Method for producing retinoid from microorganism

Abstract

The present invention relates to a method for producing retinoid from a microorganism, and more specifically, to a method for effectively obtaining retinoid, which lacks stability, from a microorganism by cultivating the microorganism capable of producing retinoid in a medium containing a lipophilic substance, and separating retinoid from the lipophilic substance.

Claims

1. A method for production of retinoid from a microorganism, comprising: culturing a microorganism having retinoid producing efficacy through 2-C-methyl-d-erythritol 4-phosphate (MEP) pathway or mevalonate (MVA) pathway in a medium containing a lipophilic substance; and isolating said retinoid from said lipophilic substance without degrading a cell wall of said microorganism.

2. The method of claim 1, wherein the microorganism is a bacteria, a fungi, a isolated animal cell, or a combination thereof.

3. The method of claim 1, wherein said microorganism is of the genus Escherichia, the genus Bacillus, the genus Corynebacterium, yeast, the genus Kluyveromyces or a combination thereof.

4. The method of claim 1, wherein said lipophilic substance is an alkane compound having 8 to 50 carbon atoms, a compound represented by Formula 1 below, a compound represented by Formula 2 below, or a combination thereof:
R.sub.1(CO)OR.sub.2[Formula 1] wherein R.sub.1 and R.sub.2 are each independently alkyl having 8 to 50 carbon atoms, and CO represents a carbonyl group; and ##STR00002## wherein R.sub.3, R.sub.4 and R.sub.5 are each independently alkyl having 8 to 50 carbon atoms, and CO represents a carbonyl group.

5. The method of claim 1, wherein said lipophilic substance is octane, decane, dodecane, tetradecane, phytosqualane, mineral oil, isopropyl myristate, cetyl ethylhexanoate, dioctanoyl decanoyl glycerol, squalane, or a combination thereof.

6. The method of claim 1, wherein a ratio by volume of said medium to said lipophilic substance ranges from 1:0.2 to 3.0.

7. The method of claim 1, wherein said culturing is performed while agitating.

8. The method of claim 1, wherein said medium further comprises glycerol.

9. The method of claim 1, wherein said medium further comprises glucose.

10. The method of claim 1, wherein said isolating includes removing cells from a culture solution and then isolating said retinoid from dodecane.

11. The method of claim 1, wherein said retinoid is selected from the group consisting of retinal, retinol, retinyl ester and retinoic acid.

12. The method of claim 1, wherein said microorganism is Escherichia coli.

13. The method of claim 12, wherein said Escherichia coli is DH5, MG1655, BL21 (DE), S17-1, XL1-Blue, BW25113 or a combination thereof.

14. The method of claim 1, wherein said microorganism is further transformed with a gene selected from the group consisting of: a gene encoding 3-carotene monooxygenase comprising the amino acid sequence of SEQ ID NO: 13 from uncultured marine bacterium 66A03; a gene encoding -carotene 15,15-monooxygenase comprising the amino acid sequence of SEQ ID NO: 14 from Mus musculus; a gene encoding brp-like protein 2 (brp 2) comprising the amino acid sequence of SEQ ID NO: 15 from Natronomonas pharaonis ATCC35678; and a gene encoding 3-carotene monooxygenase comprising the amino acid sequence of SEQ ID NO: 16 or 17 from Halobacterium salinarum ATCC700922.

15. The method of claim 1, wherein said microorganism is further transformed by a gene comprising the nucleotide sequence of SEQ ID NO: 32, wherein said gene is codon optimized for Escherichia coli.

16. The method of claim 1, wherein said microorganism is further transformed by a gene encoding 1-deoxyxylolose-5-phosphate (DXP) synthase comprising the amino acid sequence of SEQ ID NO: 11 from Escherichia coli.

17. The method of claim 1, wherein said microorganism is further transformed by a gene encoding IPP isomerase comprising the amino acid sequence of SEQ ID NO: 12 from Haematococcus pluvialis.

18. The method of claim 1, wherein said microorganism is Escherichia coli DH5/pTDHB/pSNA deposited under Accession No. KCTC 11254BP or Escherichia coli DH5/pTDHBSR/pSNA deposited under Accession No. KCTC 11255BP.

19. The method of claim 1, wherein said microorganism is of the genus Escherichia.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a view schematically illustrating an MEP path and a foreign MVA path of retinal biosynthesis;

(2) FIG. 2 illustrates retinoid conversion of -carotene into retinoid including retinal, retinol, retinoic acid and retinyl ester.

(3) FIG. 3 illustrates production of retinal, production of -carotene and cell growth of Escherichia coli including pT-HB, pT-HBblh, pT-HPbrp, pT-HBbrp2, pT-HBBCMO1 and pT-HBSR;

(4) FIG. 4 illustrates production of retinal, production of -carotene and cell growth of Escherichia coli including pT-HB, pT-HBSR, pT-DHB and pT-DHBSR, as well as Escherichia coli including pT-DHB or pT-DHBSR together with pS-NA as an MVA path plasmid;

(5) FIG. 5 illustrates retinoid production and cell growth by a variety of Escherichia coli strains including pT-DHBSR and pS-NA;

(6) FIG. 6 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on a test volume of a culture solution;

(7) FIG. 7 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on a culture temperature;

(8) FIG. 8 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on a carbon source;

(9) FIGS. 9 and 10 illustrate retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on a concentration of glycerol as a carbon source, respectively; and

(10) FIGS. 11 and 12 illustrate results of retinoid production and cell growth by a variety of Escherichia coli strains in the presence of dodecane, respectively.

(11) FIG. 13 illustrates retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) depending on a concentration of glycerol as a carbon source in a 2-phase culture system including 1 mL of dodecane in 5 ml of culture medium;

(12) FIG. 14 illustrates retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) in a 2-phase culture system depending on the volume of dodecane;

(13) FIG. 15 illustrates distributions of retinoid depending on a culturing time of Escherichia coli (pT-DHBSR/pS-NA) and a volume of dodecane in a 2-phase culture system, which are represented in terms of percentages of individual constituents to total retinoid;

(14) FIG. 16 illustrates effects of dodecane addition upon production of beta-carotene and cell growth of Escherichia coli including pT-DHB and pS-NA;

(15) FIGS. 17 and 18 illustrate results of retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) in the presence of different alkanes, respectively;

(16) FIGS. 19, 20 and 21 illustrate results of retinoid production, cell growth, and cell specific retinoids productivity of Escherichia coli (pT-DHBSR/pS-NA) in the presence of different volumes of lightweight mineral oil, respectively;

(17) FIGS. 22 and 23 illustrate results of retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) in the presence of heavy mineral oil, respectively;

(18) FIGS. 24 and 25 illustrate results of retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) when culturing was conducted by tilting a test tube, respectively;

(19) FIG. 26 illustrates cell growth and pH of Escherichia coli (pT-DHBSR/pS-NA) in the presence of skin-friendly lipophilic substance; and

(20) FIGS. 27 and 28 illustrate results of retinoid production of Escherichia coli depending on different kinds and amounts of skin-friendly lipophilic substance, respectively.

DETAILED DESCRIPTION

(21) Hereinafter, the present invention will be described in more details according to the following examples. However, these examples are proposed for illustrative purposes only and the scope of the present invention is not particularly limited thereto. In the examples, the following experimental materials and methods have been used.

(22) Bacteria Strain and Culture Conditions

(23) Escherichia coli DH5 was used for gene cloning and retinoid production. Alternatively, Escherichia coli MG1655, BL21 (DE3), XL1-Blue, S17-1 and BW25113 were used to investigate an optimum strain for retinol production. The culture for retinoid production was executed in a 2YT medium (including 16 g of tryptone, 10 g of yeast extract and 5 g of NaCl per liter) at 29 C. using an agitation incubator operating at 250 rpm. A major and additional carbon sources were glycerol and arabinose, which were added in concentrations of 0.5 to 2% (w/v) and 0.2% (w/v), respectively, to the incubator. Alternative carbon sources for retinoid production, for example, glucose, galactose, xylose and maltose were compared to glycerol. Ampicillin (100 g/mL) and chloramphenicol (50 g/mL) were optionally added to a culture solution requiring the same. The culturing was conducted in a test tube containing 7 ml of medium. Cell growth was determined by measuring an optical density at 600 nm (OD.sub.600). In a 2-phase culture method for production of retinoid, 1 mL of dodecane (Cat. No. 297879, Sigma, USA) was placed on 5 ml of the culture medium.

(24) Conditions for Analysis of -Carotene and Retinoid

(25) -carotene and retinoid were extracted from bacteria cell pellets through acetone. In the 2-phase culture method including dodecane capping, cell pieces were completely removed by collecting a dodecane phase containing retinoid and centrifuging the same at 14,000 rpm for 10 minutes. The acetone extraction product and dodecane phase were analyzed using HPLC (LC-20A, Shimadzu, Kyoto, Japan) at detection wavelengths of 370 nm (retinal), 340 nm (retinol and retinyl acetate) and 454 nm (-carotene). The analysis was performed using a Symmetry C18 type (250 mm4.6 mm, 5 m) HPLC column including Sentry Guard C18 (15 mm4.6 mm, 5 m). A mobile phase of the column was each of methanol and acetonitrile in ratios of 95:5 and 70:30, respectively, for analyzing retinoid and -carotene. HPLC analysis was performed at a flow rate of 1.5 ml/min and a column temperature of 40 C. Retinal (Cat. No. R2500), retinol (Cat. No. R7632), retinyl acetate (Cat. No. R4632) and -carotene (Cat. No. C4582) were purchased from Sigma Co. (USA) and dissolved in acetone, respectively, to prepare standard compounds, and each of the prepared standard compounds was used. Through three independent experiments, results were obtained and represented by a meanSD.

Example 1: Preparation of Vector for Producing Escherichia coli with High Productivity of a -Carotene and Retinal

(26) Conventional processes involving genome DNA preparation, restriction enzyme cleavage, transformation and standard molecular biological technologies have been executed according to description in related documents (Sambrook and Russell 2001). PCR was performed using pfu DNA polymerase according to standard protocols (Solgent Co., Korea). blh gene of uncultured marine bacteria 66A03 (Genbank accession No. AAY68319) was synthesized into Genofocus (Daejeon, Korea) according to codon-optimization by Gene Designer software (DNA 2.0, Menlo Park, USA), in order to express the above gene in Escherichia coli.

(27) According to the present example, an enzyme involved in a velocity determination process, that is., a gene encoding DXP synthase was additionally introduced into Escherichia coli having an MEP path and, at the same time, a gene encoding an enzyme associated with a mevalonate path was selected from a variety of gene resources and introduced, thus preparing Escherichia coli with hugh productivity of -carotene.

(28) (1) Preparation of pSNA Vector Including a Gene Encoding an Enzyme in a Mevalonate Path Associated with Synthesis of IPP from a Carbon Source

(29) Genes encoding an enzyme in a mevalonate path associated with IPP synthesis from a carbon source used in the present experiment are shown in Table 1 below.

(30) TABLE-US-00001 TABLE 1 Gene sequence (Genbank Name of enzyme Gene accession No.) Acetyl-CoA mvaE SEQ. ID No. 18 acetyltransferase/hydroxymethylglytaryl (AF290092) (HMG)-CoA reductase derived from Enterococcus faecalis HMG-CoA synthase derived from mvaS SEQ. ID No. 19 Enterococcus faecalis (AF290092) Mevalonate kinase derived from mvaK1 SEQ. ID No. 20 Streptococcus pneumonia (AF290099) Phosphomevalonate kinase derived from mvaK2 SEQ. ID No. 21 Streptococcus pneumonia (AF290099) Mevalonate diphosphate decarboxylase mvaD SEQ. ID No. 22 derived from Streptococcus pneumoniae (AF290099) Isopentenyl diphosphate (IPP) isomerase Idi SEQ. ID No. 23 derived from Escherichia coli (U00096)

(31) Primers and restriction enzymes to amplify genes listed in Table 1 have been described.

(32) TABLE-US-00002 TABLE 2 Restriction Primer sequence enzyme mvaE F SEQ. ID No. 37 SacI R SEQ. ID No. 38 SmaI mvaS F SEQ. ID No. 39 SmaI R SEQ. ID No. 40 BamHI mvaK1, mvaK2, F SEQ. ID No. 41 KpnI mvaD R SEQ. ID No. 42 XbaI Idi F SEQ. ID No. 43 SmaI R SEQ. ID No. 44 SphI

(33) The primer sequences and restriction enzymes used in cloning the genes listed in Table 1 are stated in Table 2. Since mvaK1, mvaK2 and mvaD are present as a single operon in a chromosome, a whole operon rather than individual genes was subjected to PCR cloning at once.

(34) The genes listed in Table 1 were amplified using the primers listed in Table 3 through PCR which uses a chromosome DNA in each strain including corresponding gene as a matrix. The amplified product was introduced into pSTV28 vector (Takara Korea, Korea) (SEQ. ID No. 45) using the restriction enzymes listed in Table 2, thereby preparing the vector pSNA. The vector pSNA includes all of genes encoding the enzyme in a mevalonate path, which can produce IPP from acetyl-CoA.

(35) (2) Preparation of Vectors pT-HB and pT-DHB Including a Gene Encoding an Enzyme Associated with Synthesis of -Carotene from IPP

(36) Genes encoding an enzyme associated with synthesis of -carotene from IPP used in the present experiment, as well as DXP synthase gene as an enzyme involved in the velocity determination process in the MEP path, are shown in Table 3 below.

(37) TABLE-US-00003 TABLE 3 Gene sequence (Genbank Name of enzyme Gene accession No.) IPP isophomerase derived from ipiHpl SEQ. ID No. 24 Haematococcus pluvialis (AF082325) 1-deoxyxylolose-5-phosphate (DXP) dxs SEQ. ID No. 25 synthase derived from Escherichia coli (U00096) Geranylgeranyl pyrophosphate (GGPP) crtE SEQ. ID No. 26 synthase derived from pantoea agglomerans (M87280) Phytoene synthase derived from pantoea crtB SEQ. ID No. 27 agglomerans (M87280) Phytoene dehydrogenase derived from crtI SEQ. ID No. 28 pantoea agglomerans (M87280) Lycopene -cyclase derived from pantoea crtY SEQ. ID No. 29 ananatis (D90087)

(38) TABLE-US-00004 TABLE 4 Restriction Gene Primer sequence enzyme ipiHpl F SEQ. ID No. 46 SmaISphI R SEQ. ID No. 47 dxs F SEQ. ID No. 48 EcoR1SnaBI R SEQ. ID No. 49 crtE F SEQ. ID No. 50 BspHIEcoRI R SEQ. ID No. 51 crtB, crtI F SEQ. ID No. 52 EcoR1SacI R SEQ. ID No. 53 crtY F SEQ. ID No. 54 SalIPstI R SEQ. ID No. 55

(39) The primer sequences and restriction enzymes used in cloning the genes listed in Table 3 are stated in Table 4. Since crtB and crtl are present as a single operon in a chromosome, a whole operon rather than individual genes was subjected PCR cloning at once.

(40) The genes listed in Table 3 were amplified using the primers listed in Table 4 through PCR which uses a chromosome DNA in each strain including corresponding gene as a matrix. The amplified product was introduced into pTrc99A vector (Genbank accession No. M22744) (SEQ. ID No. 30) using the restriction enzymes listed in Table 4, thereby preparing the vector pT-DHB. The vector pTDHB includes all of genes encoding the enzyme associated with synthesis of -carotene from IPP, as well as DXP synthase (dxs) gene as an enzyme used in the velocity determination process in the MEP path. Further, among the genes listed in Table 3, all genes other than dxs were introduced into pTrc99A vector using the restriction enzymes listed in Table 4, thereby preparing the vector pT-HB.

(41) (3) Preparation of a Vector Including a Gene Encoding an Enzyme Associated with Synthesis of Retinal from -Carotene

(42) Genes encoding an enzyme associated with synthesis of retinal from -carotene used in the present experiment are shown in Table 5 below. As a gene encoding -carotene monooxygenase derived from uncultured marine bacterium 66A03, SR gene which is an Escherichia coli codon-optimized sequence of blh was used.

(43) TABLE-US-00005 TABLE 5 Gene sequence (Genbank Name of enzyme Gene accession No.) -carotene monooxygenase derived from blh SEQ. ID No. 31 uncultured marine bacterium 66A03 (DQ065755) -carotene monooxygenase derived from SR SEQ. ID No. 32 uncultured marine bacterium 66A03 (Escheri- chia coli codon- optimized sequence of blh) -carotene 15,15-monooxygenase BcmoI SEQ. ID No. 33 derived from Mus musculus (NM_021486) Brp-like protein 2 derived from brp2 SEQ. ID No. 34 Natronomonas pharaonis ATCC35678 (CR936257) -carotene monooxygenase derived from Blh SEQ. ID No. 35 Halobacterium salinarum ATCC700922 (AE004437) -carotene monooxygenase derived from Brp SEQ. ID No. 36 Halobacterium salinarum ATCC700922 (AE004437)

(44) TABLE-US-00006 TABLE 6 Restriction Gene Primer sequence enzyme SR F SEQ. ID No. 56 EcoR1SpeI R SEQ. ID No. 57 bcmo1 F SEQ. ID No. 58 EcoR1SpeI R SEQ. ID No. 59 brp2 F SEQ. ID No. 60 EcoR1SpeI R SEQ. ID No. 61 blh F SEQ. ID No. 62 EcoR1SpeI R SEQ. ID No. 63 brp F SEQ. ID No. 64 EcoR1SpeI R SEQ. ID No. 65

(45) The primer sequences and restriction enzymes used in cloning the genes listed in Table 5 are stated in Table 6. The genes listed in Table 5 were amplified using the primers listed in Table 6 through PCR which uses a chromosome DNA in each strain including a corresponding gene as a matrix. The amplified product was introduced into pT-HB vector using the restriction enzymes listed in Table 6, respectively, thereby preparing the vectors pT-HBSR, pT-HBBcmol, pT-HBbrp2, pT-HBblh and pT-HBbrp. Such vectors pT-HBSR, pT-HBBcmol, pT-HBbrp2, pT-HBblh and pT-HBbrp are vectors formed by introducing SR, Bcmol, brp2, blh and brp genes into pT-HB vector, respectively, and have included all of genes encoding an enzyme associated with the synthesis of retinal through -carotene from IPP. After cutting SR gene from pT-HBSR using Spel, the cut gene was introduced into a corresponding part of pT-DHB, thus preparing pT-DHBSR.

Example 2: Comparison of Different BCM(D)O Genes in Relation to Retinal Production

(46) Retinal may be produced by introduction of BCM(D)O gene encoding -carotene mono(di)oxygenase, which is a recombinant Escherichia coli producing -carotene. The present inventors have conducted cloning of BCM(D)O gene from two bacteria, i.e., Halobacterium sp NRC-1 (blh and brp genes) and Natronomonas pharaonis (brp2 gene), as well as Mus musculus (Bcmol gene) of a vertebrate animal. The present inventors have synthesized codon-optimized BCDO gene (SR) on the basis of an amino acid sequence of uncultured marine bacterium 66A03 blh gene. BCM(D)O gene (SR) was used to prepare retinal synthetic plasmids pT-HBblh, pT-HBbrp, pT-HBbrp2, pT-HBBcmol and pT-HBSR, respectively. The recombinant Escherichia coli cell containing each of retinal plasmids was cultured in a 2YT medium including 0.5% (w/v) of glycerol and 0.2% (w/v) of arabinose as a carbon source at 29 C. for 48 hours.

(47) FIG. 3 illustrates production of retinal, production of -carotene and cell growth of Escherichia coli including pT-HB, pT-HBblh, pT-HBbrp, pT-HBbrp2, pT-HBBCMO1 and pT-HBSR. More particularly, white and grey bars show numerical values at 24 hours and 48 hours, respectively.

(48) As shown in FIG. 3, the recombinant Escherichia coli pT-HBblh, pT-HBbrp and pT-HBSR have produced 2.2, 0.8 or 1.4 mg/L of retinal, respectively, at 24 hours. However, retinal production by the recombinant Escherichia coli pT-HBblh or pT-HBbrp was reduced to 0.7 or 0.4 mg/L, respectively, at 48 hours, whereas Escherichia coli pT-HBSR showed a slight increase in retinal production. The decrease in retinal production after 24 hours may be caused by oxidative degradation of retinal in the cell. An amount of retinal obtained from the culture solution depends upon both intracellular synthesis and degradation of retinal.

(49) For Escherichia coli including pT-HBblh or pT-HBbrp, a retinal productivity at 24 hours after culturing may be lower than a rate of degradation of the same. In the culture of Escherichia coli strain including pT-HBbrp2 or pT-HBBcmol, a trace amount of retinal was detected. Escherichia coli without BCM(D)O gene has produced 35 mg/L of -carotene, but did not produce retinal. Since -carotene is a precursor just before retinal, a -carotene consumption by BCM(D)O may be exactly proportional to the retinal productivity if there was retinal degradation. -carotene remained in a culture solution of Escherichia coli including BCM(D)O other than SR, therefore, -carotene cleavage activity of SR was expected to be the highest level among tested BCM(D)O. Accordingly, in an additional experiment, SR enzyme was adopted for retinal production. The cell growth did not come under the influence of over-expression of BCM(D)O gene except for N. pharaonis brp gene exhibiting delayed cell growth.

Example 3: Gene Manipulation into MEP and MVA Paths for Supplying a Building Block

(50) Retinal building blocks, that is, IPP and DMAPP may be synthesized in Escherichia coli through an inherent MEP path and a foreign MVA path (FIG. 1).

(51) It was reported that synthesis of 1-deoxy-d-xylolose-5-phosphate (DXP) is an important velocity restriction process in the MEP path. Therefore, over-expression of DXP synthase (to be encoded by dxs) increased production of lycopene and -carotene in previous inventions of the present inventors. By introducing dxs gene into before the MEP path among pT-HBSR, pT-DHBSR was prepared.

(52) FIG. 4 illustrates production of retinal and -carotene, and cell growth of Escherichia coli including pT-HB, pT-HBSR, pT-DHB and pT-DHBSR, as well as Escherichia coli including pT-DHB or pT-DHBSR together with pS-NA as an MVA path plasmid. More particularly, white and grey bars show numerical values at 24 hours and 48 hours, respectively.

(53) As shown in FIG. 4, the retinal productivity of Escherichia coli pT-DHBSR was a little higher than that of Escherichia coli pT-DHB at 24 hours, while being substantially similar to the same at 48 hours. However, -carotene production by Escherichia coli pT-DHB was increased by about 1.5 times due to over-expression of dxs, as compared to Escherichia coli pT-HB. It is known that a foreign MVA path in Escherichia coli considerably increases production of isoprenoid by providing sufficient amounts of IPP and DMAPP building blocks. Escherichia coli pT-DHBSR/pS-NA including an additional foreign MVA path have produced 8.7 mg/L of retinal for 48 hours, which is 4 times higher than the productivity of Escherichia coli pT-DHBSR. For an Escherichia coli strain including SR gene, -carotene did not remain or slightly remained in the cell. This condition is presumed due to an effective cleavage reaction of -carotene by SR. There was a considerable difference between an amount of -carotene (a substrate) consumption and an amount of produced retinal (a product). This difference may be due to the presence of a cellular reaction to metabolize retinal in Escherichia coli as well as biological degradation of retinal. Accordingly, formation of an alternative retinoid derived from retinal by any enzyme in Escherichia coli may be under consideration. Since the retinal can be converted into retinol, retinoic acid and retinyl ester by cell-enzyme reaction (FIG. 2), retinal derivatives contained in an Escherichia coli culture solution were subjected to analysis. The analysis results showed that retinal derivatives other than retinoic acid were formed. According to further experiments, production of retinal, retinol and retinyl acetate was determined.

Example 4: Effects of Escherichia Coli Strain, Culture Conditions and Carbon Source in Relation to Retinoid Production

(54) (1) Strain

(55) With regard to production of retinoid including retinal, retinol and retinyl acetate, effects of Escherichia coli strains were investigated. Five Escherichia coli strains including pT-DHBSR and pS-NA, that is, MG1655, DH5, XL1-Blue, S17-1 and BL21 (DE3) were used to produce retinoid. Table 7 shows characteristics of six Escherichia coli strains including the foregoing five strains.

(56) TABLE-US-00007 TABLE 7 E. coli strain Details MG1655 K12, wild type DH5 F.sup., f80dlacZM15, (lacZYA-argF)U169, deoR, recA1endA1, hsdR17(r.sub.K.sup.m.sub.K.sup.+), phoA, supE44, -, thi-1, gyrA96, relA1 XL1-Blue hsdR17, supE44, recA1, endA1, gyrA46, thi, relA1, lac/F'[proAB.sup.+, lacI.sup.q, lacZM15::Tn10(tet.sup.r)] S17-1 recA pro hsdR RP4-2-Tc::Mu-Km::Tn7 BL21 (DE3) F.sup., ompT, hsdS.sub.B(r.sub.B.sup.m.sub.B.sup.), gal(lcI857, ind1, Sam7, nin5, lacUV5-T7gene1), dcm(DE3) BW25113 (araD-araB)567, lacZ4787(::rrnB-3), lambda.sup., rph-1, (rhaD-rhaB)568, hsdR514

(57) FIG. 5 illustrates retinoid production and cell growth of five Escherichia coli strains, respectively, of which each has pT-DHBSR and pS-NA. Culturing was conducted in a 2YT medium including 0.5% (w/v) of glycerol and 0.2% (w/v) of arabinose at 29 C. for 48 hours. Retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. Also, in case of cell growth, MG1655, DH5, XL1-Blue, S17-1 and BL21 (DE3) are represented by .square-solid., .circle-solid., .box-tangle-solidup., .Math. and .diamond-solid., respectively.

(58) As shown in FIG. 5, Escherichia coli DH5 has produced 40 mg/L, that is, the largest amount of retinoid at 36 hours, and Escherichia coli S17-1 and XL1-Blue have produced about 22 mg/L, that is, the second largest amount of retinoid. However, from Escherichia coli MG1655 and BL21 (DE3), only a trace amount of retinoid was produced. Therefore, Escherichia coli DH5 was adopted as a strain for retinoid production.

(59) (2) Culture Conditions

(60) With regard to production of retinoid, effects of dissolved oxygen were investigated with difference test volumes in a test tube having a diameter of 30 mm.

(61) FIG. 6 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on test volume. Referring to FIG. 6, retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. Also, in case of cell growth, test volumes of 3 mL, 5 mL, 7 mL and 10 mL are represented by .square-solid., .circle-solid., .box-tangle-solidup. and .Math., respectively. Culturing was conducted in a 2YT medium including 0.5% (w/v) of glycerol and 0.2% (w/v) of arabinose at 29 C. for 48 hours.

(62) As shown in FIG. 6, it was found that retinoid production more early reached the maximum level at a small test volume (corresponding to highly dissolved oxygen), and the production was deemed to more quickly decrease due to oxidative degradation. With 10 mL of test volume, both of cell growth and retinoid production were delayed, while degradation of the product was observed by a small extent. It was found that the optimum test volume for retinoid production is 7 mL.

(63) Further, the retinoid production depending upon the temperature was investigated. FIG. 7 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on a culture temperature. Referring to FIG. 7, retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. Also, in case of cell growth, culture temperatures of 29 C., 34 C. and 37 C. are represented by .square-solid., .circle-solid. and .box-tangle-solidup., respectively. Culturing was conducted in a 2YT medium including 0.5% (w/v) of glycerol and 0.2% (w/v) of arabinose at 29 C., 34 C. and 37 C., respectively, for 48 hours.

(64) As shown in FIG. 7, the retinoid production has come under an influence of culture temperature and the highest production was accomplished at 29 C.

(65) (3) Carbon Source

(66) Effects of different carbon sources upon production of retinoid were compared.

(67) FIG. 8 illustrates retinoid production and cell growth of Escherichia coli including pT-DHBSR and pS-NA depending on the carbon source. Referring to FIG. 8, retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. Also, in case of cell growth, no carbon source, and the carbon sources of glycerol, glucose, xylose, maltose and galactose are represented by .square-solid., .circle-solid., .box-tangle-solidup., .Math., .diamond-solid. and , respectively. Culturing was conducted in a 2YT medium including 0.2% (w/v) of arabinose and 0.5% (w/v) of glycerol, glucose, xylose, maltose or galactose, at 29 C. for 48 hours.

(68) As shown in FIG. 8, it was found that glycerol was the best carbon source for retinoid production. When glucose or galactose was used as the carbon source, the retinoid productivity was lower than that in case where the carbon source was not used.

(69) Next, effects of a concentration of glycerol upon the retinoid production and cell growth were investigated. Escherichia coli DH5(pT-DHBSR/pSNA) was grown in a 2YT medium including glycerol in a range of 0.0% to 2.0% (w/v), at 29 C.

(70) FIG. 9 illustrates production of retinoids (retinal, retinol and retinyl acetate) of Escherichia coli including pT-DHBSR and pS-NA. Retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively.

(71) FIG. 10 illustrates cell growth of Escherichia coli including pT-DHBSR and pS-NA. Given glycerol concentrations of 0%, 0.5%, 1% and 2% are represented by .square-solid., .circle-solid., .box-tangle-solidup. and .Math., respectively.

(72) As shown in FIGS. 9 and 10, the cell growth was proportional to the glycerol concentration and increased. With the glycerol concentration of 0.5, 1.0 and 2.0% (w/v), the cell growth has become stagnate at 36, 48 and 72 hours, respectively. At the above time, the maximum retinoid productivity was accomplished and, thereafter, the productivity was considerably reduced during stagnation. It can be seen that the retinoid production generally increases after 24 hours. The retinoid productivity was about 95 mg/L, the highest level, at 2.0% (w/v) of glycerol among various glycerol concentrations, which is substantially 2.4 times higher than the maximum retinoid productivity at 0.5% (w/v) of glycerol. An increase in glycerol concentration delayed the stagnation while extending a period of retinoid production.

(73) From all culture solutions, it was observed that the retinoid production was extremely reduced during stagnation of the cell growth, and this condition is deemed to be caused by discontinued production of retinoid during stagnation and intracellular oxidative degradation of the same.

(74) (4) Culture in the Presence of Dodecane

(75) Strains containing transformed pT-DHBSR/pSNA were used for six strains listed in Table 7. After adding 1 mL of dodecane to 5 ml of medium, culturing was conducted according to such conditions as described in bacteria strain and culture conditions. The medium used herein was a 2YT medium including 0.2% (w/v) of arabinose and 0.5% (w/v) of glycerol added thereto.

(76) FIG. 11 illustrates results of the retinoid production depending on different Escherichia coli strains for retinoid production. As shown in FIG. 11, DH5 and MG1655 showed the largest amount of retinoid production. For MG1655, the cell growth and retinoid productivity were increased, as compared to no addition of dodecane. A cell growth rate and a rate of increasing retinoid productivity for MG1655 were obviously faster than DH5. BL21 (DE3) strain showed a still high cell growth but had scarcely any production of retinyl acetate. Consequently, it was determined that DH5 and MG1655 among six strains are relatively suitable, as compared to other strains.

(77) FIG. 12 illustrates results of growth of strains for retinoid production in the presence of dodecane.

(78) FIG. 13 illustrates results of retinoid production and growth in the presence of dodecane, depending on the concentration of glycerol as a carbon source.

Example 5: 2-Phase Culture Using Dodecane for In-Situ Extraction of Retinoid

(79) In order to prevent intracellular degradation of retinoid, a 2-phase culture method was conducted using a hydrophobic solvent, that is, dodecane, in order to perform in-situ extraction of retinoid from cells. Dodecane was selected since it has a low toxicity to Escherichia coli I, a high hydrophobicity (log P.sub.o/w, 6.6) for extracting hydrophobic retinoid, and low volatile properties not to cause evaporation loss.

(80) In the present example, 1 mL of dodecane was added to 5 ml of culture solution. FIG. 13 illustrates retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) in a 2-phase culture system including 1 mL of dodecane in 5 ml of culture medium. With regard to retinoid production, retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. In the case of cell growth, given glycerol concentrations of 0.5%, 1% and 2% are represented by .square-solid., .circle-solid. and .box-tangle-solidup., respectively.

(81) Retinoid was extracted into the dodecane phase while an insignificant amount of retinoid was detected in the culture solution and cell mass (data not shown). As a result, the retinoid productivity was measured from the dodecane phase. As shown in FIG. 13, in-situ extraction could minimize intracellular degradation of retinoid by dodecane. The retinoid among the dodecane phase was deemed to be relatively stable and remained without significant oxidative degradation thereof. As comparing with the results shown in FIGS. 9 and 10 (without addition of dodecane), the retinoid production was remarkably increased even at 24 hours in case of adding the dodecane. Further, the cell growth did not come under an influence of the dodecane addition while the retinoid production was not decreased during stagnation. However, in the culture using 2% (w/v) glycerol, the retinoid production was not so higher than that in case of using 1% (w/v) glycerol, even though the cell growth was remarkably increased in proportion to an increase in glycerol concentration from 1% (w/v) to 2% (w/v). When the volume of dodecane addition is 1 mL, it is insufficient to conduct effective in-situ extraction of retinoid in the culture using 2% (w/v) of glycerol.

(82) In order to investigate effects of the volume of dodecane addition on the retinoid production and cell growth, 1 mL to 5 mL of dodecane was initially added to a culture solution including 2% (w/v) of glycerol (FIG. 14).

(83) FIG. 14 illustrates retinoid production and cell growth of Escherichia coli (pT-DHBSR/pS-NA) in a 2-phase culture system, depending on the volume of dodecane. With regard to the retinoid production, retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively. In case of cell growth, volumes of overlaying dodecane of 0 mL, 1 mL, 2 mL, 3 mL, 4 mL, 5 mL and 6 mL are represented by .square-solid., .circle-solid., .box-tangle-solidup., , , and .star-solid., respectively.

(84) FIG. 15 illustrates a distribution of retinoid depending on culturing time and volume of dodecane, in terms of percentages of individual constituents to total retinoid. Retinal, retinol and retinyl acetate are represented by bright grey, dark grey and black colors, respectively.

(85) As shown in FIGS. 14 and 15, production of overall retinoids was improved according to an increase in the volume of dodecane addition. In 72 hours culture using 5 mL of dodecane, the highest retinoid productivity of 136 mg/L was obtained, which is about 2 times higher value than that (65 mg/L) in case where 1 mL of dodecane is used. Meanwhile, in 72 hours-extended culture using 5 mL of dodecane, the retinoid productivity did not further increase but the highest level was maintained without degradation of the retinoid (data not shown). By adding 2 mL of dodecane to the culture solution at 0, 24 and 48 hours, a whole volume of dodecane addition was increased to 6 mL. In the culture using 6 mL of added dodecane, the total retinoid productivity did not increase, as compared to the culture using 5 mL of dodecane. Likewise, even in the culture using 6 mL of initially added dodecane, the retinoid productivity did not increase (data not shown). The cell growth in all of culture solutions including dodecane was slightly higher than that in case of not using dodecane (FIG. 14).

(86) FIG. 15 illustrates a distribution of produced retinoids depending on the volume of dodecane addition. With regard to ratios of obtained retinal and retinol, there is a considerable difference in retinoid distributions between addition of dodecane and no addition of dodecane. A ratio of retinal in retinoid at 48 hours was about 51% (w/v) in the dodecane-added culture and 23% (w/v) in the culture without adding dodecane. Likewise, a ratio of retinol ranged from 30 to 39% in the dodecane-added culture and was 59% in the culture without adding dodecane. Accordingly, the addition of dodecane may increase a ratio of retinal while reducing a ratio of retinol. In consideration of the order of reactions for formation of retinol from retinal in a cell, retinal is deemed to be extracted from the cell before conversion of the same into retinol by dodecane. Further, a ratio of retinyl acetate at 48 hours was less than 20% in both of the cultures with and without addition of dodecane, which is relatively lower than the ratios of retinal and retinol. In the culture with addition of dodecane, the ratio of retinyl acetate is reduced as the culturing time is extended and this indicates that activity of cells for forming retinyl acetate is reduced during culturing. Consequently, adding dodecane has prevented a decrease of retinoid production during stagnation of the cell growth, while improving the retinoid production.

(87) The in-situ extraction of retinoid according to the present invention does not need lysozyme used for degrading a cell wall. Retinoid (C20, isoprenoid molecule) may be efficiently released from the cell without loss of the cell wall. In 2-phase culture for production of retinoid, -carotene must be continuously maintained in the cell since it is a direct precursor of the retinoid. If -carotene is extracted from the dodecane phase, it can be cut by BCD(M)O placed in cytosol.

(88) Due to a size of molecule, -carotene can neither be released from the cell nor extracted by dodecane, therefore, can be continuously maintained in the cell during 2-phase culture of -carotene (FIG. 16).

(89) FIG. 16 illustrates effects of dodecane addition depending on -carotene production and cell growth of Escherichia coli including pT-DHB and pS-NA. Culturing was conducted in 5 ml of 2YT medium including 0.5% (w/v) of glycerol and 0.2% (w/v) of arabinose at 29 C. for 48 hours while adding 1 mL of dodecane to the medium. Grey and black bars indicated 24 hours and 48 hours, respectively.

(90) As shown in FIG. 16, an insignificant amount of -carotene was detected in the dodecane phase and a whole -carotene has been almost retained in the cell. There was not a noticeable difference in -carotene production and cell growth between cultures with and without addition of dodecane.

(91) In the culture with addition of 5 mL of dodecane, a total 122 mg/L of retinoid productivity was attained at 48 hours. However, in the culture without addition of dodecane, only half of the above productivity (60 mg/L) was obtained at the same time period. Accordingly, the dodecane-added 2-phase culture system may be appropriately applied to an alternative transformation system to produce small lipophilic molecules.

Example 6: Production of Retinoid in Medium Including Lipophilic Substance

(92) The present example was performed to identify as to whether a variety of lipophilic substances have effects of increasing retinoid production.

(93) (1) Production of Retinoid in Medium Including Alkane

(94) A strain DH5 including transformed pT-DHBSR/pSNA (DH5(pT-DHBSR/pSNA)) was used, and after adding 5 mL of each of octane, decane, dodecane and tetradecane to 5 ml of medium, culturing was conducted according to such conditions as described in Bacteria strain and culture conditions. The medium used herein was a 2YT medium including 0.2% (w/v) of arabinose and 2.0% (w/v) of glycerol added thereto.

(95) FIG. 17 illustrates results of retinoid production in the presence of alkane. FIG. 18 illustrates results of growth of the retinoid producing strain in the presence of alkane.

(96) As shown in FIG. 17, a total 108 mg/L of retinoid was produced in case of using decane. Alternatively, bacterial cell proliferation, pH and an amount of -carotene in the bacterial cell did not show a considerable difference depending upon the presence of alkanes. Therefore, it is considered that decane may be more advantageous in retinoid production, as compared to dodecane. When using octane, production of retinal and retinol was similar to other alkanes, whereas retinyl acetate was almost not produced. Tetradecane showed a lower productivity of whole retinoids, as compared to other alkanes.

(97) (2) Production of Retinoid in Medium Including Mineral Oil

(98) (2.1) Lightweight Mineral Oil

(99) The lightweight mineral oil is cheap and has an economical advantage, as compared to alkanes. A strain DH5 including transformed pT-DHBSR/pSNA (DH5 (pT-DHBSR/pSNA)) was used, and after adding the lightweight mineral oil in different volumes to 5 ml of medium, respectively, culturing was conducted according to such conditions as described in Bacteria strain and culture conditions. The medium used herein was a 2YT medium including 0.2% (w/v) of arabinose and 2.0% (w/v) of glycerol added thereto.

(100) FIG. 19 illustrates results of retinoid production in the presence of lightweight mineral oil. FIG. 20 illustrates results of strain growth in the presence of lightweight mineral oil.

(101) As shown in FIG. 19, 158 mg/L of retinoid was produced in the presence of the lightweight mineral oil in an amount of 2 ml, as compared to 136.1 mg/L of retinoid produced using 5 mL of dodecane. As shown in FIG. 20, pH was not considerably different other than the case of using dodecane. On the other hand, bacterial cell growth was reduced as an amount of the lightweight mineral oil increased. The reason of this condition was deemed because the medium and mineral oil were not sufficiently admixed due to a high viscosity and specific gravity of the lightweight mineral oil. Owing to a decrease in growth of the bacterial cell, the retinoid production was also reduced.

(102) FIG. 21 illustrates cell specific retinoids productivity. As shown in FIG. 21, a specific productivity of about 5 mg/L/OD.sub.600 nm was observed regardless of an amount of mineral oil.

(103) (2.2) Heavy Mineral Oil

(104) The heavy mineral oil is cheaper than the lightweight mineral oil. A strain DH5 including transformed pT-DHBSR/pSNA (DH5(pT-DHBSR/pSNA)) was used, and after adding 2 ml of heavy mineral oil to 5 ml of medium, culturing was conducted according to such conditions as described in Bacteria strain and culture conditions. The medium used herein was a 2YT medium including 0.2% (w/v) of arabinose and 2.0% (w/v) of glycerol added thereto.

(105) FIG. 22 illustrates results of retinoid production in the presence of heavy mineral oil. FIG. 23 illustrates results of strain growth in the presence of heavy mineral oil. As shown in FIGS. 22 and 23, the heavy mineral oil involved lower cell growth, as compared to the lightweight mineral oil and dodecane. Further, 104.6 mg/L of retinoid was produced. The reason of this condition was deemed because the medium and mineral oil were not sufficiently admixed due to a viscosity of the heavy mineral oil.

(106) Except that a test tube was tilted and mounted on an incubator, cell culture was performed by the same procedures as described above. By tilting the test tube, effects of agitation were improved to thus allow the medium and mineral oil to be admixed more effectively.

(107) FIG. 24 illustrates results of retinoid production when the culturing was conducted in a tilted test tube. FIG. 25 illustrates results of strain growth when the culturing was conducted in the tilted test tube. As shown in FIGS. 24 and 25, the cell growth and retinoid production were increased when the culturing was conducted in the tilted test tube. More particularly, the retinoid was produced in an amount of 88.2 mg/L at 96 hours in a vertically-mounted test tube, while the retinoid productivity reached 173.9 mg/L in the tilted test tube.

(108) The above results indicated that mixing the lightweight and/or heavy mineral oils with the medium is an important factor in retinoid production since the mineral oils have a high viscosity. Accordingly, the foregoing lightweight and/or heavy mineral oils may be used for retinoid by properly agitating the same during culturing.

(109) (3) Production of Retinoid in Medium Including Skin-Friendly Lipophilic Substance

(110) Retinoid was produced in a medium including a skin-friendly lipophilic substance. As the skin-friendly lipophilic substance, isopropyl myristate (IPM), dioctanoyl-decanoyl glycerol (ODO), cetyl ethylhexanoate (CEH) and phytosqualane were used.

(111) A strain DH5 including transformed pT-DHBSR/pSNA (DH5(pT-DHBSR/pSNA)) was used, and after adding 2 ml of heavy mineral oil to 5 ml of medium, culturing was conducted according to such conditions as described in Bacteria strain and culture conditions. The medium used herein was a 2YT medium including 0.2% (w/v) of arabinose and 2.0% (w/v) of glycerol added thereto. A control was prepared by adding 5 mL of dodecane to the medium.

(112) FIG. 26 illustrates cell growth and pH in the presence of skin-friendly lipophilic substance. FIGS. 27 and 28 illustrate results of retinoid production depending on an amount of skin-friendly lipophilic substance. As shown in FIGS. 27 and 28, in case of lipophilic substances other than dodecane, using 2 ml of lipophilic substance has achieved an increase in retinoid productivity, as compared to 5 ml of the same. In other words, when 2 ml of lipophilic substance was added to 5 ml of medium including the lightweight mineral oil, IPM, ODO, CEH and phytosqualane, large amount of retinoid was produced. Especially, when using IPM among IPM, ODO, CEH and phytosqualane, the largest amount of retinoid was produced. More specifically, when adding 2 ml of IPM, 180 mg/L of retinoid was produced. For IPM, in consideration of similar growth of bacterial cell, it is presumed to have a high specific productivity per bacterial cell.