic acid fragments encoding CrtU have been isolated from a strain of Brevibacterium linens and expressed in Rhodococcus erythropolis ATCC 47072. The Rhodococcus erythropolis ATCC 47072 naturally produces 4-keto-y- carotene as the major carotenoid, and CrtO was identified to be the carotenoid ketolase responsible for converting y-carotene to 4-keto-y- carotene (Figure 2). Because biosynthesis of aromatic carotenoids catalyzed by CrtU proceeds by desaturation and methyltransferation of the (3-ionone rings of carotenoids, either ß-carotene or y-carotene are potential substrates. To make a ß-ionone ring structure available for the CrtU, the natural CrtO activity was removed from the strain by a knockout mutation.

In one example, the crtU gene from B. linens was expressed with the Rhodococcus dxs gene (encoding 1-deoxyxylulose-5-phosphate synthase) which was used to improve flux of isoprenoid pathway metabolites. In another example, the cite gene was expressed using the chloramphenicol resistance gene promoter, which improved expression levels.

EXAMPLES The present invention is further defined in the following Examples.

It should be understood that these Examples, while indicating preferred embodiments of the invention, are given by way of illustration only. From the above discussion and these Examples, one skilled in the art can ascertain the essential characteristics of this invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions.

GENERAL METHODS Standard recombinant DNA and molecular cloning techniques used in the Examples are well known in the art and are described by Sambrook, J. , Fritsch, E. F. and Maniatis, T. Molecular Cloning : A Laboratory Manual ; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, (1989) (Maniatis) and by T. J. Silhavy, M. L. Bennan, and L. W. Enquist, Experiments with Gene Fusions, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1984) and by Ausubel, F. M. et al., Current Protocols in Molecular Biology, pub. by Greene Publishing Assoc. and Wiley- lnterscience (1987).

Materials and methods suitable for the maintenance and growth of bacterial cultures are well known in the art. Techniques suitable for use in the following examples may be found as set out in Manual of Methods for General Bacteriology (Phillipp Gerhardt, R. G. E. Murray, Ralph N.

Costilow, Eugene W. Nester, Willis A. Wood, Noel R. Krieg and G. Briggs Philips, eds), American Society for Microbiology, Washington, DC. (1994)) or by Brock, supra. All reagents, restriction enzymes and materials used for the growth and maintenance of bacterial cells were obtained from Aldrich Chemicals (Milwaukee, WI), DIFCO Laboratories (Detroit, MI), GIBCO/BRL (Gaithersburg, MD), or Sigma Chemical Company (St. Louis, MO) unless otherwise specified.

Manipulations of genetic sequences were accomplished using the suite of programs available from the Genetics Computer Group Inc.

(Wisconsin Package Version 9.0, Genetics Computer Group (GCG), Madison, WI). Where the GCG program"Pileup"was used the gap creation default value of 12, and the gap extension default value of 4 were used. Where the CGC"Gap"or"Bestfit"programs were used the default gap creation penalty of 50 and the default gap extension penalty of 3 were used. Multiple alignments were created using the FASTA program

incorporating the Smith-Waterman algorithm (W. R. Pearson, Comput.

Methods Genome Res. , [Proc. Int. Symp. ] (1994), Meeting Date 1992, 111-20. Editor (s): Suhai, Sandor. Publisher : Plenum, New York, NY). In any case where program parameters were not prompted for, in these or any other programs, default values were used.

The meaning of abbreviations is as follows : "h"means hour (s), "min"means minute (s), "sec"means second (s), "d"means day (s),"mL" means milliliters,"IlL"means microliters,"L"means liters,"lig"means micrograms, "mM"means millimolar,", uM"means micromolar,"ppm" means parts per million, and"nm"means nanometer.

Bacterial Strains Bacterial strains used in the following examples and their sources are given in the table below : Bacterial Strain Use Source Rhodococcus erythropolis Production of American Type Culture Collection ATCC 47072 chlorobactene P. O. Box 1549 Manassas, VA 20108, U. S. A Rhodococcus erythropolis Source of crtO, Example 1 AN12 dxs genes Brevibacterium linens ATCC Source of crtU American Type Culture Collection 9175 gene P. O. Box 1549 Manassas, VA 20108, U. S. A Vectors Transformation and expression vectors used in the following examples and their sources are given in the table below : Vector Purpose Source/description Designation pCR2. 1 TOPO Sub-cloning vector Invitrogen Carlsbad, CA pBR328 Used as an integration American Type Culture Collection vector to create #37517 Rhodococcus crtO knockout P. O. Box 1549 strain Manassas, VA 20108, U. S. A Genbank Accession #L08858 pDA71 E. coli-Rhodococcus shuttle American Type Culture Collection vector #77474 P. O. Box 1549 Manassas, VA 20108, U. S. A pDCQ102 pBR328 (crt0 R. erythropolis Example 2 ATCC 47072) pRhBR17 E. coli-Rhodococcus shuttle Example 3 vector pRhBR171 E coli-Rhodococcus shuttle Example 3 vector pDCQ22 pRhBR171 (dxs R. Example 3 erythropolis AN12) pDCQ23 pRhBR171 (dxs R. Example 3 erythropolis AN12) pDCQ138 pCR2. 1 (B. linens cftU Exampfe 4 pDCQ139pDCQ23 (B. linens crtU) Example 4 DCQ140 pDA71(dxs-crtU) Example 4 pDCQ141 pCR2. 1 (cmR) Example 5 DCQ142 pRhBR171 (cmR) Example 5 DCQ143 PDCQ142 (crtU) Example 5

EXAMPLE 1 Isolation and Characterization of Strain AN12 This Example describes the isolation of strain AN12 of Rhodococcus erythropolis on the basis of being able to grow on aniline as the sole source of carbon and energy. Analysis of a 16S rRNA gene sequence indicated that strain AN12 was related to high G + C Gram positive bacteria belonging to the genus Rhodococcus.

Bacteria that grow on aniline were isolated from an enrichment culture. The enrichment culture was established by inoculating 1 mL of activated sludge into 10 mL of S12 medium (10 mM ammonium sulfate, 50 mM potassium phosphate buffer (pH 7.0), 2 mM MgCl2, 0.7 mM Cal2, 50 µM MnCl2, 1 µM FeCl3, 1 pM ZnCI3, 1. 72 µM CuSO4, 2.53 µM CoC12, 2. 42 FM Na2MoO2, and 0. 0001% FeSO4) in a 125 mL screw cap Erlenmeyer flask. The activated sludge was obtained from a wastewater treatment facility. The enrichment culture was supplemented with 100 ppm aniline added directly to the culture medium and was incubated at 25°C with reciprocal shaking. The enrichment culture was maintained

by adding 100 ppm of aniline every 2-3 days. The culture was diluted every 14 days by replacing 9.9 mL of the culture with the same volume of S12 medium. Bacteria that utilize aniline as a sole source of carbon and energy were isolated by spreading samples of the enrichment culture onto S12 agar. Aniline was placed on the interior of each petri dish lid. The petri dishes were sealed with parafilm and incubated upside down at room temperature (25°C). Representative bacterial colonies were then tested for the ability to use aniline as a sole source of carbon and energy.

Colonies were transferred from the original S12 agar plates used for initial isolation to new S12 agar plates and supplied with aniline on the interior of each petri dish lid. The petri dishes were sealed with parafilm and incubated upside down at room temperature (25°C).

The 16S rRNA genes of each isolate were amplified by PCR and analyzed as follows. Each isolate was grown on R2A agar (Difco Laboratories, Bedford, MA). Several colonies from a culture plate were suspended in 1001lL of water. The mixture was frozen and then thawed.

The 16S rRNA gene sequences were amplified by PCR by using a commercial kit according to the manufacturer's instructions (Perkin Elmer, Boston, MA) with primers HK12 (5'-GAGTTTGATCCTGGCTCAG-3') (SEQ ID NO : 1) and HK13 (5'-TACCTTGTTACGACTT-3') (SEQ ID NO : 2). PCR was performed in a Perkin Elmer GeneAmp 9600. The samples were incubated for 5 minutes at 94°C and then cycled 35 times at 94°C for 30 seconds, 55°C for 1 minute, and 72°C for 1 minute. The amplified 16S rRNA genes were purified using a commercial kit according to the manufacturer's instructions (QlAquick PCR Purification Kit, Qiagen, Valencia, CA) and sequenced on an automated ABI sequencer. The sequencing reactions were initiated with primers HK12, HK13, and HK14 (5'-GTGCCAGCAGYMGCGGT-3') (SEQ ID NO : 3, where Y=C or T, M=A or C). The 16S rRNA gene sequence of each isolate was used as the query sequence for a BLAST search (Altschul et al., Nucleic Acids Res. , 25: 3389-3402 (1997) ) of GenBank@ for similar sequences.

A 16S rRNA gene of strain AN12 was sequenced (SEQ ID NO : 4) and compared to other 16S rRNA sequences in the GenBank@ sequence database. The 16S rRNA gene sequence from strain AN12 was 98% similar to the 16S rRNA gene sequences of high G + C Gram positive bacteria Rhodococcus erythropolis.

EXAMPLE 2 Construction of the Rhodococcus cet0 Knockout Mutant as the Production Host The crtO knockout mutant of Rhodococcus erythropolis ATCC 47072, hereafter referred to as 47072, was generated by homologous recombination using the following procedure. PCR primers AN12_12_F (5'-CCATGGTCTGCGCACCTCATGATCCGA-3' : SEQ ID NO : 5) and AN12_12_R (5'-CCATGGMTGMGCGGTCGAGGACGGA-3': SEQ ID NO : 6) were designed based on the Rhodococcus erythropolis AN12 crtO sequence (SEQ ID NO : 7), the underlined sequences were the engineered Ncol sites. The primers were used to amplify the 1151 bp crtO internal fragment from 47072 with a 275 bp truncation at the N-terminal end and 173 bp truncation at the C-terminal end of the gene. The crtO amplified from 47072 (SEQ ID NO : 8) was confirmed by sequencing and showed 95% identity at the DNA level with the AN12 crfO The crtO fragment was first cloned into pCR2.1 TOPO vector (Invitrogen, Carlsbad, CA). The TOPO clones were then digested with Ncol and the crtO fragment was subsequently cloned into the Ncol site of pBR328 (GenBankO Accession # L08858). The resulting construct was confirmed by sequencing and designated as pDCQ102. Approximately 1 pg DNA of pDCQ102 was introduced into 47072 by electroporation and plated on NBYE (0.8% nutrient broth and 0.5% yeast extract) plates with 10 pg/mL tetracycline.

The pBR328 vector does not replicate in Rhodococcus and the tetracycline resistant transformants obtained after 3-4 days of incubation at 30° C were generated by chromosomal integration. Integration into the targeted crtO gene on chromosome of 47072 was confirmed by PCR. The vector specific primers PBR3 (5'-AGCGGCATCAGCACCTTG-3' ; SEQ ID NO : 9) and PBR5 (5'-GCCAATATGGACAACTTCTTC-3' ; SEQ ID NO : 10), paired with the gene specific primers (outside of the insert on pDCQ102) 12_0P5 (5t-ACCTGAGGTGTTCGACGAGGACMCCGA-3'; SEQ ID NO : 11) and 12_0P3 (5'-GTTGCACAGTGGTCATCGTGCCAGCCGT-3'; SEQ ID NO : 12) were used for PCR with chromosomal DNA prepared from the tetracycline resistant transformants as template. PCR fragments of the expected size were amplified from the tetracycline resistant transformants, but no PCR product was obtained from the wild-type 47072. When the two gene specific primers were used, no PCR fragment was obtained with the tetracycline resistant transformants due to the insertion of the large vector DNA. The PCR fragment obtained with the vector specific primers

and the gene specific primers was sequenced. Sequence analysis of the junction of the vector and the crtO gene confirmed that the single crossover recombination occurred at the expected site and disrupted the targeted crtO gene.

TABLE 1 Gene SEQ % % E-Citation Name and Similarity ID Identitya Similarityb valuec Organism of Identified Isolation, l cet0 phytoene 7 35 50 6e-75 White, O. et Rhodococcus dehydrogenase, aL, Genome erythropolis putative sequence of AN12 [Deinococcus the radio- radiodurans] resistant (AE001872) bacterium Deinococcus radiodurans R1 Science. 286 (5444), 1571-1577 (1999) crtO phytoene 8 36 51 1e-72 White, O. et Rhodococcus dehydrogenase, a/., Genome erythropolis putative sequence of ATCC47072 [Deinococcus the radiodurans] radioresistant (AE001872) bacterium Deinococcus radiodurans R1 Science. 286 (5444), 1571-1577 (1999ì The sequence of crtO was compared for similarity with other sequences by conducting BLAST (Basic Local Alignment Search Tool ; Altschul, S. F. , et al., J. Mol. Biol., 215: 403-410 (1993) ) searches against the"nr"database (comprising all non-redundant GenBankO CDS translations, sequences derived from the 3-dimensional structure Brookhaven Protein Data Bank, the SWISS-PROT protein sequence database, EMBL, and DDBJ databases). These sequences were analyzed for similarity to all publicly available DNA sequences contained in the"nr"database using the BLASTN algorithm provided by the National Center for Biotechnology Information (NCBI). The DNA sequences were translated in all reading frames and compared for similarity to all publicly

available protein sequences contained in the"nr"database using the BLASTX BLOSUM62 algorithm with a gap existence cost of 11 per residue gap cost of 2, filtered, gap alignment (Gish, W. and States, D. J., Nature Genetics, 3: 266-272 (1993) ) provided by the NCBI.

The results of the BLAST comparisons are given in Table 1, which summarize the sequence to which each sequence has the most similarity.

Table 1 displays data based on the BLASTXnr algorithm with values reported in expect values. The Expect value estimates the statistical significance of the match, specifying the number of matches, with a given score, that are expected in a search of a database of this size absolutely by chance.

Although the BLAST analysis indicated that CrtO had the highest homology to phytoene dehydrogenases, knockout mutants of CrtO suggested that CrtO encodes a carotenoid ketolase.

Colonies of the cet0 mutant, 47072 (cet0-), were yellow compared to the pink color of the 47072 wild-type strain, which suggested that different carotenoid pigments were produced in the crtO mutant, 47072 (cftO-). To extract the carotenoids from 47072 (crut0-), 100 mL of cell culture in NBYE (0. 8% nutrient broth + 0.5% yeast extract) was grown at 26° C to the stationary phase in aerobic conditions. Cells were spun down at 4000 g for 15 min, and the cell pellets were resuspended in 10 mL acetone. Carotenoids were extracted into acetone with constant shaking at room temperature for 1 hour. The cells were spun down and the supernatant was collected. The extraction was repeated once, and the supernatants of both extractions were combined and dried under nitrogen. The dried material was re-dissolved in 0.5 mL methanol and insoluble material was removed by centrifugation at 16,000 g for 2 min. The sample was analyzed by HPLC using a Beckman System Golds HPLC with Beckman Gold Nouveau Software (Columbia, MD). The extract (0.1 mL) was loaded onto a 125 x 4 mm RP8 (5 um particles) column with corresponding guard column (Hewlett- Packard, San Fernando, CA). The flow rate was 1 mL/min. The solvent program was: 0-11.5 min linear gradient from 40% water/60% methanol to 100% methanol, 11.5-20 min 100% methanol, 20-30 min 40% water/60% methanol. The spectra data were collected by a Beckman photodiode array detector (model 168). HPLC analysis showed that 47072 (cffO-) lacked the major

carotenoid peak of the wild-type strain. The major peak observed in 47072 (crt0-) was at an elution time of 15.5 and the absorption maxima were 435 nm, 458 nm and 486 nm, which is identical to the characteristicsofy-carotene. Thus, ATCC47072 (cffO-) accumulated y-carotene and could be used as an expression host for crtU to synthesize aromatic carotenoids.

EXAMPLE 3 Construction Of pRhBR171 and pDCQ23 Shuttle Vectors An E coli-Rhodococcus shuttle vector requires a set of replication function and antibiotic resistance markers that functions both in E. coli and in Rhodococcus. Applicants have identified a cryptic pAN12 plasmid which encodes the replication function for Rhodococcus. To identify an antibiotic resistance marker for Rhodococcus, E. coli plasmid pBR328 (ATCC &num 37517, GenBank@ Accession #L08858) was tested to see whether it would function in Rhodococcus. Plasmid pBR328 carries ampicillin, chloramphenicol, and tetracycline resistance markers that function in E. coli. pBR328 was linearized with Pvull which disrupted the chloramphenicol resistance gene and ligated with pAN12 digested with Sspi. The resulting clone was designated pRhBR17 (SEQ ID NO : 13).

Transposon mutagenesis of the shuttle plasmid pRhBR17 suggested that certain regions of the shuttle plasmid may not be essential for the plasmid function (Table 2). One of these regions was at the junction of pBR328 and pAN12. It was decided to examine whether this region of the plasmid was dispensable and if the size of the shuttle plasmid could be trimmed. Shuttle plasmid pRhBR17 was digested with Pst I (2 sites/2520,3700 bp) and Mlu I (1 site/4105 bp), yielding three fragments of the following sizes: 9656,1180 and 405 bp. The digested DNA fragments were blunt ended with mung bean nuclease (New England Biolabs, Beverly, MA) following manufacturer's instruction. The largest 9.7 kb fragment was separated by size on an agarose gel, and purified using QIAEX ll Gel Extraction Kit (Qiagen). This 9.7 kb DNA fragment with deletion of region 2520-4105 bp of pRhBR17 was self-ligated to form a circular plasmid designated pRhBR171. Plasmid isolation from the E. coli DH10B transformants and restriction enzyme characterization showed the correct size and digest pattern of pRhBR171. E. coli cells harboring the pRhBR171 plasmid lost the ability to grow in the presence of ampicillin (100 pg/mL), since the Pst I and M/u I digest removed part of the coding region for the ampicillin resistant gene on the parental plasmid.

The tetracycline resistance gene on pRhBR171 served as the selection marker for both E. coli and Rhodococcus. Transformation of pRhBR171 to Rhodococcus was tested. It transformed competent Rhodococcus erythropolis ATCC 47072 and AN12 cells with similar frequency by electroporation as compared with its parental plasmid pRhBR17. These results demonstrate that this region (2520-4105 bp) of pRhBR17 was not essential as suggested by transposon mutagenesis. It also provided a smaller shuttle vector that is more convenient for cloning.

TABLE 2: Transposon insertion mapping of pRhBR17 and the effects on transformation of Rhodococcus ATCC 47072 Clone Site inserted Strand inserted Gene Transformation number inserted ability pRhBR17 No insertion N/AN/A+++ 30,31 2092 bp Forward pBR328 26,27 3120 bp Reverse pBR328 ND 29 3468 bp Reverse pBR328 ND 24 3625 bp Reverse pAN12 +++ 2 4030 bp Reverse pAN 12 +++ 38, 39 4114 bp Forward pAN12 +++ 20 4442 bp Reverse pAN12 +++ 1 4545 bp Reverse pAN12 +++ 35 4568 bp Forward pAN12 +++ 13 4586 bp Forward pAN12 + 17,33 <4920 bp Forward pAN12 + 7 5546 bp Forward pAN12 rep + 11 5739 bp Reverse pAN12 rep - 12 5773 bp Forward pAN12 rep - 16 5831 bp Forward pAN12 rep - 5 5883 bp Reverse pAN12 rep - 9 6050 bp Reverse pAN12 rep - 28 6283 bp Forward pAN12 rep - 6 6743 bp Reverse pAN12 - 12 5773 bp Forward pAN12 rep- 16 5831 bp Forward pAN12 rep- 5 5883 bp Reverse pAN12 rep- 9 6050 bp Reverse pAN12 rep- 28 6283 bp Forward pAN12 rep- 6 6743 bp Reverse pAN12 rep 28 6283 bp Forward pAN12 rep - 6 6743 bp Reverse pAN12 - 37 <6935 bp Forward pAN12 +++ 32 6965 bp Forward pAN12 +++ 15 6979 bp Forward pAN12 + 3 7285 bp Reverse pAN12 +++

Clone Site inserted Strand inserted Gene Transformation number inserted ability 4 7811 bp Reverse pAN12 +++ 22,23 8274 bp Forward pAN12 div +++ 21 8355 bp Forward pAN12 div +++ 18 8619 bp Reverse pAN12 div +++ 10 10322 bp Reverse pBR328 +++ 36 11030 bp Forward pBR328 ND +++ the transformation frequency was comparable to that of the wild type plasmid.

+ the transformation frequency decreased about 100 fold.

- the transformation frequency was zero.

ND the transformation frequency was not determined.

The dxs gene (SEQ ID NO : 14) with its native promoter was amplified from the RhodococcusAN12 strain by PCR. Two upstream primers, New dxs 5'primer, 5'-ATT TCG TTG AAC GGC TCG CC-3' (SEQ I D NO : 15), and New2 dxs 5'primer, 5'-CGG CAA TCC GAC CTC TAC CA-3' (SEQ ID NO : 16), were designed to include the native promoter region of dxs with different lengths. The downstream primer, New dxs 3' primer, 5'-TGA GAC GAG CCG TCA GCC TT-3' (SEQ ID NO : 17), included the underlined stop codon of the dxs gene. PCR amplification of AN12 total DNA using New dxs 5'+ New dxs 3'yielded one product of 2519 bp in size, which included the full length AN12 dxs coding region and about 500 bp of immediate upstream region (nt. #500-#3019). When using New2 dxs 5'+ New dxs 3'primer pair, the PCR product is 2985 bp in size, including the complete AN12 dxs gene and about 1 kb upstream region (nt. #34-#3019). Both PCR products were cloned in the pCR2.1- TOPO cloning vector according to manufacturer's instruction (Invitrogen).

Resulting clones were screened and sequenced. The confirmed plasmids were digested with EcoRl and the 2.5 kb and 3.0 kb fragments containing the dxs gene and the upstream region from each plasmid were treated with the Klenow enzyme and cloned into the unique Ssp I site of the E. coli - Rhodococcus shuttle plasmid pRhBR171. The resulting constructs pDCQ22 (clones #4 and &num 7) and pDCQ23 (clones #10 and #11) were electroporated into Rhodococcus erythropolis ATCC 47072 with tetracycline 10 pg/mL selection.

The pigment of the Rhodococcus transformants of pDCQ22 and pDCQ23 appeared darker as compared with those transformed with the

vector control. To quantify the carotenoid production of each Rhodococcus strain, 1 mL of fresh cultured cells were added to 200 mL fresh LB medium with 0.05% Tween-80 and 10 ug/mL tetracycline, and grown at 30° C for 3 days to stationary phase (Table 3). Cells were pelleted by centrifugation at 4000 g for 15 min and the wet weight was measured for each cell pellet. Carotenoids were extracted from the cell pellet into 10 mL acetone overnight with shaking and quantitated at the absorbance maximum (465 nm). The 465 nm wavelength was diagnostic of the absorbance peak for the carotenoid isolated from Rhodococcus sp.

ATCC 47072. The absorption data was used to calculate the amount of carotenoid produced. This amount was calculated and normalized in each strain based either on the cell paste weight or the cell density (OD600).

Carotenoid production calculated by either method showed about 1. 6-fold increase in 47072 with pDCQ22, which contained the dxs gene with the shorter promoter region.

Table 3: Carotenoids production by Rhodococcus strains. Strain OD600 weight (g) OD465 % a % (wt) ° % (OD600) ° % (avg) 47072 1.992 2.82 0.41 100 100 100 100 (pRhBR171) 47072 1.93 2.9 0.642 157 161 152 156 (pDCQ22) #4 47072 1.922 2.76 0.664 162 159 156 157 (pDCQ22) #7 47072 1.99 2.58 0.958 234 214 233 224 (pDCQ23) #10 47072 1. 994 2. 56 0. 979 239 217 239 228 (pDCQ23) #11 a % of carotenoid production based on OD465nm. b % of carotenoid production (OD465nm) normalized with wet cell paste weight. c % of carotenoid production (OD465nm) normalized with cell density (OD600nm). d % of carotenoid production (OD465nm) averaged from the normalizations with wet cell paste weight and cell density.

Carotenoid production increased even more (2. 2-fold) when the dxs gene was expressed with the longer promoter region as in pDCQ23 (Table 3). It is likely that the 1 kb upstream DNA contains the promoter and some elements for enhancement of the expression. HPLC analysis also verified that the same carotenoids were produced by the wild-type strain, 47072 (pRhBR171), and the dxs expression strain, 47072 (pDCQ22) or (pDCQ23).

EXAMPLE 4 Expression of Brevibacterium crtU Downstream of dxs gene in Rhodococcus The dxs gene (SEQ ID NO : 14) encodes the 1-deoxyxylulose-5- phosphate synthase that catalyzes the first step of the synthesis of 1-deoxyxylulose-5-phosphate from glyceraldehyde-3-phosphate and pyruvate precursors in the isoprenoid pathway for carotenoid synthesis.

As indicated in Example 3, placement of the Rhodococcus dxs gene with its native promoter on a multicopy shuttle vector (pDCQ23) increased Rhodococcus carotenoid production approximately 2-fold. Initially, the Brevibacterium crû gene (SEQ ID NO : 18) (Krubasik et al., Mol. Gen.

Genet., 263: 423-432 (2000) ) was cloned downstream of the dxs gene on pDCQ23. The rational behind this strategy was two-fold. First, the crû gene could be expressed in Rhodococcus by the dxs promoter, which had been demonstrated to be functional on pDCQ23. Second, the Rhodococcus strain with pDCQ23 produced twice as much carotenoid as compared with 47072 not carrying dxs (Table 3). A unique Mscl site on pDCQ23 was identified immediately downstream of dxs into which the crtU gene was cloned to form a polycistronic transcript. The crû gene with its native ribosome binding site was amplified from genomic DNA of Brevibacterium linens (ATCC 9175) by PCR, using forward primer crtURBS (Brevi) (5'-GTGCTCATGCTGTGGCAGTGGCAA-3'SEQ ID NO : 20) and reverse primer crtU_R (5'- TCATCGACGTCTCCTGATGAGCCCGAGCACT-3'SEQ ID NO : 21). The 1554 bp PCR product was first cloned in the pCR2.1-TOPO cloning vector (Invitrogen), resulting in plasmid pDCQ138. The 1.6 kb EcoRl fragment of pDCQ138 DNA containing the cru gene was filled in by Klenow DNA polymerase and ligated to the Mscl site in pDCQ23. In the resulting construct pDCQ139, the crfU gene was in the same orientation as the dxs gene. No transcription termination site was found between dxs and crû, which are separated by 131 bp of DNA sequence. The crtU presumably was co-transcribed with the dxs gene. The pDCQ139 construct contained a tetracycline resistant marker, which was the same marker as the one used to create the crfO knockout host as described in Example 2. The 6.0 kb Hind III fragment containing the dxs and crû gene cluster was cut out from pDCQ139 and ligated to the Hindlll site of another Rhodococcus shuttle vector pDA71 (ATCC 77474) (Figure 4), which has a chloramphenicol resistant marker (GenBank@ Accession # AJ308376,

SEQ ID NO : 22) for selection in Rhodococcus. The resulting construct, pDCQ140, (Figure 3) was verified by sequencing. Plasmid pDCQ140 and the vector control pDA71 were electroporated into 47042 (crtO-). Cells were grown in 200 mL LB with 0.05% Tween-80,10 pg/mL tetracycline and 40 ug/mL chloramphenicol at 30°C for 2 days. Carotenoids were extracted from cell pellets with 20 mL acetone, dried under nitrogen and dissolved in 1 mL methanol. Each sample of 0.1 mL was used for HPLC analysis as described in Example 2. In 47042 (crut0-), carrying the pDA71 vector control (Figure 4), the y-carotene eluted at 15.4 min. In the 47042 (crtO-), with pDCQ140 containing cffU, the majority of the carotenoids was y-carotene, which eluted at 15.4 min. However, a new carotenoid representing 7% of the total carotenoids eluted at 15.1 min. This new carotenoid was the conversion product of CrtU, chlorobactene, and was confirmed as is described in subsequent examples.

EXAMPLE 5 Expression of Brevibacterium crfU Downstream of Cm'Gene in Rhodococcus Only a small fraction of y-carotene was converted by crû downstream of dxs on pDCQ140. Since dxs was shown to be a limiting step for carotenoid synthesis in E. coli (Albrecht, M. , et al., Biotechnol.

Letters, 21: 791-795 (1999) ), it is likely that the promoter of dxs was also not strong in Rhodococcus. One way to improve the crtU expression was to use a stronger promoter. The promoter of the Cm'gene on pDA71 (SEQ ID NO : 22) was a likely candidate. Another advantage of using the promoter of the Cmr gene was that this Cm'gene only expresses in Rhodococcus and not in E. coli, which circumvented toxicity or stability problems in E. coli during sub-cloning steps. The Cmr gene on pDA71 has been used as an antibiotic marker cassette and the gene and its promoter was previously localized to the 1.8 kb BbrPI-Stul fragment (De Mot, R. , et al., Microbiology, 143: 3137-3147 (1997); GenBankE Accession &num AJ308376). The Cm'gene including 500bp of its upstream region was PCR amplified from pDA71 plasmid DNA, using forward primer cm_F (5'- ccatcgaagtaccgtcacgtgcac-3'; SEQ ID NO : 23) and reverse primer cm_R (5'-ccatcaaffgtcaggotgggacggtttcot-3'; SEQ ID NO : 24). The forward primer, cm_F, spanned the BbrPi site (bold). The reverse primer, cm_R, contained the complementary sequence of a stop codon of the gene shown in bold. Both primers contained Ncol sites (underlined) and cm_R contained a Mfel site (italicized). The 1640 bp PCR product was

cloned in the pCR2.1-TOPO cloning vector (Invitrogen), resulting in pDCQ141. The 1.6kb Ncol fragment from pDCQ141 was subeloned into the Ncol site in pRhBR171 to construct the new E. coli-Rhodococcus shuttle plasmid pDCQ142 with the Cmr gene. The 1.6 kb EcoRl fragment of pDCQ138 DNA containing the cttu gene and its ribosomal binding site was subcloned into the unique Mfel site immediately downstream of Cmr gene in pDCQ142. The resulting construct pDCQ143 contained the cite gene downstream of the Cmr gene, with both Cmr and crû oriented in the same direction. The two genes were separated by 62 bp DNA without an apparent transcriptional termination site. The crtU gene could presumably be transcribed as a polycistronic message with the Cor gene by its promoter. Plasmid pDCQ143 and the vector control pDCQ142 were electroporated into 47042 (crut0-). Cells were grown in 200 mL LB with 0.05% Tween-80, 10 ug/mL tetracycline and 40 ug/mL chloramphenicol at 30° C for 2 days. Carotenoids were extracted from cell pellets with 20 mL acetone, dried under nitrogen and dissolved in 1 mL methanol. Each sample of 0.1 mL was used for HPLC analysis as described in Example 2.

In 47042 (cato-), with the pDCQ142 vector control, y-carotene eluted at 15.3 min. In 47042 (cHO-), with pDCQ143 containing crû, no carotenoid eluted at 15.3 and all carotenoid eluted at 15.0 min. Since the carotenoid that eluted at 15.0 min has the same absorption spectra as y-carotene (which eluted at 15.3 min), it was not certain that the carotenoid at 15.0 min is a new carotenoid produced by pDCQ143 or whether the y-carotene eluted at slightly different position due to HPLC variation. To distinguish these two possibilities, equal amounts of carotenoids extracted from the pDCQ142 control strain and the strain with pDCQ143 were mixed and loaded on HPLC. Two carotenoids peaks were observed with 41% eluting at 15.0 min and 59% eluting at 15.3 min. This strongly suggested that the carotenoid produced from the strain containing pDCQ143 was different from the y-carotene produced from the pDCQ142 vector control strain. The crtUexpressed downstream of Cm'gene on pDCQ143 achieved nearly 100% conversion of y-carotene to the new carotenoid, which was analyzed as described in Example 6.

EXAMPLE 6 Confirmation of Synthesis of Chtorobactene in Rhodococcus by LC-MS Chlorobactene exhibits the same absorption spectra as y-carotene.

In order to confirm synthesis of chlorobactene in Rhodococcus, molecular

weight of the carotenoids from different Rhodococcus strains were determined by LC-MS. Carotenoids were extracted as described above from wild type Rhodococcus erythropolis ATCC 47072,47072 (crut0-), and 47072 (cato-) containing pDCQ143. Each sample of 50 p. L was run on a Zorbax 2.1 x 150 mm SB-C18 LC column isocraticly with 75% acetonitrile and 25% acetone for 30 minutes (Agilent Technologies, CA). The mass spectrometer (a Micromass Quattro LC triple quadrapole, Micromass Limited, UK) was scanned from 100 to 1000 AMU's in 0.9 seconds with an 0.1 second interscan delay in APCI (Atomopheric Pressure Chemical lonization) mode with the corona discharge needle at 3 KV and the APCI probe at 450 degrees centigrade. Results of LC-MS analyses are shown in Figure 5. As expected, wild type ATCC 47072 produced 4-keto-y- carotene with a molecular weight of 550 Dalton (Figure 5), 47072 (crut0-) produced y-carotene with molecular weight of 536 Dalton, and 47072 (crtO-) containing pDCQ143 produced a new carotenoid with molecular weight of 532 Dalton. The 4 Dalton difference of molecular weight between y-carotene and the new carotenoid was consistent with the aromatization of the 0-ionone ring of the y-carotene by two additional steps of desaturation. The new carotenoid produced in 47072 (crtO-) containing pDCQ143 was chlorobactene, the aromatized product from y-carotene catalyzed by CrtU.

EXAMPLE 7 Production of Chlorobactene in the Wild Type Rhodococcus erythropolis As mentioned in the Example 2, the carotenoid ketolase CrtO (SEQ ID NO : 8) found in the wild type Rhodococcus erythropolis ATCC 47072 would compete with the heterologous aromatase CrtU for the y-carotene substrate. The crfU gene was therefore expressed in the 47072 (cffO-) as described above. The question of whether significant amount of chlorobactene could be produced from the wild type ATCC 47072 host was also addressed. The wild type strain has no antibiotic selection for the host and higher growth rate. Plasmid pDCQ143 containing the Brevibacterium crtU expressed downstream of the Cm'gene was transformed into the wild type Rhodococcus ATCC 47072 strain and transformants were selected on LB with 40 pg/mL chloramphenicol plates.

Carotenoids were extracted from the transformants and analyzed as described in Example 2. HPLC analysis indicated that majority of the carotenoids produced was chlorobactene. Less than 10% of carotenoids

present was 4-keto-y-carotene, which eluted at 14.5 min (absorption maximum at 465 nm). Samples were also concentrated and spotted on a 250 pm thickness Silica Gel 60 plate (EM Separations Technology, Gibbstown, NJ) for thin layer chromatography (TLC). Figure 6 illustrated comparison of carotenoids from wild type Rhodococcus 47072,47072 (cet0-), 47072 with pDCQ143, and 47072 (cftO-) with pDCQ143 separated on the Silica TLC in 7.5% acetone + 92.5% hexane. 4-keto-y-carotene in the wild type, y-carotene in 47072 (crtO-), and chlorobactene in cato- (pDCQ143) was previously confirmed by LC/MS as shown in Figure 5. In the 47072 carrying pDCQ143, a mixture of chlorobactene and 4-keto-y- carotene was produced as expected. The fact that majority of carotenoids in the mixture was chlorobactene indicated that it was feasible to use the wild type ATCC 47072 as the crtU expression host for the production of chlorobactene. The crû expressed from the pDCQ143 plasmid competed well with the single copy cet0 encoded on the chromosome of ATCC 47072.