Patent Application
20050089884
The present application includes a Sequence Listing filed on one (1) CD-R disc, containing a single file named PO13-B015 (Sequence) Copy Sent to USPTO.doc, having 980 kilobytes, last modified on Jan. 30, 2004 and recorded on Jan. 30, 2004. The Sequence Listing contained in said file on said disc is incorporated herein by reference in its entirety.
The present invention relates to genes for the synthesis of an FR-008 polyketide derived from Streptomyces sp. FR-008.
Polyketides, which are compounds synthesized from 2-carbon units by the action of polyketide synthases through a series of condensations and subsequent modifications, are produced in several bacteria, including molds or actinomycetes (Robinson, J. A., Philos. Trans. R. Soc. Lond. B. Biol. Sci., 332:107-14, 1991; Hopwood, D. A., Curr. Opin. Biotechnol., 4:531-537, 1993). The polyketides are physiologically active molecules having highly various structures and the species thereof includes many compounds having various activities.
Generally, a production method using such natural polyketides has encountered a difficulty in industrial application because of its low efficiency, and also faced many economic and technical difficulties in producing polyketide compounds by the prior chemical method.
Owing to such concerns, studies to efficiently biosynthesize the natural polyketides were conducted, and a recombinant production method using related genes started to be developed. Namely, studies, including cloning, assay and operation using a recombinant technology of genes coding for polyketide synthases, were performed, and owing to such technologies, the production of polyketides at a higher level than those occurring in an environment or host, which does not produce polyketides, became possible (WO 93/13663; 95/08548; 96/40968; 97/02358; 98/27203; 98/49315; U.S. Pat. Nos. 4,874,748, 5,063,155; 5,098,837; 5,149,639; 5,672,491; 5,712,146; 5,830,750; 5,843,718; Fu, H. et al., Biochem., 33:9321-6, 1994; McDaniel, R. et al., Science, 262:1546-50, 1993; Kuhstoss, S. et al., Gene, 183:231-6, 1996).
The natural polyketides is synthesized by the continuous activity of about 50 species of enzymes, so-called polyketide synthases (PKSs), and carrier proteins, in a similar manner as the synthesis of fatty acids (Robinson, J. A., Philos. Trans. R. Soc. Lond. B. Biol. Sci., 332:107-14, 1991; Hopwood, D. A., Curr. Opin. Biotechnol., 4:531-7, 1993). For example, the core structure of erythromycin is made by about 25 continuous enzymatic reactions. PKSs include two types, one of which is a modular type and the other one is an iteractive type (Khosla, C., Chem. Rev., 97:2577-90, 1997).
In the modular PKSs (type I), all steps for the assembly and modification of carbon chains occur by different catalytic sites, but in the iteractive PKSs (type II), one catalytic site is utilized one or more times on biosynthetic pathway. A typical modular PKS consists of several large peptides, which are divided into a loading module, multiple extension modules and a releasing module in the direction from the N-terminal end to the C-terminal end (Hopwood and Sherman, Annu. Rev. Genet., 24:37-66, 1990).
The loading module consists of acyl-transferase (AT) and acyl carrier protein (ACP). Furthermore, the multiple extension modules basically comprise ketosynthase (KS), AP and ACP, and often comprise an enzyme for modifying the β-carbon of the extended polyketide chain in addition to such basic modules. Finally, the releasing module comprises thioesterase (TE) and in some cases, cyclase activity.
Generally, the loading module binds to a first building block used in synthesizing polyketides, and acts to transfer the building block to the first extension module. AT on the multiple extension modules recognize a certain acyl-CoA (acetyl or propionyl) and transfer it to ACP in the form of thiol ester. At the same time, AT on each module transfers the malonyl group of a certain malonyl-CoA (malonyl or α-substituted malonyl) to ACP on the module in a thiol ester form. Then, the acyl group on the loading module is transferred to KS on the first module by transesterification. The transferred acyl group covalently binds to the α-carbon of the malonyl group, and dicarboxylation occurs to produce a new acyl group with a backbone two carbons longer than the loaded unit (extension or elongation).
The polyketide chain grown by two carbon atoms of each extension module is transferred from the extension module to the next extension module in the form of covalently bonded thiol ester, and subjected to the above-mentioned procedure. In addition to KS, AT and ACP involved in the formation of the C—C bonds, each module often comprises enzymes for modifying the β-keto group of 2-carbon units just added to the extended polyketide chain before this chain is transferred to the next module. These modifier enzymes include ketoreductase (KR) reducing a keto group into alcohol, dehydroreductase (DH) forming a double bond by dehydration of alcohol, and enoyl reductase (ER) converting a double bond into a single bond.
There are modules having no modifier enzymes, and modules comprising KR(1), KR+DH(2) or KR+DH+ER(3). According to such modifier enzymes, the oxidation state of the β-carbon of each 2-carbon unit is determined (0=ketone; 1=alcohol; 2=double bond; and 3=single bond). Polyketide products will vary depending on the specificity of AT of each module and the kind of the modifier enzymes. If the extended polyketide chain is transferred to the last module of PKS, it will meet the releasing module or thioesterase active site in which polyketide is truncated into a ring form. Furthermore, polyketide can be further modified by adding a carbohydrate or methyl group to the core molecule thereof or by a tailoring enzyme (TE) inducing other modifications.
As described above, polyketides are formed by the condensation of carboxylic acid units using the continuous action of PKS. Most of PKS genes in microorganisms are known as being present in a cluster form in the microbial chromosome (Hopwood, Chem. Rev., 97:2465-97, 1997). In 1960, DEBS genes involved in the formation of 6-deoxyerythronolide B (6 dEB) were first reported (Cortes et al., Nature, 348:176-8, 1990; Donadio et al., Science, 252:675-9, 1991), and then, the cloning and base sequence of whole or partial genes for the synthesis of eleven polyketides were reported up to now. The genetic information of the polyketides reported up to now is summarized in Table 1 below.
FR-008 polyketide produced in Streptomyces sp. FR-008 is a heptaene macrolide having aglycone containing 4-aminoacetophenone, as in candicidin D (Yuan and Zhou, J. Huazhong Agricult. Univ., 9:209, 1990). Since the FR-008 polyketide has antifungal activity and also high toxicity against mosquito larva, they are highly useful in agricultural and medical fields (Liang and Zhou, Chinese J. Biotech., 3:130-6, 1987). The FR-008 polyketide has the following formula:
Genes for the synthesis of the FR-008 polyketide was first cloned in 1994 (Hu et al., Mol. Microbiol., 14:163-72, 1994). Hu et al. constructed a chromosomal library of Streptomyces sp. FR-008 using a cosmid vector and screened 16 cosmid clones containing PKS genes from the same, but the base sequence of the genes was not reported.
Therefore, the present inventors have cloned genes for the biosynthesis of FR-008 polyketides derived from Streptomyces sp. FR-008, and identified their base sequence, thereby reaching the present invention.
Accordingly, a primary object of the present invention is to provide the base sequence of genes coding for proteins involved in the biosynthesis of FR-008 polyketides derived from Streptomyces sp. FR-008.
Another object of the present invention is to provide a method for producing FR-008 polyketide variants comprising the step of cultivating a mutant strain where one or more of the genes in the just above-mentioned base sequence were deleted or inactivated.
In the present invention, the base sequence of whole genes (138,203 bp) coding for proteins involved in the biosynthesis of FR-008 polyketides derived from Streptomyces sp. FR-008 was analyzed. This base sequence comprises genes coding for ketosynthase (KS), acyl transferase (AT), acyl carrier protein (ACP), ketoreductase (KR), dehydratase (DH) and enoyl reductase (ER) domains involved in FR-008 polyketide biosynthesis, and also genes coding for modifier enzymes, such as ABC transporter, cytochrome P-450 monooxygenase, ferredoxin, thioesterase, sugar biosynthetic protein, FAD-dependent monooxygenase, 4-amino-4-deoxychorismate (ADC) synthase and ADC lyase.
In one embodiment, the present invention provides the whole base sequence (SEQ ID NO: 1) of a gene cluster for the biosynthesis of FR-008 polyketides. SEQ ID NO. 1 is a 138,203 bp gene base sequence having 21 open reading frames consisting of fscA, fscB, fscC, fscD, fscE, fscF, fscTI, fscTII, fscRI, fscRII, fscRIII, fscRIV, fscP, fscFE, fscTE, fscMI, fscMII, fscMIII, fscO, pabAB and pabC involved in FR-008 polyketide biosynthesis.
In the present invention, FscA, FscB, FscC, FscD, FscE and FscF represent modular (type I) polyketide synthase, FscTI and FscTII represent ABC transporter protein, FscRI, FscRII, FscRIII and FscRIV represent regulator protein, FscP represents cytochrome P-450 monooxygenase, FscFE represents ferredoxin protein, FscTE represents thioesterase, FscMI represents glycosyltransferase, FscMII represents GDP-ketosugar aminotransferase, FscMIII represents GDP-mannose-4,6-dehydratase, FscO represents FAD-dependent monooxygenase, ADC represents 4-amino-4-deoxychorismate synthase, and PabC represents ADC lyase.
In another embodiment, the present invention provides a method for producing FR-008 polyketide variants comprising the step of cultivating a mutant strain where one or more genes in the above-mentioned base sequence were deleted or inactivated, as well as FR-008 polyketide variants produced by this method.
In this embodiment, the FR-008 polyketide variants are preferably FR-008-V, FR-008-VI, FR-008-VII, FR-008-VIII, FR-008-IX, FR-008-X, FR-008-XI and FR-008-XII, the variants having the following formula:
In another embodiment, the present invention provides a modular polyketide synthase (PKS) of FR-008 comprising the following amino acid sequences (a)-(f), as well as a gene coding for the same:
In this embodiment, the above-mentioned gene preferably has the base sequence of SEQ ID NO: 1, residues 20,927-26,158 (fscA), 28,983-60,860c (fscC), 60,962-77,587c (fscB), 77,983-84,132c (fscF), 84,170-107,485c (fscE) and 107,496-136,148c (fscD). The indication “c” on the sequence according to the present invention shows that transcription occurs in a complementary strand in the opposite direction.
In yet another embodiment, the present invention provides an ABC transporter having the amino acid sequence of SEQ ID NO: 8 or 9, as well as a gene (fscTI or fscTII) coding for the same. The fscTI and fscTII genes preferably have the base sequences of residues 26,333-27,340 and 27,561-28,280 of SEQ ID NO: 1, respectively.
In still another embodiment, the present invention provides a regulator protein having the amino acid sequence of SEQ ID NO: 10, 11, 12 or 13, and a gene (fscRI, fscRII, fscRIII or fscRIV) coding for the same. The fscRI, fscRII, fscRIII and fscRIV genes preferably have the base sequences of residues 3,150-3,818c, 4,377-7,205c, 7,210-10,320c and 10,298-13,315c of SEQ ID NO: 1, respectively.
Furthermore, the present invention provides a cytochrome P-450 monooxygenase having the amino acid sequence of SEQ ID NO: 14, as well as a gene (fscP) coding for the same. The fscP gene preferably has the base sequence of residues 16,106-17,287 of SEQ ID NO: 1.
In another embodiment, the present invention provides a ferredoxin protein having the amino acid sequence of SEQ ID NO: 15, as well as a gene (fscFE) coding for the same. The fscFE gene preferably has the base sequence of residues 17,334-17,528 of SEQ ID NO: 1.
In still another embodiment, the present invention provides a thioesterase having the amino acid sequence of SEQ ID NO: 16, and a gene (fscTE) coding for the same. The fscTE gene preferably has the base sequence of residues 17,556-18,413 of SEQ ID NO: 1.
In further another embodiment, the present invention provides a glycosyltransferase having the amino acid sequence of SEQ ID NO: 17, and a gene (fscMI) coding for the same. The fscMI gene preferably has the base sequence of residues 13,522-14,898 of SEQ ID NO: 1.
In yet another embodiment, the present invention provides a GDP-ketosugar aminotransferase having the amino acid sequence of SEQ ID NO: 18, and a gene (fscMII) coding for the same. The fscMII gene preferably has the base sequence of residues 14,953-16,011 of SEQ ID NO: 1.
In still another embodiment, the present invention provides a GDP-mannose-4,6-dehydratase having the amino acid sequence of SEQ ID NO: 19, and a gene (fscMIII) coding for the same. The fscMIII gene preferably has the base sequence of residues 136,558-137,766c of SEQ ID NO: 1.
Moreover, the present invention provides a FAD-dependent monooxygenase having the amino acid sequence of SEQ ID NO: 20, and a gene (fscO) coding for the same. The fscO gene preferably has the base sequence of residues 574-1,950c of SEQ ID NO: 1.
In another embodiment, the present invention provides an ADC synthase having the amino acid sequence of SEQ ID NO: 21, and a gene (pabAB) coding for the same. The pabAB gene preferably has the base sequence of residues 18,610-20,781 of SEQ ID NO: 1.
Also, the present invention provides an ADC lyase having the amino acid sequence of SEQ ID NO: 22, and a gene (pabC) coding for the same. The pabC gene preferably has the base sequence of residues 2,264-3,037c of SEQ ID NO: 1.
Hereinafter, the present invention will be described in further detail.
As a result of prior studies on the cloning of genes for the synthesis of FR-008 polyketides, a chromosomal library of Streptomyces sp. FR-008 was constructed using a cosmid vector, and 16 cosmid clones containing PKS genes were screened from the chromosomal library (Hu et al., Mol. Microbiol., 14:163-72, 1994). Meanwhile, a Streptomyces sp. FR-008 strain used in the present invention was deposited under the accession number KCTC 10529BP on Oct. 20, 2003 with the Korean Collection for Type Cultures (KCTC), Korea Research Institute of Bioscience and Biotechnology, 52, Oun-dong, Yusong-Ku, Taejon 305-333, Republic of Korea.
In the present invention, all of the gene base sequences of cosmid clones pHZ145, pHZ137, pHZ220 and pHZ194 among 16 cosmid clones constructed in the prior art and also plasmid clone pJTU6 were analyzed. Also, the gene base sequence of a new plasmid clone pJTU10 was analyzed. On the basis of this sequencing, a gene base sequence (138,203 bp) involved in the biosynthesis of FR-008 polyketides was identified from base sequences other than the cosmid and plasmid DNA base sequences (see,
The present invention will hereinafter be described in further detail by examples. It will however be obvious to a person skilled in the art that these examples are given for illustrative purpose only, and the scope of the present invention is not limited to or by these examples.
pHZ145, pHZ137, pHZ220 and pHZ194 among 16 cosmid clones constructed by Hu et al. (Hu et al., Mol. Microbiol., 14:163-72, 1994) and also plasmid clone pJTU6 were used in gene sequencing.
Cosmid DNA was fragmented by ultrasonic treatment and then subjected to agrose gel electrophoresis. The produced DNA fragments of suitable sizes were isolated from agarose gels, and their both ends were polished with T4 DNA polymerase and T4 DNA kinase. The polished DNA fragments were cloned into plasmid pUC19 digested with restriction enzyme SmaI. For cosmids pHZ145, pHZ137 and pJTU6, the isolated DNA fragments of about 5 kb size were used. For pHZ220 and pHZ194, the isolated DNA fragments of about 1 kb size were used.
Gene sequencing reaction was carried out in each pUC19 clone using a BigDye Terminator Cycle Sequencing kit, version 2 (Applied Biosystems, USA) by dye terminator sequencing chemistry. For gene sequencing, an ABI Prism 3700 DNA analyzer (Applied Biosystems, USA) was used. As an analysis program, the phrap version 0.990329 (www.phrap.org) was used. For contig viewing and editing, the consed version 11 (www.phrap.org) was used. The gene sequencing was performed at least six times for each base.
A whole base sequence analyzed by the method as described above was 138,203 bp in length. The G+C content in the whole base sequence was about 74.6%, and whole open reading frames were 20 in number. Gaps in shotgun sequencing were filled either by primer walking or extra clone sequencing. Gene base sequences for open reading frame analysis and conserved domain database search were analyzed with FramePlot 2.3.2 (www.nih.go.jp/˜jun/cgi-bin/frameplot.pl).
This gene sequencing revealed that parts of FR-008 polyketide biosynthetic genes were deleted in the four cosmid clones resulted from the study of Hu et al. and the one-plasmid clone. Hu et al. reported that the fourth plasmid pHZ194 contained the latest part of the FR-008 polyketide biosynthetic genes (Hu et al., Mol. Microbiol., 14:163-72, 1994), but the actual gene sequencing revealed that the pHZ194 cosmid clone had parts of the FR-008 polyketide biosynthetic genes, parts of chromosomes derived from Streptomyces sp. FR-008, and two backbone cosmids pHZ132.
Thus, in order to secure the whole base sequence of the FR-008 polyketide biosynthetic genes, a cosmid clone containing the last part of the FR-008 polyketide biosynthetic genes was searched from a cosmid library reported on the article of Hu et al. In a search method, Southern hybridization was performed using probes constructed by polymerase chain reaction (PCR) from parts of the FR-008 polyketide biosynthetic genes contained in the pHZ194 cosmid clone. Namely, PCR was performed using primers of SEQ ID NOs: 23 and 24 as described below, thereby obtaining a PCR product of a 218 bp length. This PCR product was subjected to Southern hybridization using probes, and as a result, two cosmid clones (pJTU10 and pJTU11) could be identified. The cosmid clone pJTU10 was selected and selected to gene sequencing as described above, thereby securing a whole gene sequence (138,203 bp) involved in FR-008 polyketide biosynthesis.
The analysis result of each gene is as follows. The base sequences of residues 20,927-26,158 and 28,983-107,485 of SEQ ID NO: 1 code for PKS genes involved in the synthesis of FR-008 macrolide backbones. In such two regions, there are present six open reading frames (fscA, fscB, fscC, fscD, fscE and fscF), and between the two regions, there are present genes coding for two ABC transporters. The base sequence and amino acid sequence position of each of the PKS genes are summarized in Table 2 below.
fscA codes for a loading module and extension module 1. Each of amino acid sequences for PKS domains of the loading module and extension module in fscA is summarized in Table 3 below.
fscC codes for six extension modules (extension modules 5, 6, 7, 8, 9 and 10). Each of amino acid sequences for PKS domains of the six extension modules in fscC is summarized in Table 4 below.
fscB codes for three extension modules (extension modules 4, 3 and 2). Each of amino acid sequences for PKS domains of the three extension modules in fscB is summarized in Table 5.
FscF codes for one extension module (extension module 21) and one thioesterase. Each of amino acid sequences for PKS domains of the one extension module and one thioesterase in FscF is summarized in Table 6 below.
FscE codes for four extension modules (extension modules 17, 18, 19 and 20). Each of amino acid sequences for PKS domains of the four extension modules in fscE is summarized in Table 7 below.
fscD codes for six extension modules (extension modules 16, 15, 14, 13, 12, and 11). Each of amino acid sequences for PKS domains of the six extension modules in fscD is summarized in Table 8.
There are the total 15 open reading frames for PKS enzymes, and such open reading frames comprise modifier enzymes, transcription regulatory proteins, and ABC transporters. The base sequence and amino acid sequence position of each gene are summarized in Table 9 below.
3,150-3,818c
4,377-7,205c
10,298-13,315c
136,558-137,766c
2,264-3,037c
The function of the proteins that are coded by genes involved in FR-008 biosynthesis is summarized in Tables 10 and 11 below.
The amino acid sequence of proteins for which fscA through fscF genes code was analyzed by comparison with the amino acid sequence of the existing type I PKS protein. As a result, the function of each gene is shown in
The transcription of the fscA gene occurred in the opposite direction to the fscC, fscB, fscF, fscE and fscD genes, and 21 condensation reactions were required in FR-008 polyketide biosynthesis. Thus, it could be found that there were the total 21 modules within the fscA through fscF genes (see,
The FscA protein has a loading module involved in the starting of FR-008 aglycone biosynthesis and also a first condensation module consisting of KS-AT-ACP. The loading module has CoA ligase (CoL) and ACP domains. The CoL domain was also found in genes for the biosynthesis of rifamycin, pimaricin and candicidin (Schwecke, T. et al., Proc. Natl. Acad. Sci. USA, 92:7839-43, 1995; Aparicio, J. F. et al., J. Biol. Chem., 274:10133-9, 1999; Campelo and Gil, Microbiology, 148:51-9, 2002).
The FscB protein has 2-4 extension modules involved in 2-4 extension procedures of FR-008 agylcone biosynthesis. Three AT domains within FscB have the amino acid sequence of RVDVVxxxxxxxMxS(A)xAxxW as a conservative sequence at the N-terminal end of the active site GHSxG, in which this amino acid sequence is specific for methylmalonate (Tang, L. et al., Gene, 216:255-65, 1998; Haydock, S. F. et al., FEBS Lett., 374:246-8, 1995). Module 2 has KS, mAT, KR and ACP domains, and module 3 has a DH-ER-KR reduction loop. DH4 within module 4 is found to be deleted for the conservative sequence HPLL at its N-terminal end and also 56 amino acids shorter at its C-terminal end, and thus, regarded as having no activity. However, a CanP2 protein involved in 2-4 extension procedures of candicidin aglycone biosynthesis was reported to have no inactive DH domain described above, although the N-terminal end of a KS domain was not identified.
FscC and FscD proteins provide two other examples of hexamodular proteins. FscC has a 10,625-amino acid length, and FscD has a 9,550-amino acid length. FscC combines six double bonds of the total seven double bonds. Modules 5-10 involved in condensation steps 5-10 all have a DH-KR reduction loop. The total six DH domains have the conservative active site GXGXXGXXXA. All the AT domains within FscC are specific for acetate. FscC that is a protein having the same six DH-KR reduction loops, is in a form which was not found up to date. On the other hand, NysC, Raps2 and AmphC all have five DH-KR reduction loops and also DH-ER-KR reduction loops.
FscD consists of modules 11-16 involved in extension procedures 11-16 in the formation of the FR-008 macrolactone ring. The module 11 combines a KS-AT-DH-KR-ACP domain structure with the last double bond. The extension procedures 10-11 are involved in the formation of two adjacent double bonds (last two of seven double bonds), but interestingly, a gene for the module 11 on the Streptomyces FR-008 chromosome is located about 110-kb apart from the gene for the module 10.
The module 13 has a methylmalonate-specific AT domain, and acts to construct a KS-mAT-KR-ACP domain in the extension procedure so as to insert a methylmalonate-CoA extender. The modules 12, 14, 15 and 16 all have the same KS-AT-KR-ACP domain structure, but exceptionally KR15 is inactivated. Furthermore, KR15 is deleted of two conservative G amino acid residues at its NADP-binding site, and also deleted of 10 amino acids at its inside. In a similar manner to nystatin, amphotericin B, and pimaricin, the inactive KR domain is a carboxyl group at C-28. This illustrates the formation of six ketalic rings between positions C-25 and C-28 of most polyketides (Brautaset et al., Chem. Biol., 7:395-403, 2000; Caffrey et al., Chem. Biol., 8:713-23, 2001; Aparicio et al., Chem. Biol., 7:895-905, 2000).
An FscE protein forms modules 17-20 which is involved in extension procedures 17-20 of FR-008 polyketide biosynthesis. The ER domain of the module 17 is 67 amino acids shorter than the N-terminal end of the ER domain of the modules 3, 18 and 20, and thus becomes inactive. The module 19 has the domain structure of KS-AT-KR-ACP. At the active site of the KR domain of the module 19, tyrocine is replaced with leucine, so that the active site becomes inactive.
FscF forms the module 21, which has a chain terminator thioesterase domain at the C-terminal end. The activity of a DH domain in FscF is not required in the last extension procedure. This inactivity also appears in Nystanin (NysK), rifamycin (RifB and RifC) and amphotericin (AmphJ) polyketides. The KS-AT-KR-ACP domain of FscF adds an —OH group to position C-40 of the FR-008 macrolactone ring. Furthermore, a TE domain enables FscF involved in the release of FR-008 polyketides to form lactone. The KR domain of the module 21 has no difference in active motifs and residues from the known KR domain, but it is guessed from the structure of FR-008 that this KR domain is inactive.
The TE domain of FscF and also an fscTE gene upstream of a pabAB gene were known as coding for thioesterase. Like two other TE activities present in gene clusters for antibiotic biosynthesis, such as pikromycin (PicAV), tyrosine (TylO), nystatine (NysE) and pimaricin (PimI), FscTE has high homology with TEII. However, FscTE has low homology with the TEI domain.
TEII is known as serving to remove a bad intermediate which may also interfere with extension procedures in polyketide biosynthesis. This maintains antibiotic production at a normal level (Butler, et al., Chem. Biol., 6:287-92, 1999; Tang, et al., Chem. Biol., 6:553-8, 1999). This is also supported by recent study results on the action of tyrosine TEII.
p-aminobenzoic acid(PABA) is a precursor substance of p-aminoacetophenone acting as a starting unit of FR-008 biosynthesis. Between fscTE and fscA, there is a pabAB gene which codes for 4-amino-4-deoxychorismate (ADC) synthase (see,
pabC upstream of the transcription regulatory gene fscRI consists of 257 amino acids and codes for 4-amino-4-deoxychorismate (ADC) lyase. It is presumed that PabC acts to convert ADC produced by PabAB into PABA or pyruvic acid or acts as aminotransferase or pyridoxal-5′-phosphate-dependent enzyme.
Four transcription regulatory genes (fscRI, fscRII, fscRIII, and fscRIV) are involved in FR-008 polyketide biosynthesis. All proteins for which such genes code belong to LuxR family transcription regulatory proteins. FscRI consists of 222 amino acids and has 60% homology with PleR present in a gene cluster for Streptomyces avermitilis polyene macrolide biosynthesis (GenBank accession number AB070949), and 59% homology with proteins coded by ORF4 regarded as being involved in the regulation of nystadin biosynthesis (Brautaset et al., Chem. Biol., 7:395-403, 2000). Also, it shows homology with several two-component response regulatory proteins.
The FscRII protein (942 amino acids) shows more than 84% homology with the first 595 amino acid sequence of a Streptomyces griseus-derived cho-pseudo protein known as being involved in candicidin synthesis. It is surprising that, in spite of the fact the FscRII protein is 347 amino acids longer than the cho-pseudo protein, this high homology is shown in the N-terminal 595 amino acids. FscRII exhibits 42% homology with a NysRII regulator protein (927 amino acids) involved in nystatin biosynthesis.
FscRIII has 1,036 amino acids and shows significant homology to NysRII (953 amino acids). An FscRIV protein has 1,005 amino acids and shows high homology to NysRI (966 amino acids). It is noteworthy that FscRII, FscRIII and FscRIV have homology to a transcription regulatory protein found in regions adjacent to a cholesterol oxydase-cytochrome P450 operon derived from Streptomyces sp. SA-COO, as in NysRIII, NysRII and NysRI.
FscMI(458 amino acids) shows high homology with UDP-glucuronosyltransferase of eukaryotic cells, and is presumed to be involved in the attachment of sugar to aglycone of FR-008 at position C-21 (see,
FscMII(352 amino acids) is similar to aminotransferase involved in the biosynthesis of perosamine (4,6-dideoxy-4-aminomannose). An fscMIII gene codes for 402 amino acids similar to GDP-mannose-4,6-dehydratase. It was reported that the biosynthetic pathway of mycosamine (3,6-dideoxy-3-amino-D-mannose) included the isomerization of GDP-6-deoxy-4-ketomannose into GDP-6-deoxy-3-ketomannose and the production of GDP-mycosamine by amino group transfer (Caffrey et al., Chem. Biol., 8:713-23, 2001). Thus, each of FscMIII and FscMII is presumed to be involved in the biosynthetic pathway of mycosamine.
As a result of searching in the GenBank database, FscMI has 54.7% homology with AmphDI, 52.4% homology with NysDI and 60.5% homology with PimK, FscMII has 92% homology with CanA, and 74-76% homology with AmphDII, NysDII and PimC, and FscMIII has 77-79% homology with AmphDIII, NysDIII and PimJ.
All the proteins as described above are presumed to be involved in the biosynthesis and attachment of mycosamine in the amphotericin, nystatin, pimaricin or candicidin gene cluster. This agrees with the fact that FscMI, FscMII and FscMIII are involved in the biosynthesis and attachment of mycosamine in the FR-008 gene cluster.
An fscP gene and an fscFE gene code for cytochrome P450 monooxygenase and ferredoxin, respectively. These genes form a P450 monooxygenase system, and it is presumed that FscP acts to produce a carboxyl group at position C-18, and FscFE is involved in electron transfer in the P450 system (O'Keefe & Harder, Mol. Microbiol., 5:2099-105, 1991).
An fscO gene downstream of a pabC codes for FAD-dependent monooxygenase (see,
It is noteworthy that FscFE (64 amino acids) has 100% homology with CanF involved in candicidin biosynehtsis. FscP shows 96% homology with CanC. FscTI and FscTII show 85% and 76% homology to CanRA and CanRB, respectively. Such a high homology between the candicidin and the protein involved in FR-008 polyketide biosynthesis suggests that two synthetic gene clusters have a relatively recent common origin.
FR-008 polyketides have aglycone containing the same 4-aminoacetophenone as that of candicidin D which is a polyketide produced in Streptomyces griseus (Yuan and Zhou, Chinese J. Biotechnol., 7:142-7, 1991; Yuan and Zhou, J. Huazhong Agricult. Univ. 9:209, 1990). Streptomyces sp. FR-008 shows significant biological differences from the candicidin-producing Streptomyces griseus IMRU3570, in that Streptomyces sp. FR-008 has two linear plasmids pHZ227 (130 kb) and pHZ228 (30 kb) whereas Streptomyces griseus IMRU3570 has no any linear plasmid, and in that sporulation easily occurs in the Streptomyces sp. FR-008 but does not easily occur in the Streptomyces griseus IMRU3570. In spite of such significant differences between the two strains, polyketides produced by the two strains have aglycones containing the same 4-aminoacetophenone, respectively, and thus have very high homology with genes involved in aglycone synthesis.
Campelo et al. reported the gene base sequence of parts (39,314 bp) of genes for candicidin synthesis (Campelo and Gil, Microbiology, 148:51-9, 2002). The result of the comparison between this gene base sequence and the base sequence of the present invention showed a very high homology between the two sequences. The comparison results are summarized in Table 12 below.
At the intervals between the compared positions in Table 12, the two gene base sequences did not show high homology with each other. The results of the comparison between the two gene base sequences at such intervals are given in Table 13 below. A 39,414 bp gene of candicidin codes for proteins corresponding to the KR domain of the module 4 in fscRII, fscRIII, fscRIV, fscMI, fscMII, fscP, fscFE, fscTE, pabAB, fscA, fscTI, fscTII and fscB present in positions 5,421-44,805 of SEQ ID NO: 1 for FR-008. This codes for canA (aminotransferase), canC (cytochrome P450), canF (ferredoxine), canT (thioesterase), pabAB (PABA synthase), canRA (ABC transporter), canRB (ABC transporter) and moiety orf1 (Cho-like protein).
FR-008 produced by Streptomyces sp. and candicidin complex produced by Streptomyces griseus were analyzed by HPLC. The analysis results showed that the two antibiotics could not be definitely distinguished from each other (see,
Since the m/z 1108 of FR-008II is equal to the m/z 1108 of FR-008III, FR-008II is regarded as the isomer of FR-008III. Moreover, the two substances showed the same peaks as those of candicidin-II and candicidin-III (m/z 1108) (
It was found that the hemiketal ring was not found between positions C-15 and C-19 of FR-008II, ketal at δ96.9 was located at position C-1 of mycosamine, and four ketone groups between δ202.2 and δ196.7 were located at positions C-3, C-7, C-15 and C-43, respectively. Moreover, it was found that six hemiketal rings attributed to the isomer FR-008-III which equilibrates in the FR-008 complex.
That amphotericin B, polyhydroxy ketone, is present in a structure equilibrating in the form of various cyclic ketals was demonstrated by the fact that a six-membered ring cyclic ketal structure and a structure having no six-membered ketal ring are present together.
A difference in molecular weight between FR-008-I and FR-008-II (III) is 2 Daltons, and this difference can be explained by the fact that a mixture having a keto group and a hydroxy group occurs at position C-3, since K21 activity is insufficient to catalyze keto reduction and thus incompletely reduced polyketide chains are passed to a stage mediated by TE(thioesterase), or TE can not distinguish the polyketides of a form reduced by KR21 from the polyketides of a oxidized form before cyclization.
FR-008-IV/candicidin-IV with MS1 m/z 1092 shows a molecular weight difference of 16 Daltons from FR-008-II(III)/candicidin-II(III)(MS1 m/z 1108). There are several alternative possibilities for the structural between the FR-008-IV/candicidin-IV and the FR-008-II(III)/candicidin-II(III) functioning as FscO (FAD-dependent monooxygenase).
To disrupt FscMI, a 6.6 kb KpnI fragment of pHZ145 containing fscMI was inserted into pIJ2925 (Janssen and Bribb, Gene, 124:133-4, 1993) so as to construct pJTU26. A 1.4 kb EcoRV-SmaI DNA fragment containing an apramycine-resistant gene was inserted into the BalI site (located 564 bp downstream of the start codon of fscMI) of the pJTU26 so as to construct pJTU38. A 8.0 kb BglII DNA fragment of the pJTU38 was inserted into the BamHI site of pHZ1358 (Sun et al, Microbiology, 148:361-71, 2002) so as to construct pJTU56. Streptomyces sp. FR-008 was transformed with pJTU56 by conjugation using E. coli ET12567 (MacNeil et al., Gene, 115:119-25, 1992; and Mazodier et al., J. Bacteriol., 171:3583-5, 1989) containing RP4 derivative pUZ8002, thereby giving mutant strain C101. This mutant strain was identified by PCR using primers of CS3 (SEQ ID NO: 25) and CS4 (SEQ ID NO: 26).
When the chromosomal DNA of a wild-type Streptomyces sp. FR-008 gene was used as a template, a 840 bp fragment was produced, but when its mutant strain was used, a 2.2 kb fragment was produced by insertion of a 1.4 kb apramycine-resistant gene (acc(3)IV).
A culture medium of this mutant strain was analyzed by LC/MSn, and the results showed that polyketides V, VI, VII, VIII and IX having no sugar attached thereto were produced, in addition to polyketides produced in wild-type Streptomyces sp. FR-008 (see,
To disrupt FscMII, a 5.5 kb SacI-KpnI fragment of pHZ145 containing an fscMII gene was inserted into pIJ2925 so as to construct pJTU27. A 1.4 kb PstI DNA fragment of the pJTU27 containing an apramycine-resistant gene of pHZ1358 was inserted into the PstI site (located 436 bp downstream of the start codon of fscMII) of the pJTU27 so as to construct pJTU31. A 6.9 kb BglII DNA fragment of the pJTU31 was inserted into the BamHI site of pHZ1358 so as to construct pJTU58. Streptomyces sp. FR-008 was transformed with pJTU58 by conjugation using E. coli ET12567 containing RP4 derivative pUZ8002, thereby giving mutant strain CS102. This mutant strain was identified by PCR using primers of CS5 (SEQ ID NO: 27) and CS6 (SEQ ID NO: 28).
When the chromosomal DNA of a wild-type Streptomyces sp. FR-008 gene was used as a template, a 508 bp fragment was produced, but when its mutant strain was used, a 1.9 kb fragment was produced by insertion of a 1.4 kb apramycine-resistant gene (acc(3)IV).
A culture medium of this mutant strain was analyzed by LC/MSn, and the results showed that polyketides V, VI, VII, VIII and IX having no sugar attached thereto and also polyketites X, XI and XII where amine transfer did not occur were produced, in addition to polyketides produced in wild-type Streptomyces sp. FR-008 (see,
The polyketide variants produced in Examples 8 and 9 were tested for antifungal activity, and the results showed that turbid inhibition zones where the growth of Saccharomyces cerevisiae is inhibited were formed at sites to which the culture mediums of the Streptomyces sp. FR-008 mutant strains CS101 and CS102 were added (see,
To disrupt transcription regulatory genes fscRII and fscRIII and moiety fscRI (6,414 bp DNA; 3,811-10,225 of GenBank accession number AY310323), cosmid pHZ2007 derived from cosmid pHZ1358 was constructed. The cosmid pHZ2007 was constructed in such a manner that 2.8 kb of an apramycine (aac(3)IV)- and erythromycin (ermE)-resistant gene cassette derived from cosmid pHZ1358 (Hu et al, Mol. Microbiol., 14:163-72, 1994) is interposed between a 1,518 bp BglII-BamHI DNA fragment serving as a left arm and a 1,952 bp SacI-BamHI DNA fragment serving as a right arm. This cosmid was transformed into Streptomyces sp. FR-008 by conjugation in the same manner as in Example 8, thereby giving a mutant strain. In this mutant strain, a 4.3 kb BglII-BamHI DNA fragment was identified by Southern hybridization using left and right arms derived from pHZ2007, and an acc(3)IV/ermE gene cassette, as probes. This mutant strain did not produce substances related with FR-008. From this result, it could be found that fscRII, fscRIII and fscRI were involved in FR-008 biosynthesis.
A 4.6 kb BamHI fragment was obtained from a pHZ145 cosmid and cloned into suicide vector pOJ260. Then, it was digested with FscPI and ligated to delete a 1,491 bp fragment containing a pabAB gene (in frame deletion, 497 amino acid). The resulting cosmid was transformed into Streptomyces sp. FR-008 by conjugation in the same manner as in Example 8. The pabAB-deleted mutant strain was analyzed by Southern hybridization. FR-008 polyketides were not biosynthesized by culturing of this mutant strain, and when p-aminobenzoic acid was added, the biosynthetic capability of the mutant strain was restored. From this result, it could be found that pabAB is a gene coding for enzymes for the synthesis of p-aminobenzoic acid.
When comparing KR domains related with double bonds, it can be seen that, in KR8 and KR9, aspartate (D) residues that are conserved amino acids known as producing D configuration (Tang et al, Chem. Biol., 6:553-8, 1999), are changed into glycine (G) and asparagine (N) by reduction, respectively. KR having conserved aspartate forms trans double bonds by dehydratase activity after reduction. The presence of two double bonds in KR was found by the NMR analysis of vacidin A (Altschul et al, Trends Biochem. Sci., 23:444-7, 1998), and it was reported that a KR domain was connected with the planar structure of double bonds (Tang et al, Chem. Biol., 6:553-8, 1999). As a result, it is predicted that two double bonds of C28-C29 and C30-C31 of FR-008 is in cis as in other similar KRs (Pearson, Methods Enzymol., 183:63-98, 1990; August et al, Chem. Biol., 5:69-79, 1998).
As described and demonstrated in detail as described above, the present invention provides the base sequence of genes involved in the biosynthesis of new FR-008 polyketides. The whole or parts of the gene base sequence provided in the present invention can be used to develop recombinant microorganisms capable of producing FR-008 polyketides or to search for the active site of FR-008 polyketides from Streptomyces sp. FR-008 or other species. Moreover, the whole or parts of the gene base sequence provided in the present invention can be applied to increase the productivity of the existing FR-008 polyketides or to produce new FR-008 polyketide variants, by its modification.
In particular, the whole or parts of the gene base sequence provided in the present invention can be used to produce FR-008 polyketides or variants thereof in cells other than source producer strain Streptomyces sp. FR-008. In addition, it can be used in molecular evolution by DNA shuffling between the DNA fragments of the whole or parts of the gene base sequence according to the present invention and other genes having similar homology, and allows new polyketides to be produced using such a procedure.
While the present invention has been described in detail with reference to the specific features, it will be apparent to those skilled in the art that this description is only for a preferred embodiment and does not limit the scope of the present invention. Thus, the substantial scope of the present invention will be defined by the appended claims and equivalents thereof.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2003-0074035 | Oct 2003 | KR | national |
