Advanced
Synthesis of Chlorogenic Acid and p-Coumaroyl Shikimates from Glucose Using Engineered Escherichia coli
Synthesis of Chlorogenic Acid and p-Coumaroyl Shikimates from Glucose Using Engineered Escherichia coli
Journal of Microbiology and Biotechnology. 2014. Aug, 24(8): 1109-1117
Copyright © 2014, The Korean Society For Microbiology And Biotechnology
  • Received : March 13, 2014
  • Accepted : April 27, 2014
  • Published : August 30, 2014
Download
PDF
e-PUB
PubReader
PPT
Export by style
Share
Article
Author
Metrics
Cited by
TagCloud
About the Authors
Mi Na Cha
Department of Bioscience and Biotechnology, Bio/Molecular Informatics Center, Konkuk University, Seoul 143-701, Republic of Korea
Hyeon Jeong Kim
Department of Bioscience and Biotechnology, Bio/Molecular Informatics Center, Konkuk University, Seoul 143-701, Republic of Korea
Bong Gyu Kim
Department of Forest Resources, Gyeongnam National University of Science and Technology, Jinju 660-758, Republic of Korea
Joong-Hoon Ahn
Department of Bioscience and Biotechnology, Bio/Molecular Informatics Center, Konkuk University, Seoul 143-701, Republic of Korea
jhahn@konkuk.ac.kr

Abstract
Chlorogenic acid and hydroxylcinnamoyl shikimates are major dietary phenolics as well as antioxidants, with recently discovered biological, activities including protection against chemotheraphy side effects and prevention of cardiovascular disease and cancer. Certain fruits and vegetables produce these compounds, although a microbial system can also be utilized for synthesis of chlorogenic acid and hydroxylcinnamoyl shikimates. In this study, we engineered Escherichia coli to produce chlorogenic acid and p -coumaroyl shikimates from glucose. For the synthesis of chlorogenic acid, two E. coli strains were used; one strain for the synthesis of caffeic acid from glucose and the other strain for the synthesis of chlorogenic acid from caffeic acid and quinic acid. The final yield of chlorogenic acid using this approach was approximately 78 mg/l. To synthesize p -coumaroyl shikimates, wild-type E. coli as well as several mutants were tested. Mutant E. coli carrying deletions in three genes ( tyrR , pheA , and aroL ) produced 236 mg/l of p -coumaroyl shikimates.
Keywords
Introduction
Hydroxycinnamic acids are phenolic acids found in plants, and contain a C6-C3 backbone. They are synthesized using the phenylpropanoid pathway and include p -coumaric acid, caffeic acid, ferulic acid, and sinapinic acid [3] . Hydroxycinnamic acids serve as building blocks for the biosynthesis of flavonoid, stilbene, coumarin, curcumin, and lignin [19] . Hydroxycinnamic acids can be conjugated with several compounds, including quinic acid or its derivatives such as shikimic acid [6] , 4-deoxyquinic acid [17] , and methyl or butyl quinate [16] , hydroxyl acids (malic acid or glucuronic acid), amino compounds (anthranilic acid or aromatic amino acids), and sugar alcohols (glycerol or myo -inositol) [16] .
Hydroxycinnamic acid conjugates, especially chlorogenic acid and hydroxycinnamoyl shikimates, are natural key dietary source of phenolic compounds and are found abundantly in coffee, fruits, and vegetables, and also serve as antioxidants [7] . A growing number of biological activities of hydroxycinnamate conjugates have been described [4 , 18] , including the protection against side effects of chemotherapy and prevention of cardiovascular disease and cancer. Chlorogenic acids are regarded as alternative antioxidants, which may be an alternative for traditional synthetic antioxidants such as butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT) [18] .
Hydroxycinnamic acid conjugates are synthesized by hydroxycinnamoyl transferase (HCT). The activated form of hydroxycinnamate (hydroxycinnamoyl-coenzyme A[CoA]) is used as an acyl donor, while quinic acid or shikimic acid is used as the acyl acceptor [8] . Hydroxycinnamic acid is derived from phenylalanine or tyrosine, and is a substrate of HCT. Tyrosine is converted to p -coumaric acid by tyrosine ammonia lyase (TAL). Subsequent hydroxylation and O -methylation lead to the formation of caffeic acid, ferulic acid, and sinapicic acid [19] . Before conjugation with acyl acceptors, hydroxycinnamic acids are activated by the attachment of CoA via 4-coumarate:CoA ligase (4CL). The pathway from tyrosine to hydroxycinnamoyl-CoA biosynthesis is well characterized in plants [19 , 20] , and has been introduced in Escherichia coli for the synthesis of hydroxycinnamic acids using E. coli strains that overproduce tyrosine [9 , 13 , 21] . The other substrate, either quinic acid or shikimic acid, is an intermediate in the shikimic acid pathway, leading to the synthesis of aromatic amino acids. Quinic acid is synthesized from a shikimate pathway intermediate, 3-dehyroquinate, by quinate dehydrogenase [14] . In E. coli , most of the 3-dehydroquinate is converted into 3-dehydroshikimate, and only a limited amount of 3-dehydroquinate is converted into quinic acid. Only small amounts of chlorogenic acid are therefore produced from caffeic acid in wild-type E. coli expressing HCT and 4CL even after overexpression of quinate dehydrogenase [10] . Thus, to accumulate 3-dehydroquinate, 3-dehydroquinate dehydratase (AroD), an enzyme that catalyzes the conversion of 3-dehydroquinate into 3-dehydroshikimate, was deleted [10] , blocking the synthesis of aromatic amino acids, including tyrosine, phenylalanine, and tryptophan. It makes it impossible to synthesize p -coumaric acid and eventually caffeic acid in the aroD deletion mutant. However, shikimic acid is synthesized from 3-dehydroshikimic acid and is phosphorylated by either AroK (shikimate kinase I) or AroL (shikimate kinase II) in E. coli [5] . Mutation of either aroK or aroL therefore does not prevent the shikimate pathway, and mutants with a deletion in either aroK or aroL can synthesize tyrosine, a precursor for p -coumaric acid.
Previously, 450 mg/l of chlorogenic acid and 235 mg/l of p -coumaroyl shikimates were synthesized from caffeic acid and p -coumaric acid, respectively, using engineered E. coli [10] . In this study, we attempted to synthesize hydroxycinnamic acid conjugates (chlorogenic acid and p -coumaroyl shikimates) from glucose in E. coli . However, the two substrates for HCT are synthesized using the shikimate pathway. Since an aroD deletion has to be made to supply quinic acid, the chlorogenic acid biosynthesis from glucose was not possible. In this mutant, the synthesis of tyrosine, which is a precursor of caffeic acid, was blocked. In order to circumvent this problem, two different E. coli strains were used; one strain to synthesize caffeic acid from glucose, and the other strain to synthesize chlorogenic acid from caffeic acid. Using this approach, 78 mg/l of chlorogenic acid was successfully synthesized from glucose. Using a single mutant that accumulates shikimic acid but does not block tyrosine biosynthesis, we synthesized p -coumaroyl shikimates from glucose in E. coli . Using combination of mutations in various genes, including aroL , aroK , and pheA in E. coli , up to 236 mg/l of p -coumaroyl shikimates could be produced.
Materials and Methods
- E. coli Strains and Plasmids
For the synthesis of caffeic acid, three genes ( aroG (2-dehydro-3-deoxyphosphoheptonate aldolase, AM946981.2), tyrA (chorismate mutase/prephenate dehydrogenase, AM946981.2), and SeTAL (tyrosine amino lyase, DQ357071.1)) were cloned as described previously [11] . These three genes were subcloned into pAYCDuet vector (Novagen). Each gene contained an independent T7 promoter and ribosome binding site. The resulting construct was named pA-aroG-SeTAL-tyrA ( Table 1 ). The HpaBC gene (Z29081) was cloned from E. coli using genomic DNA as template. Primers were as the following; 5’-AT GAATTC GATGAAACCAGAAGATTTC CGC-3’ used as the forward primer and 5’-CAT GCGGCCGC TTAAATCGCAGCTTCCATT-3’ as the reverse primer. EcoRI and NotI sites (underlined) were inserted into the forward and the reverse primer, respectively. The resulting PCR product was sequenced, digested with Eco RI/ Not I and subcloned into the corresponding sites in the pETDuet-1 vector. The resulting construct was named pE-HpaBC. The E. coli strain for caffeic acid synthesis was B-TP [11] , which contained deletions in both the tyrR and pheA genes in E. coli BL21(DE3). Strain B-TP was transformed with pA-aroG-SeTAL-tyrA and pE-HpaBC. The resulting transformant, B-TP-CA2, was used for the production of caffeic acid from glucose.
Plasmids and strains used in the present study.
PPT Slide
Lager Image
Plasmids and strains used in the present study.
Constructs for the synthesis of chlorogenic acid from caffeic acid were as previously described [11] . Both hydroxycinnamoyl-CoA quinate transferase from Nicotiana tobacco (NtHQT) and 4-coumarate:CoA ligase from Oryza sativa (Os4CL) were subcloned into pCDFDuet-1 (pC-HQT-Os4CL in Table 1 ). ydiB , encoding quinate dehydrogenase from E. coli , was subcloned into the pETDuet-1 vector (pE-EcydiB in Table 1 ). B-aroD, in which the aroD gene was deleted in E. coli BL21(DE3), was used [11] . pC-NtHQT-Os4CL and pE-EcydiB were transformed into B-aroD (B-102 in Table 1 ).
For the synthesis of p -coumaroyl shikimates from p -coumaric acid and shikimate, hydroxycinnamate-CoA shikimate transferase ( NtHST ) and Os4CL were subcloned into pCDFDuet-1 (pC-NtHST-Os4CL in Table 1 ). E. coli strains, B-aroK and B-aroL, which contain a deletion in shikimate kinase I ( aroK ) and II ( aroL ), respectively, were as previously described [11] . E. coli triple mutants, B-TPL and B-TPK, were constructed using the Quick and Easy Conditional Knockout Kit (Gene Bridges, Heidelberg, Germany) according to the manufacturer’s protocol ( Table 1 ). Each strain (BL21(DE3), B-aroL, B-aroK, B-TPL, and B-TPK) was transformed with pA-aroG-SeTAL-tyrA and pC-NtHST-Os4CL and the transformant was named B-CS, B-aroL-CS, B-aroK-CS, B-TPL-CS, or B-TPK-CS ( Table 1 ).
- Production of Caffeic Acid, Chlorogenic Acid, and p-Coumaroyl Shikimates
For comparison of the caffeic acid production, strains B-CA1, B-CA2, B-TP-CA1, and B-TP-CA2 ( Table 1 ) were grown in LB containing 50 μg/ml ampicillin and 50 μg/ml chloramphenicol at 37℃ for 18 h. Then 500 μl of preculture was inoculated into 50 ml of LB containing the same antibiotics and grown for 6 h. Cells were harvested and the cell concentration was adjusted to OD 600 = 1 with M9 medium containing 2% glucose, 50 μg/ml ampicillin, 50 μg/ml chloramphenicol, and 1 mM IPTG. Cells were then cultured for 12 h at 30℃. The production of caffeic acid from each culture was analyzed using high-performance liquid chromatography (HPLC, Varian).
For the production of chlorogenic acid, strain B-TP-CA2 was grown as described above. After cultivation for 10 h, cells were removed by centrifugation, and the supernatant containing caffeic acid was used for resuspension of strain B-102. Strain B-102 was grown in LB containing 50 μg/ml ampicillin and 50 μg/ml chloramphenicol at 37℃, overnight and then inoculated in 50 ml of LB containing 50 μg/ml ampicillin and 50 μg/ml chloramphenicol. After the OD 600 reached 0.8, IPTG was added to the final concentration of 1 mM. The mixture was grown at 18℃ for 48 h with shaking at 180 rpm. Cells were then harvested, resuspended with culture filtrate from strain B-TP-CA2, and incubated at 30℃. Chlorogenic acid production was periodically monitored. The reaction was stopped by boiling for 5 min and the product was centrifuged for 15 min at 13,000 × g to remove the cell debris. The supernatant was analyzed using HPLC.
For the production of p -coumaroyl shikimates, an overnight culture of E. coli strain B-CS, B-aroL-CS, B-aroK-CS, B-TPL-CS, or B-TPK-CS was inoculated in LB containing 50 μg/ml spectinimycin and 50 μg/ml chloramphenicol, and grown until the OD 600 reached 1. The cells were harvested, resuspended in M9 containing 2% glucose, 50 μg/ml spectinimycin, 50 μg/ml chloramphenicol, and 1 mM IPTG, and incubated at 30℃ for 48 h with shaking at 180 rpm. The reaction product was extracted with two volume of ethylacetate. The supernatant was dried and dissolved with dimethylsulfoxide (DMSO).
Analysis of metabolites was performed using HPLC as described previously [11] . The mean and standard error of the mean were calculated from triplicate experiments. An analysis of variance (ANOVA) was carried out with Tukey’s test of the pairwise comparisons of the experimental strains with a significance level of p = 0.01, using Excel 2010 (Microsoft Corp., Redmond, WA, USA).
Results and Discussion
- Production of Chlorogenic Acid from Glucose in E. coli
Chlorogenic acid is a conjugate of caffeic acid and quinic acid. Two E. coli strains were used to produce chlorogenic acid; one E. coli strain synthesized caffeic acid from glucose, and the other strain was designed to synthesize chlorogenic acid using caffeic acid synthesized by the first E. coli strain ( Fig. 1 A). Using B-CA1, which contains two genes, TAL and HpaBC ( Fig. 1 A), the production of caffeic acid was examined. As shown in Fig. 2 A- 3 , a new peak at 5.3 min was observed with the same HPLC retention time as caffeic acid. In addition, MS/MS data showed a similar fragmentation pattern between the reaction product and caffeic acid. We also detected p -coumaric acid (a peak at 6.5 min in Fig. 2 A- 3 ). However, E. coli harboring empty vectors did not make any product ( Fig. 2 A).
PPT Slide
Lager Image
Schemas of the biosynthesis of caffeic acid, chlorogenic acod, and p-coumaroyl-shikimates using E. coli. (A) Illustration of the biosynthesis of chlorogenic acid using two engineered Escherichia coli strains. The first E. coli strain was engineered to synthesize caffeic acid and the second strain to synthesize chlorogenic acid. (B) Synthesis of p-coumaroyl-shikimates from glucose using engineered E. coli.
PPT Slide
Lager Image
Analysis of the caffeic acid produced by different engineered strains of E. coli. (A) Analysis of Escherichia coli strain B-CA1 caffeic acid production using HPLC. 1, Culture of E. coli harboring empty vectors; 2, standard caffeic acid. (Inset is the MS/MS profile of caffeic acid.); 3, culture of strain B-CA1. (Inset is the MS/MS profile of the reaction product.) The MS analysis was done in negative electrospray ionization (ESI-) mode. (B) Production of caffeic acid in various E. coli strains. 1, Strain B-CA1; 2, strain B-TP-CA1; 3, strain B-CA2; 4, strain B-TP-CA2. (C) Production of caffeic acid using strain B-TP-CA2.
PPT Slide
Lager Image
Production of chlorogenic acid using two E. coli strains. Strain B-TP-CA2 was used for the production of caffeic acid for 9 h and strain B-102 was then used to produce chlorogenic acid.
Both aroG and tyrA were overexpressed to increase the production of tyrosine, as it serves as a precursor of caffeic acid. AroG utilizes phosphoenolpyruvate and erythrose 4-phosphate as substrates to produce 3-deoxy-D-arabino-heptulosonate 7-phosphate, the first molecule in the shikimate pathway. Overexpression of tyrA leads to the increased production of 4-hydroxyphenylpyruvate, a precursor of tyrosine. Strain B-CA2 expressing aroG and tyrA produced approximately 39 mg/l, whereas strain B-CA1 produced approximately 16 mg/l, indicating that the amount of tyrosine is important for the production of caffeic acid. We also used an E. coli mutant strain B-TP ( Table 1 ), in which two genes, tyrR and pheA , were deleted. This strain was transformed with pA-SeTAL and pE-HpaBC or pA-aroG-SeTAL-tyrA and pE-HpaBC. The resulting strains were named B-TP-CA1 and B-TP-CA2, respectively ( Table 1 ). B-TP-CA1 produced 32 mg/l of caffeic acid and B-TP-CA2 produced 54 mg/l ( Fig. 2 B). Therefore, strain B-TP-CA2 was chosen as the optimal strain for the production of caffeic acid and used for further studies. We also monitored the production of caffeic acid in strain B-TP-CA for 30 h. A maximal production of 341 mg/l of caffeic acid was achieved at 24 h growth, with a decrease in production observed after 24 h ( Fig. 2 C).
Two E. coli strains were employed to produce chlorogenic acid. The first strain, B-TP-CA2, was used for the production of caffeic acid. To determine the optimal concentration of caffeic acid on the production of chlorogenic acid, we harvested B-TP-CA2 culture filtrate at 6, 9, and 12 h. The caffeic acid concentration was 30 mg/l at 6 h, 48 mg/l at 9 h, and 54 mg/l at 12 h. Strain B-102 ( Table 1 ), which was used for the production of chlorogenic acid from caffeic acid, was resuspended with each culture filtrate and the resulting mixture was cultured for 24 h. Caffeic acid was fully converted to chlorogenic acid in B-102 resuspended with 6 h culture filtrate. In the culture with B-102 resuspended with 9 or 12 h culture filtrate, caffeic acid was still present but more chlorogenic acid was produced in B-102 cultivated with 9 h culture filtrate. The higher concentration of caffeic acid likely inhibited the production of chlorogenic acid. B-TP-CA2 with 9 h culture filtrate was used to produce chlorogenic acid with B-102 as it is likely that unknown metabolite(s) from the culture filtrate of B-TP-CA2 would influence the production of chlorogenic acid and the concentration of the metabolites would increase as cells grow. At 9 h, approximately 54 mg/l caffeic acid was synthesized. The culture filtrate was used to resuspend B-102 to an OD 600 of 3. Strain B-102, in which aroD w as deleted, accumulated higher levels of quinic acid and/or dehydroquinic acid and was used to produce chlorogenic acid from caffeic acid [10] . The cells were cultured and the production of chlorogenic acid was monitored. The chlorogenic acid synthesis rapidly increased until 33 h and reached a maximal at 45 h, at which approximately 78 mg/l of chlorogenic acid was produced ( Fig. 3 ).
- Production of p-Coumaroyl Shikimates from Glucose Using E. coli
Next, we synthesized p -coumaroyl shikimates from glucose using E. coli . Shikimic acid and p -coumaric acid are needed to synthesize p -coumaroyl shikimates ( Fig. 1 B). Therefore, modulation of the quantities of p -coumaric acid and shikimic acid in E. coli is necessary. The amount of shikimate in E. coli was increased by deletion of either aroL or aroK , both of which encode enzymes catalyzing the conversion of shikimate to shikimate 3-phosphate [2 , 12] . To increase the supply of p -coumaric acid, increased cellular tyrosine was necessary. Previous studies showed that deletion of both tryR and pheA led to increased tyrosine [15] . Therefore, we tested four mutants (BaroL, BaroK, B-TPL, and B-TPK in Table 1 ) for the production of p -coumaroyl shikimates. After incubation for 48 h, p -coumaroyl shikimates production was measured. Our previous study showed that the four reaction products were 5- p -coumaroyl shikimate (P1), 3,4-di- p -coumaroyl shikimate (P2), 3,5-di- p -coumaroyl shikimate (P3), and 4,4-di- p -coumaroyl shikimate (P4) [10] . B-aroL-CS produced 204.7 mg/l p -coumaroyl shikimates, which was greater than that by BaroK-CS (194.7 mg/l). B-TPL-CS produced 236.0 mg/l and B-TPK-CS produced 221.3 mg/l ( Fig. 4 A). Thus, strain B-TPL was the most effective among the tested strains for the production of p -coumaroyl shikimates.
PPT Slide
Lager Image
Production of p-coumaroyl shikimates by engineered E. coli strains. (A) Production of p-coumaroyl shikimates using different E. coli strains. (B) Production of p-coumaroyl shikimates using B-TPL-CS. Product 1, 5-p-coumaroyl shikimate; product 2, 3,4-di-p-coumaroyl shikimate; product 3, 3,5-di-p-coumaroyl shikimate; and product 4, 4,5-di-p-coumaroyl shikimate.
Using B-TPL-CS, the production of p -coumaroyl shikimates was monitored for 48 h. Di- p -coumaroyl shikimates (3,4-di- p -coumaroyl shikimate, 3,5-di-p-coumaroyl shikimate, and 4,4-di- p -coumaroyl shikimate) were rapidly produced whereas mono- p -coumaroyl shikimate was not observed until 9 h of incubation. After 9 h, production of di- p -coumaroyl shikimates was almost to saturation and production slowed. However, mono- p -coumaroyl shikimate production continued to increase until 33 h. The final yield of p -coumaroyl shikimates was approximately 236 mg/l at 48 h ( Fig. 4 B), but after 48 h, total synthesized p -coumaroyl shikimates decreased slightly.
We successfully synthesized chlorogenic acid and p -coumaroyl shikimates from glucose in engineered E. coli . This is the first report that shows the biological synthesis of chlorogenic acid and p -coumaroyl shikimates from glucose. The price of glucose is much cheaper than that of hydroxycinnamic acids such as caffeic acid and p -coumaric acid. Therefore, it had advantages to the previous study, in which chlorogenic acid and p -coumaroyl shikimates were synthesized from hydroxycinnamic acids [10] . The final yield of chlorogenic acid was approximately 78 mg/l, which was much less than that produced by feeding caffeic acid (450 mg/l) [10] . The low yield is likely due to the fact that two E. coli strains were used, and the higher concentration of caffeic acid produced by the first E. coli strain may have inhibited the production of chlorogenic acid. However, the yield of p -coumaroyl shikimate was comparable with that produced by feeding p -coumaric acid, due to the further catalyzation of AroK or AorL. Deletion of either aroK or aroL did not affect the production of tyrosine, which is a precursor of p -coumaric acid. Additionally, engineering of the shikimic acid pathway increased the tyrosine supply for the synthesis of p -coumaric acid.
The yield of chlorogenic acid synthesized from glucose was lower than from caffeic acid; however, this yield is still important because the price of caffeic acid is higher than that of glucose. Other groups have also reported the production of caffeic acid in E. coli [9 , 13 , 21] . The yield in each study ranged from 50-150 mg/l, which is much less than that observed in this study (341 mg/l). The difference in tyrosine levels in each E. coli strain used in each study may be associated with the final yield of caffeic acid, although the highest tyrosine production did not give the highest yield of caffeic acid. Therefore, a metabolic balance between tyrosine biosynthesis and caffeic acid is crucial for a greater production of caffeic acid.
Acknowledgements
This work was supported by a grant from the Next- Generation BioGreen 21 Program (PJ00948301), Rural Development Administration, Republic of Korea, and by the Priority Research Centers Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2009-0093824).
References
Clifford MN 1999 Chlorogenic acids and other cinnamates - nature, occurrence and dietary burden. J. Sci. Food Agric. 79 362 - 372    DOI : 10.1002/(SICI)1097-0010(19990301)79:3<362::AID-JSFA256>3.0.CO;2-D
Draths KM , Knop DR , Frost JW 1999 Shikimic acid and quinic acid: replacing isolation from plant sources with recombinant microbial biocatalysis. J. Am. Chem. Soc. 121 1603 - 1604    DOI : 10.1021/ja9830243
Dixon RA , Paiva NL 1995 Stress-induced phenylpropanoid metabolism. Plant Cell 7 1085 - 1097    DOI : 10.1105/tpc.7.7.1085
El-Seedi HR , El-Seed AMA , Khalifa SAM , Bohlin L , Borg-Karlson AK , Verpoorte R 2012 Biosynthesis, natural sources, dietary intake, pharmacokinetic properties, and biological activities of hydroxycinnamic acids. J. Agric. Food Chem. 60 10877 - 10895    DOI : 10.1021/jf301807g
Ely B , Pittard J 1979 Aromatic amino acid biosynthesis: regulation of shikimate kinase in Escherichia coli K-12. J. Bacterol. 138 933 - 943
Goupy PM , Varoquaux PJA , Nicolas JJ , Macheix JJ 1990 Identification and localization of hydroxycinnamoyl and flavonol derivatives from endive (Cichorium endivia L cv Geante Maraichere) leaves. J. Agric. Food Chem. 38 2116 - 2121    DOI : 10.1021/jf00102a003
Herrmann K 1989 Occurrence and content of hydroxycinnamic acid and hydoxybenzoic acid compounds in foods. Crit. Rev. Food Sci. Nutr. 28 315 - 347    DOI : 10.1080/10408398909527504
Hoffmann L , Maury S , Martz F , Geoffroy P , Legrand M 2003 Purification, cloning, and properties of an acyltransferase controlling shikimate and quinate ester intermediates in phenylpropanoid metabolism. J. Biol. Chem. 278 95 - 103    DOI : 10.1074/jbc.M209362200
Kang S-Y , Choi O , Lee JK , Hwang BY , Im T-B , Hong Y-S 2012 Artificial biosynthesis of phenylpropanoic acids in a tyrosine overproducing Escherichia coli s train. Microb. Cell Fact. 11 153 -    DOI : 10.1186/1475-2859-11-153
Kim BG , Jung WD , Mok H , Ahn J-H 2013 Production of hydroxycinnamoyl-shikimates and chlorogenic acid in Escherichia coli: production of hydroxycinnamic acid conjugates. Microb. Cell Fact. 12 15 -    DOI : 10.1186/1475-2859-12-15
Kim MJ , Kim B-G , Ahn J-H 2013 Biosynthesis of bioactive O-methylated flavonoids in Escherichia coli. Appl. Microbiol. Biotechnol. 97 7195 - 7204    DOI : 10.1007/s00253-013-5020-9
Krämer M , Bongaerts J , Bovenberg R , Kremer S , Müller U , Orf S 2003 Metabolic engineering for microbial production of shikimic acid. Met. Eng. 5 277 - 283    DOI : 10.1016/j.ymben.2003.09.001
Lin Y , Yan Y 2012 Biosynthesis of caffeic acid in Escherichia coli using its endogenous hydroxylase complex. Microb. Cell Fact. 11 42 -    DOI : 10.1186/1475-2859-11-42
Lindner HA , Nadeau G , Matte A , Michel G , Ménard R , Cygler M 2005 Site-directed mutagenesis of the active site region in the quinate/shimate 5-dehydrogenase YdiB of Escherichia coli. J. Biol. Chem. 280 7162 - 7169    DOI : 10.1074/jbc.M412028200
Lütke-Eversloh T , Stephanopoulos G 2007 L-Tyrosine production by deregulated strains of Escherichia coli. Appl. Microbiol. Biotechnol. 75 103 - 110    DOI : 10.1007/s00253-006-0792-9
Rumbero-Sanchez A , Vazquez P 1991 Quinic acid esters from Isertia haenkeana. Phytochemistry 30 311 - 313    DOI : 10.1016/0031-9422(91)84144-H
Stevenson PC , Anderson JC , Blaney M , Simmonds MSJ 1993 Developmental inhibition of Spodoptera litura (Fab) larvae by a novel caffeoylquinic acid from wild groundnut Arachis paraguariensis (Chod et Hassl). J. Chem. Ecol. 19 2917 - 2933    DOI : 10.1007/BF00980592
Upadhyay R , Rao LJM 2013 An outlook on chlorogenic acids - occurrence, chemistry, technology, and biological activities. Crit. Rev. Food Sci. Nutr. 53 968 - 984    DOI : 10.1080/10408398.2011.576319
Vogt T 2010 Phenylpropanoid biosynthesis. Mol. Plant 3 2 - 20    DOI : 10.1093/mp/ssp106
Winkel-Shirley B 2001 Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol. 126 485 - 493    DOI : 10.1104/pp.126.2.485
Zhang H , Stephanopoulos G 2013 Engineering E. coli for caffeic acid biosynthesis from renewable sugars. Appl. Microbiol. Biotechnol. 97 3333 - 3341    DOI : 10.1007/s00253-012-4544-8