Characterization of catechol-O-methyltransferases through de novo biosynthesis of phenylpropanoid derivatives 

Catechol-O-methyltransferases (COMTs) catalyze regioselective O-methylation of phenolic substrates that are valuable biocatalysts to produce functionalized aromatic compounds. In this study, four COMTs originating from diverse biological sources – LwOMT10 (Lophophora williamsii), AtOMT (Arabidopsis thaliana), sCOMT (Homo sapiens), and MtbOMT (Mycobacterium tuberculosis) – were evaluated for enzymatic activity and regiospecificity toward cinnamic acid derived substrates from the lignin pathway. Each enzyme was expressed in Escherichia coli BL21(DE3*) and screened against caffeic acid, ferulic acid, isoferulic acid, and 3,4-dimethoxycinnamic acid. Product formation and regioselectivity were quantified using high-performance liquid chromatography-mass spectrometry (HPLC-MS). Based on screening performance, selected COMTs were integrated into engineered biosynthetic pathways that enabled de novo production of 3,4-dimethoxycinnamic acid from L-tyrosine. Multigene operons incorporating tyrosine ammonia-lyase (RgTAL), hydroxyphenyl acetate monooxygenase (EcHpaBC) in either wild type or mutant variants were assembled with sCOMT and AtOMT using plasmid construction. Recombinant strains were fermented under inducible conditions, and metabolites were analyzed using HPLC-MS to assess pathway efficiency and methylation. Comparative analysis discovered distinct substrate preference and regiospecific methylation profiles among the four COMTs, with sCOMT and AtOMT demonstrating the strongest catalytic activity and effective conversion to demethylated products. The engineered strains were successful in achieving de novo biosynthesis of 3,4-dimethoxycinnamic acid directly from tyrosine, which confirms functional pathway integration and enzyme compatibility. The results highlight the importance of COMT selection for pathway optimization and demonstrate a versatile microbial platform for sustainable biosynthesis of high-value aromatic compounds.

Catechol- O -methyltransferases (COMTs) are a diverse class of enzymes that catalyze the regioselective O-methylation of phenolic compounds, playing a pivotal role in the biosynthesis and modification of aromatic metabolites across biological systems. Their ability to functionalize aromatic rings makes them valuable biocatalysts for the synthesis of structurally diverse and bioactive molecules. While COMTs are widespread in nature, systematic comparative analyses of their regiospecificity and compatibility within engineered biosynthetic pathways remain limited. In cellular metabolism, S-adenosylmethionine (SAM) serves as the universal methyl donor, providing the activated methyl group utilized by methyltransferases—including COMTs—to methylate aromatic hydroxyl groups (Fig. 1). In this reaction, SAM donates its methyl group to a nucleophilic hydroxyl on the aromatic substrate, yielding meta- and para- O-methylated products and generating S-adenosyl-L-homocysteine (SAH) as a byproduct. Notably, SAH often acts as a competitive inhibitor of methyltransferases, underscoring the need to balance methyl donor availability and product inhibition for optimal pathway function. This methylation process is especially critical for the modification of metabolic intermediates in engineered microbial systems such as Escherichia coli.

The present study aims to systematically characterize the catalytic activity and regiospecific preferences of four phylogenetically diverse COMTs: LwOMT from Lophophora williamsii, AtOMT from Arabidopsis thaliana, sCOMT from Homo sapiens, and MtbOMT from Mycobacterium tuberculosis. We sought to assess each enzyme’s ability to support de novo biosynthesis of 3,4-dimethoxycinnamic acid from L-tyrosine in a recombinant microbial host. Specifically, our objectives were to: (i) evaluate substrate specificity and methylation patterns across a panel of cinnamic acid-derived intermediates, (ii) identify COMTs capable of sequential O-methylation leading to the formation of dimethoxylated products, and (iii) functionally integrate selected COMTs into engineered E. coli strains for the biosynthetic production of 3,4-dimethoxycinnamic acid. Phenylpropanoid compounds were chosen as preferred substrates due to their conjugated aromatic ring systems bearing variable numbers of hydroxyl groups, which are amenable to O-methylation by these enzymes.

Beyond their metabolic relevance, phenylpropanoid derivatives are increasingly recognized for their potent antioxidant and anti-inflammatory activities, which position them as promising candidates for the development of nutraceuticals and therapeutic agents targeting oxidative stress, inflammation, and associated chronic diseases. Recent studies suggest these compounds exhibit potential in anticancer, anti-inflammatory, and anti-atherosclerotic applications (Afnan et al.), though further clinical evaluation is necessary to validate their efficacy.

An additional motivation for this research lies in the structural similarity of these phenylpropanoid compounds to mescaline, a molecule of interest for its potential role in treating neurodegenerative and psychiatric disorders such as Parkinson’s disease, schizophrenia, and anxiety. Given the involvement of COMTs in the metabolism of such bioactive molecules, further investigation into their regioselective preferences and functional integration is essential for advancing metabolic engineering strategies. Comprehensive understanding of COMT activity and substrate scope will facilitate the rational design of biosynthetic pathways for the sustainable production of high-value aromatic compounds and support the development of novel therapeutic agents.

Project Goals

1. Characterization of activity and regioselectivity of 4 COMT enzyme candidates

2. Assembly of complete biosynthetic pathway for 3,4-dimethoxycinnamic acid

Motivation

Methoxy functionality is common on many bioactive compounds and characterization of these OMTs will enable novel biosynthetic routes to be developed.

COMT selection and activity screening. A variety of catechol-O-methyltransferases (COMTs) were initially screened to evaluate enzymatic activity and regiospecificity for phenolic substrates. Four COMTs were selected for further characterization: LwOMT10 (Lophophora williamsii), AtOMT (Arabidopsis thaliana), sCOMT (Homo sapiens), and MtbOMT (Mycobacterium tuberculosis).

Each COMT expression construct was transformed into Escherichia coli BL21*(DE3) electrocompetent cells and plated on LB agar supplemented with ampicillin (80µg/mL). Plates were incubated overnight at 37℃ to obtain a screen of discrete colonies. Individual colonies were inoculated into Andrew’s Magic Media (AMM) containing ampicillin and the appropriate substrate and grown for approximately 14 hours at 37℃ with shaking (250rpm).

Substrates evaluated included caffeic acid, ferulic acid, and isoferulic acid. Overnight cultures were transferred into 48-well plates containing fresh AMM, substrate, and ampicillin. Protein expression was induced with isopropyl β-D-1-thiogalactopyranoside (IPTG) 4 hours post-inoculation. Cultures were incubated for an additional 48-hour period at 37℃ with shaking. Plates were centrifuged for 10 min, and supernatants were collected for metabolite analysis by High-Performance Liquid Chromatography-Mass Spectrometry (HPLC-MS).

Fermentation of pETM6-EcHpaBC(wt)-RgTAL(syn)–sCOMT-AtOMT. Recombinant plasmids were transformed into E. coli expression strain BL21*(DE3) by electroporation and selected on LB agar plates containing ampicillin (80µg/mL). Single colonies were assumed to have identical plasmid constructs and were used to inoculate overnight cultures in a 48-well plate containing AMM supplemented with ampicillin with or without 200mg/L L-tyrosine and ± casamino acids. An empty vector BL21*(DE3) strain was included as a negative control.

Overnight cultures were incubated for approximately 14 hours at 37℃ with shaking. Aliquots (40µL) of each overnight culture were transferred into a fresh 48-well plate containing ampicillin, AMM ± tyrosine and ± casamino acids. Cultures were induced with IPTG 4 hours post-inoculation and incubated for an additional 72-hours to enable growth. Cells centrifuged and the supernatants were transferred to a 96-well sampling plate for HPLC-MS analysis.

Plasmid reconstruction and DNA manipulation. Based on COMT activity screening results, additional cloning was performed to enable de novo biosynthesis of 3,4-dimethoxycinnamic acid from tyrosine. Using a pETM6 backbone, the genes encoding EcHpaBC (wild type) and RgTAL (synthetic) were digested, extracted, and ligated in operon configuration to generate the pETM6-EcHpaBC(wt)-RgTAL(syn) intermediate construct. In parallel, sCOMT and AtOMT were cloned into pETM6 to generate pETM6-sCOMT-AtOMT fragment.

A mutant variant, EcHpaBC(YI), was similarly ligated with RgTAL(syn) to produce pETM6-EcHpaBC(YI)-RgTAL(syn). The intermediate fragments were digested and ligated to assemble the full biosynthetic plasmids pETM6-EcHpaBC(wt)-RgTAL(syn)-sCOMT-AtOMT and pETM6-EcHpaBC(YI)-RgTAL(syn)-sCOMT-AtOMT. Final constructs were transformed into E. coli BL21*(DE3) via electroporation and cultured under fermentation conditions for 72-hours in AMM supplemented with ampicillin, with and without 200mg/L tyrosine. Culture supernatants were collected and analyzed by HPLC-MS. The top three producing colonies for sCOMT and AtOMT were then chosen for further evaluation of methylation for aromatic compounds.

Restriction digestion, agarose gel electrophoresis, and DNA purification. Plasmid backbones and insert fragments were generated by restriction enzyme digestion using sequence-specific endonucleases selected based on compatible overhangs and internal cut sites. Digestion reactions were performed by agarose gel electrophoresis using 0.8% agarose gels prepared in 1x TAE buffer containing ethidium bromide. Electrophoresis was performed at 120V for 25 minutes. DNA bands were visualized using a UV transilluminator, and bands corresponding to the expected fragment sizes were extracted using blades. Excised gel fragments were purified using a gel extraction kit according to the manufacturer’s protocol.

DNA Ligation, cloning, and plasmid verification. Purified DNA inserts and linearized plasmid backbones were ligated using T4 DNA ligase following the standard ratios of insert to backbone. Ligation reactions were assembled into a total volume of 10µL using ligase buffer. Ligated products were initially transformed into E. coli DH5α cells for plasmid propagation and sequence verification. The transformed products were recovered in SOC medium at 37℃ and placed in a shaking incubator for 45 minutes. The cells were plated onto LB agar plates supplemented with ampicillin (80µg/mL). Individual colonies were selected and cultured overnight with high concentration LB and ampicillin for plasmid isolation using a miniprep kit. Isolated plasmids were screened by restriction enzyme digestions to confirm correct insert size and orientation. Digestion products were resolved by 0.8% agarose gel electrophoresis and visualized using UV transillumination as described above. Correctly assembled plasmids were selected for further testing and subsequently transformed into electrocompetent E. coli BL21*(DE3) cells for fermentation experiments.

Control cloning for product verification. To further validate pathway performance and quantify 3,4-dimethoxycinnamic acid production, control constructs were generated by ligating the pETM6-EcHpaBC(wt)-RgTAL(syn) backbone individually with either sCOMT or AtOMT. These constructs were transformed into BL21*(DE3), fermented under identical conditions, and analyzed by HPLC-MS.

Analytical methods for altering HPLC-MS viewing window. HPLC-MS conditions were adjusted to identify the optimal window for metabolite detection and quantification. The (column name) was consistently used, and the sequence was modified to negative mode with 30% acetonitrile (ACN) for improved analysis of intermediates and products.

LwOMT10

LwOMT10 exhibited substrate-dependent activity consistent with prior expectations. No detectable methylated intermediate or downstream products were observed under the conditions tested. When cultures expressing LwOMT10 were supplemented with 200mg/L caffeic acid, HPLC-MS analysis revealed no formation of methylated products or pathway intermediates above the background levels. Similarly, incubation with 200mg/L ferulic acid resulted in no detectable product formation, indicating limited catalytic activity toward this substrate in vivo. Given the lack of observable conversion of caffeic and ferulic acid, LwOMT10 was not further evaluated using isoferulic acid as a substrate and further analysis was not performed.

MtbOMT

MtbOMT displayed catalytic activity consistent with previously reported regiospecific methylation of hydroxylated aromatic substrates. With supplementation of 200mg/L caffeic acid, E. coli strains expressing MtbOMT produced both ferulic and isoferulic acid, indicating methylation at the para and meta hydroxyl positions, respectively. The relative abundance of these products suggested approximately equal conversion at each position. However, when cultures were supplemented with either ferulic acid or isoferulic acid, no formation of 3,4-dimethoxycinnamic acid was detected by HPLC-MS. These results indicate that while MtbOMT is capable of single-step methylation at either the meta or para position, it does not catalyze sequential di-methylation of monomethylated intermediates under the conditions tested.

sCOMT

sCOMT demonstrated robust catalytic activity with regiospecific preference toward meta-position methylation of hydroxylated aromatic substrates. When E. coli cultures expressing sCOMT were supplemented with 200mg/L caffeic acid, HPLC-MS analysis detected the formation of both ferulic acid and isoferulic acid, with ferulic acid present at higher relative abundance, indicating preferential para-to-meta methylation dynamics consistent with sCOMT specificity. No formation of 3,4-dimethoxycinnamic acid was observed under these conditions. Consistent with these findings, supplementation with either 200mg/L ferulic acid or isoferulic acid resulted in single-step methylation only, with no detectable demethylated product formation. These results indicate that sCOMT efficiently catalyzes mono-O-methylation but does not support sequential methylation leading to dimethoxylated products in vivo.

AtOMT

AtOMT exhibited pronounced regiospecific catalytic activity with a strong preference for meta-position O-methylation of hydroxylated aromatic substrates. When E. coli strains expressing AtOMT were supplemented with 200mg/L caffeic acid, HPLC-MS analysis detected exclusive formation of ferulic acid, with no accumulation of isoferulic acid or demethylated products. Consistent with this specificity, supplementation with 200mg/L ferulic acid did not result in further methylation or production of 3,4-dimethoxycinnamic acid. In contrast, cultures supplemented with 200mg/L isoferulic acid yielded substantial levels of 3,4-dimethoxycinnamic acid, indicating that AtOMT efficiently catalyzes metal-position methylation when the alternate hydroxyl group is already methylated. These findings demonstrate that AtOMT supports sequential O-methylation under defined substrate configurations and uniquely enables dimethoxylated product formation among the COMTs evaluated.

To further evaluate the capacity of AtOMT to catalyze double methylation, a colony screening assay was performed. The highest-producing colonies—specifically colonies #35, #39, and #40—were selected for detailed analysis. Each colony was cultured under identical conditions as previously described to confirm and characterize the formation of the doubly methylated product. The presence of the final product was verified in all selected colonies, supporting the consistent and robust activity of AtOMT in facilitating sequential O-methylation.

Comprehensive analysis of all catechol-O-methyltransferases (COMTs) tested revealed distinct regiospecific methylation behaviors. Specifically, both MtbOMT and sCOMT demonstrated the ability to methylate caffeic acid at either the meta or para hydroxyl positions, with quantitative results showing a preference for meta methylation (62%) over para methylation (38%). In contrast, AtOMT exhibited exclusive activity at the meta position, producing only ferulic acid and non-isoferulic acid under the same assay conditions. These findings are visually summarized in Figure 7, which presents the experimental biosynthetic pathway following the introduction of COMT selectivity. The figure illustrates the regiospecific methylation outcomes for each enzyme, with substrate conversion quantified to highlight the distinct regioselectivity observed for MtbOMT, sCOMT, and AtOMT.

pETM6-EcHpaBC(wt)-RgTAL-sCOMT-AtOMT

Following successful pathway assembly, the five-gene construct was evaluated under multiple conditions to optimize production of the dimethoxylated compound, 3,4-dimethoxycinnamic acid. Initial optimization trials included standard fermentation experiments with supplementation of 200 mg/L tyrosine to assess whether de novo biosynthesis of the target molecule could be achieved. The cultures were induced 4h after inoculation with IPTG. When tyrosine was combined with casamino acids—which naturally contain tyrosine—HPLC-MS analysis confirmed and quantified the formation of 3,4-dimethoxycinnamic acid. To rigorously demonstrate that the product could be synthesized exclusively from supplemented tyrosine, control experiments were conducted in the absence of casamino acids, using only 200 mg/L tyrosine as the precursor. This approach verified that the engineered pathway enabled efficient de novo biosynthesis under defined conditions.

Additional experiments were conducted utilizing leave-one-out control constructs, including pETM6-EcHpaBC(wt)-RgTal(syn), pETM6-EcHpaBC(wt)-RgTal(syn)-sCOMT, and pETM6-EcHpaBC(wt)-RgTal(syn)-AtOMT. These assays were systematically performed under varying conditions: methionine supplementation at 1 g/L and 0 g/L, tyrosine supplementation at 200 mg/L and 300 mg/L, as well as induction at three distinct time points (3 hours, 4 hours, and 5 hours). This comprehensive approach enabled a rigorous evaluation of pathway performance and enzyme-specific activity under defined metabolic and induction parameters.

In summary, this study elucidated the regiospecific activities and substrate preferences of four COMTs within engineered E. coli systems, providing critical insights into how enzyme selection shapes pathway outcomes in microbial metabolic engineering. Through comprehensive screening and quantitative HPLC-MS analyses, it was demonstrated that AtOMT stands out as the only COMT capable of catalyzing sequential O-methylation, thereby enabling the biosynthesis of dimethoxylated phenylpropanoid derivatives such as 3,4-dimethoxycinnamic acid. In contrast, LwOMT10 exhibited negligible catalytic activity, likely due to factors such as poor expression or missing cofactors, highlighting the importance of enzyme compatibility in pathway design. MtbOMT and sCOMT both efficiently catalyzed mono-methylation of caffeic acid—with MtbOMT displaying a preference for meta methylation and sCOMT exclusively producing ferulic acid via meta methylation—but neither enzyme facilitated sequential methylation or dimethoxylated product formation. The successful identification and validation of AtOMT as a robust and versatile regiospecific methyltransferase not only expands the metabolic engineering accessible but also establishes a practical foundation for the sustainable microbial production of valuable aromatic compounds. This work demonstrates the feasibility of engineering microbial platforms for complex phenylpropanoid biosynthesis and paves the way for future efforts aimed at optimizing enzyme performance, enhancing metabolic flux, and scaling up production for industrial applications. Continued research into the kinetic and structural determinants of COMT regiospecificity, as well as the discovery or engineering of improved enzyme variants, will be essential for advancing the field of biosynthetic aromatic compound manufacturing.

1. Bai, H. W., Shim, J. Y., Yu, J., & Bao, T. Z. (2007). Biochemical and molecular modeling studies of the O-methylation of various endogenous and exogenous catechol substrates catalyzed by recombinant human soluble and membrane-bound catechol-O-methyltransferases. Chemical Research in Toxicology, 20(10), 1409–1425. https://doi.org/10.1021/tx700174w

2. Barros, J., Escamilla-Trevino, L., Song, L., Rao, X., Serrani-Yarce, J. C., Palacios, M. D., Engle, N., Choudhury, F. K., Tschaplinski, T. J., Venables, B. J., Mittler, R., & Dixon, R. A. (2019). 4-Coumarate 3-hydroxylase in the lignin biosynthesis pathway is a cytosolic ascorbate peroxidase. Nature Communications, 10(1). https://doi.org/10.1038/s41467-019-10082-7

3. Cao, Y., Chen, Z. J., Jiang, H. di, & Chen, J. Z. (2014). Computational studies of the regioselectivities of COMT-catalyzed meta -/ Para -O methylations of luteolin and quercetin. Journal of Physical Chemistry B, 118(2), 470–481. https://doi.org/10.1021/jp410296s

4. Dawling, S., Roodi, N., Mernaugh, R. L., Wang, X., & Parl, F. F. (n.d.). Catechol-O-Methyltransferase (COMT)-mediated Metabolism of Catechol Estrogens: Comparison of Wild-Type and Variant COMT Isoforms 1. http://aacrjournals.org/cancerres/article-pdf/61/18/6716/2486951/ch1801006716.pdf

5. del Río, J. C., Rencoret, J., Gutiérrez, A., Elder, T., Kim, H., & Ralph, J. (2020). Lignin Monomers from beyond the Canonical Monolignol Biosynthetic Pathway: Another Brick in the Wall. In ACS Sustainable Chemistry and Engineering (Vol. 8, Issue 13, pp. 4997–5012). American Chemical Society. https://doi.org/10.1021/acssuschemeng.0c01109

6. Kopycki, J. G., Stubbs, M. T., Brandt, W., Hagemann, M., Porzel, A., Schmidt, J., Schliemann, W., Zenk, M. H., & Vogt, T. (2008). Functional and structural characterization of a cation-dependent O-methyltransferase from the cyanobacterium Synechocystis sp. strain PCC 6803. Journal of Biological Chemistry, 283(30), 20888–20896. https://doi.org/10.1074/jbc.M801943200

7. Lee, S., Kang, J., & Kim, J. (2019). Structural and biochemical characterization of Rv0187, an O-methyltransferase from Mycobacterium tuberculosis. Scientific Reports, 9(1). https://doi.org/10.1038/s41598-019-44592-7

8. Li, Y., Ding, W., Yin, J., Li, X., Tian, X., Xiao, Z., Wang, F., & Yin, H. (2024). 2,3-Dimethoxycinnamic Acid from a Marine Actinomycete, a Promising Quorum Sensing Inhibitor in Chromobacterium violaceum. Marine Drugs, 22(4). https://doi.org/10.3390/md22040177

9. Lv, H., Zhang, Y., Shao, J., Liu, H., & Wang, Y. (2021). Ferulic acid production by metabolically engineered Escherichia coli. Bioresources and Bioprocessing, 8(1). https://doi.org/10.1186/s40643-021-00423-0

10.  Menke, M. J., Schneider, P., Badenhorst, C. P. S., Kunzendorf, A., Heinz, F., Dörr, M., Hayes, M. A., & Bornscheuer, U. T. (2023). A Universal, Continuous Assay for SAM-dependent Methyltransferases. Angewandte Chemie - International Edition, 62(51). https://doi.org/10.1002/anie.202313912

11.  Pinçon, G., Chabannes, M., Lapierre, C., Pollet, B., Ruel, K., Joseleau, J.-P., Boudet, A. M., & Legrand, M. (2001). Simultaneous Down-Regulation of Caffeic/5-Hydroxy Ferulic Acid-O-Methyltransferase I and Cinnamoyl-Coenzyme A Reductase in the Progeny from a Cross between Tobacco Lines Homozygous for Each Transgene. Consequences for Plant Development and Lignin Synthesis. Plant Physiology, 126(1), 145–155. https://doi.org/10.1104/pp.126.1.145

12.  Shang, H. Q., Bo Yang, Q., Qiang, S., Zheng, R., Zhang, C. Q., Hu, C. Y., Chen, Q. H., & Meng, Y. H. (2025). Engineering Caffeic Acid O-Methyltransferase for Efficient De Novo Ferulic Acid Synthesis. Engineering in Life Sciences, 25(4). https://doi.org/10.1002/elsc.70018

13.  Tsao, D., Liu, S., & Dokholyan, N. v. (2011). Regioselectivity of catechol O-methyltransferase confers enhancement of catalytic activity. Chemical Physics Letters, 506(4–6), 135–138. https://doi.org/10.1016/j.cplett.2011.03.048

14.  Wang, J., Liao, N., Liu, G., Li, Y., Xu, F., & Shi, J. (2024). Diversity and regioselectivity of O-methyltransferases catalyzing the formation of O-methylated flavonoids. In Critical Reviews in Biotechnology (Vol. 44, Issue 6, pp. 1203–1225). Taylor and Francis Ltd. https://doi.org/10.1080/07388551.2023.2280755

15.  Wickramasinghe, P. C. K., & Munafo, J. P. (2020). Fermentation Dynamics and Benzylic Derivative Production in Ischnoderma resinosum Isolates. ACS Omega, 5(35), 22268–22277. https://doi.org/10.1021/acsomega.0c02550

16.  Zubieta, C., Kota, P., Ferrer, J.-L., Dixon, R. A., & Noel, J. P. (2002). Structural Basis for the Modulation of Lignin Monomer Methylation by Caffeic Acid/5-Hydroxyferulic Acid 3/5-O-Methyltransferase. The Plant Cell, 14(6), 1265–1277. https://doi.org/10.1105/tpc.001412

As a Biochemistry major studying in a research lab, I have had a lot of opportunities to demonstrate leadership and gain further confidence in critical thinking skills. I have also had many interactions with peers, which leads to further development of my ability to work in a team to achieve a common goal. Overall, I have gained many NACE competencies that I will use within a workplace after Miami in order to be successful.