Please cite this paper as:

Yang, L. and Palsson, B. 2023. Biomanufacturing of melatonin is now possible. Melatonin Research. 6, 1 (Feb. 2023), 72-78. DOI:https://doi.org/https://doi.org/10.32794/mr112500142.


Commentary

 

Biomanufacturing of melatonin is now possible

 Lei Yang1*, Bernhard O Palsson1,2* 

1Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Kemitorvet, Building 220, 2800 Kgs. Lyngby, Denmark

2Department of Bioengineering, University of California, San Diego, La Jolla, CA 92093, USA  

*Correspondence: leiya@biosustain.dtu.dk or palsson@ucsd.edu, Tel: +45 3196 4881

Running title: Biomanufactured melatonin

Received: January 5, 2023; Accepted: February 11, 2023

 

ABSTRACT

     Currently, the majority of the melatonin manufactured is chemically synthesized replacing the previous method that extracted it from biological materials. Due to environmental concerns, biomanufacturing of chemical-related products is now gaining favor over chemical synthesis. Melatonin is synthesized from tryptophan in a four-step pathway that includes a hydroxylase and methyl transferase, both representing major challenges for metabolic engineering. Using novel technologies, including genome editing and laboratory evolution, the metabolic pathway of melatonin synthesis has been reconstructed in a microbial host which has the capacity to produce melatonin in gram-level quantities.  

Key words: Melatonin, 5-hydroxytryptophan, serotonin, biomanufacture, biosynthesis, metabolic pathway, fermentation, scale-up.

 ____________________________________________________________________________________________________

 

1.      INTRODUCTION

     Melatonin is an indoleamine naturally produced by the pineal gland in vertebrates including humans to regulate the circadian rhythm. It is best known as a sleep aid, but its applications have been expanded to other areas including beverages (1, 2), cosmetics (3), horticulture (4), AgBios (5), and antioxidant and anti-aging (6, 7). Since melatonin exhibits antiviral activity which can be used for COVID-19 treatment, this drove up its sale in the US market to reach $1.26 billion in 2021 (8). This tendency will likely continue, judging from the increased awareness of its health benefits (9–12). The rapidly increasing melatonin demand requires an environmentally-friendly process to replace the chemically synthesized melatonin which inevitably generates some environmental biohazards (4). Therefore, the biomanufacturing of melatonin using microbial cell factories and renewable feedstocks will eventually become the dominant process to synthesize melatonin.


2.      MELATONIN PATHWAY ENGINEERING

     Melatonin is biosynthesized from tryptophan by four consecutive steps: 1) hydroxylation of tryptophan to form 5-hydroxytryptophan (5-HTP), 2) decarboxylation to produce 5-hydroxytryptamine (5-HT, or serotonin), 3) acetylation to form acetyl-5-hydroxytryptamine (AcHT), and 4) methylation to generate melatonin (Figure 1). Achieving functional expression of these enzymes and maintaining a supply of the substrates in microbial hosts are among the current challenges.

     Microbial production of the intermediates of the melatonin pathway has been reported, including 5-HTP and serotonin (13–15). Production of melatonin using microbial cell factories was first demonstrated in Saccharomyces cerevisiae, reporting a titer of 14 mg/L use  glucose as the initial substrate (16). Later, an Escherichia coli melatonin  production host was constructed with the melatonin yield up to 1g/L in fed batch fermentation co-feed with  glucose and tryptophan (17). In this study, Hao et al. reconstituted the melatonin synthetic pathway in E. coli using heterologous enzymes selected from various origins. The Ddc and Aanat are promiscuous enzymes which can result in adverse effects on other metabolic processes and generate unwanted toxic metabolic products. It is thus necessary  to fine-tune their expression levels to limit the accumulation of unwanted byproducts while maintaining high metabolic fluxes through the targeted reactions (Figure 1).

     Further engineering of a host genome, such as deleting tnaA gene encoding tryptophanase that catalyzes the degradation of both tryptophan and 5-HTP, is also necessary to enable the melatonin synthetic pathway to function at high generating rate. Knocking out the trpR (tryptophan transcription repressor) gene also enhanced melatonin production by the derepression of genes involved in tryptophan synthesis and uptake. Numerous other adjustments are needed to fine-tune the gene expression of melatonin production hosts (17).

     Among the four steps of melatonin biosynthesis, hydroxylation of tryptophan to form 5-HTP and methylation of acetyl-serotonin to form melatonin are difficult to manipulate. These two enzymes do not function well in E. coli and require metabolic cofactors (see Figure 1) that have their own complex synthesis and regeneration machinery. Here we briefly describe the novel strategies used to overcome these bottlenecks (18, 19).

2.1. Tryptophan hydroxylation.

     The first step of melatonin synthesis is the production of 5-HTP through hydroxylation of tryptophan, using tetrahydrobiopterin (BH4), Fe2+ and O2 as cofactors. There are two major engineering challenges with this reaction: the efficiency of hydroxylase and cofactor tetrahydrobiopterin (BH4) regeneration.

     Hao et. al have expressed human TpH in E. coli with an N-terminal truncation, which is necessary for the solubility of this enzyme (19). Lin et. al. have alternatively used prokaryotic phenylalanine 4-hydroxylases, which is better expressed in E. coli, but has a preference of phenylalanine as the substrate. They altered its substrate preference to better utilize tryptophan through protein engineering guided by rational design (13).

     BH4 is the cofactor for TpH to synthesize 5-HTP. It is derived from guanosine triphosphate (GTP). Wang et. al. have reconstituted the 5-step human BH4 synthesis and regeneration system in E. coli, which enabled production of high level (1.2g/L) 5-HTP in fed batch fermentation (14). This was the highest 5-HTP titer reported using microbial cell factories. Hao et. al have designed a novel strategy using growth-coupling and adaptive laboratory evolution (ALE) to repurpose E. coli native pterin as an efficient cofactor for TpH. They also obtained beneficial mutations that enhanced TpH abundance, allowing for considerably higher production of 5-HTP (19).

2.2. SAM-dependent methylation.  

     Another bottleneck in the melatonin biosynthesis pathway is the methylation of acetyl-serotonin to form melatonin. This reaction is catalyzed by methyltransferase (Mtase) using S-adenosylmethionine (SAM) as a cofactor, one of the most abundant cofactors used in metabolism across different kingdoms. SAM donates a methyl group to the substrate and gets regenerated in the SAM cycle from methionine. Both SAM availability and Mtase efficiency are critical but prove difficult to metabolically engineer.

     In a study examining the production of a different methylated product, vanillin, Kunjapur et al. have improved SAM synthesis and regeneration by deleting metJ, a gene encoding a repressor of methionine synthetic  genes as well as  overexpressing the feedback resistant variants of two key enzymes in SAM biosynthesis (20). Another strategy they employed was to overexpress two enzymes, luxS and mtn, which are for SAM regeneration. This strategy also results in enhanced SAM availability and vanillin titer. All these approaches are potentially beneficial for melatonin production. It is worth noting that GTP is involved in both SAM and pterin biosynthesis and transformation (Figure 1), thus, optimizing pterin-dependent hydroxylation and SAM-dependent methylation in the melatonin biosynthesis pathway can be much more challenging due to the potential limitation of GTP or other related reactions.

     In order to enhance methyltransferase enzyme activities, Luo et al. designed a growth-coupling system by rewiring the SAM cycle to cysteine synthesis (18). First, they deleted the cysE gene to make E. coli a cysteine auxotroph. Second, they added two genes from yeast to enable the synthesis of cysteine from homocysteine, which is an intermediate of the SAM cycle. The resulting strain requires a high flux of SAM-dependent methylation to provide sufficient cysteine to grow. Improved variants of the Mtases or upstream enzymes can be obtained from ALE of the growth-coupled strains. This method provided a fast and efficient workflow for improving the biosynthesis of melatonin and other methylated products.

2.3. Transporter engineering.

     Membrane transporters play an important role in achieving a high-performance production strain. Enrichment of metabolic precursor supply is favorable for the downstream reactions, while high intracellular concentration of the product may inhibit enzymes or cellular growth. It is beneficial to identify and engineer membrane transporters to prevent the leakage of precursors or intermediates, as well as increase the efflux of the product.

     In a process where tryptophan is fed as the precursor, deleting the known tryptophan transporter gene yddG increases melatonin titer significantly (Figure 1) (17). YddG plays a major role in export of aromatics amino acids including tryptophan. Deleting it can help increase the intracellular tryptophan concentration and improve melatonin production.  Melatonin inhibits the growth of E. coli (21), so it is beneficial to increase strain tolerance to melatonin. However, it is generally difficult to identify transporters for xenobiotic chemicals in microbial hosts. Several studies have succeeded in addressing this difficulty, using methods like adaptive evolution or screening metagenomics libraries (22–25). 

     Yang et al. have developed a high throughput workflow in which they screened  an E. coli single gene knockout library containing more than 400 predicted membrane transporters in the presence of a high concentration of melatonin (26). In total, five knockout strains showed growth defects in melatonin, indicating that these five transporters play a role in exporting melatonin. One of the transporters, YhjV, was annotated as a putative transporter. Its function was previously uncharacterized. They further over-expressed genes encoding the five transporters and found that yhjV overexpression increased melatonin titer by 27% in E. coli. Enhancing production strain tolerance to melatonin is important for developing an efficient process to commercial-relevant scales.

     A practical metabolic production strain was generated after meeting all these metabolic engineering requirements that included over 10 genomic edits (26).

Figure-3.jpg

Fig. 1. Melatonin biosynthetic pathway.

     Melatonin can be synthesized from tryptophan in four steps by four enzymes: hydroxylation by tryptophan hydroxylase (TpH), decarboxylation by decarboxylase (Ddc), acetylation by aralkylamine acetyltransferease (Aanat), and methylation by N-Acetylserotonin O-methyltransferase(Asmt). Hydroxylation requires either tetrahydrobiopterin or tetrahydromonopterin (both being GTP derivatives, indicated as PtH4) as a cofactor. The last reaction methylation is dependent on S-adenosylmethionine (SAM) as a cofactor, which is mostly regenerated in the SAM cycle. Intermediates of the SAM cycle include S-adenosyl-L-homocysteine (SAH), S-ribosyl-L-homosystein (SRH), homocysteine (HCY), and methionine (MET). The enzymes involved in the SAM cycle include SAH nucleosidase (mtn), SAH lyase (luxS), methionine synthase (metE or metH) and methionine adenosyltransferase (metK). The SAM cycle is also linked to GTP through tetrhydrofolate synthesis and transformation. Transporter engineering is also beneficial for melatonin production, e.g., knockout yddG and overexpression yhjV, which encode transporters responsible for efflux of tryptophan and melatonin, respectively.


3.      CHALLENGES IN SCALING-UP MELATONIN PRODUCTION

     To biomanufacture melatonin at commercial scales, a robust production strain with high titer, yield, or productivity is needed. One critical challenge during scale-up is product inhibition. On one hand, melatonin is an antioxidant agent playing an important role in the evolution of aerobic metabolism (27), which might be beneficial in fermentation in oxygen-rich conditions (28). On the other hand, melatonin exerts a toxic effect on E. coli, especially at high concentrations. Interestingly, it appears to repel E. coli, as it upregulates the chemotaxis machinery, causing E. coli to swim away from high melatonin concentrations (21).

     It is thus essential to understand the mechanisms of melatonin inhibition in order to develop robust production strains that can perform well in large-scale fermentation. Gulcin et. al. have reported that melatonin can effectively chelate metal ions including zinc, copper, and iron as a way of reducing lipid peroxidation in human cells (29). However, the ion chelating effect can be detrimental for the production strain of microbes, causing poor fermentation performance as melatonin accumulates in the fermenter. Moreover, melatonin was reported to inhibit citrate synthase in the TCA cycle in gram negative bacteria, which could also lead to growth arrest of the production strain (30). Finally, melatonin has a low solubility (2g/L) in water (Scientific committee on consumer safety, opinion on melatonin, 23 March 2010). A commercially feasible titer should exceed 2g/L. In that case, melatonin will most likely precipitate and form insoluble particles in the fermentation broth, which might also decrease strain performance in the reactors as melatonin accumulates (31). Improving strain tolerance towards these stresses by genome engineering can help increase fermentation performance. Alternatively, removing melatonin during fermentation using methods like liquid extraction is another option to reduce product inhibition (32).

     In conclusion, the challenges of making melatonin through a bioprocess are great. The introduction of four heterologous enzymes and multiple genome edits were needed to get the pathway to function in a microbial host. Thus, the practical biomanufacturing of melatonin is now possible. Although the current production strain can go into a biomanufacturing process, the deployment of additional advanced biological tools such as transcriptomics, metabolomics, and integrated multi-omics data analysis are likely to continuously improve the productive strain and the fermentation process. Biomanufactured melatonin is thus on its way into myriad products that are anticipated to appear in the coming years.

 

ACKNOWLEDGMENTS

     The authors would like to thank the Novo Nordisk Foundation for financial support (grants NNF20CC0035580 and NNF10CC1016517). We would also like to thank Dr. Andreas Worberg and Pre-Pilot Plant team, including Dr. Ann Dorrit Enevoldsen, Dr. Milica Randelovic, Dr. Minkle Jain and Dr. Gossa Garedew Wordofa for their efforts and discussion on the scale-up and downstream process development. We also appreciate the comments from Dr. Emre Özdemir and Dr. Se Hyeuk Kim on this manuscript.

 

AUTHORSHIP

      LY and BOP both wrote the paper.


CONFLICT OF INTERESTS

      L. Yang is an inventor of the following patent application: WO/2020/187739. 2020. B.O. Palsson sits on the Board of Directors of Conarium BioWorks.


REFERENCES

 

  1. Cheng G, et al. (2021) Plant-derived melatonin from food: a gift of nature. Food Funct. 12: 2829–2849.

  2. Meng J-F, T-C Shi, S Song, Z-W Zhang, Y-L Fang (2017) Melatonin in grapes and grape-related foodstuffs: A review. Food Chem. 231: 185–191.

  3. Rusanova I, et al. (2019) Protective effects of melatonin on the skin: Future perspectives. Int. J. Mol. Sci. 20: 4948.

  4. Gao T, et al. (2022) Introducing melatonin to the horticultural industry: physiological roles, potential applications, and challenges. Hortic. Res. 9: uhac094.

  5. nawaz k, et al. (2020) melatonin as master regulator in plant growth, development and stress alleviator for sustainable agricultural production: current status and future perspectives. Sustainability 13: 294.

  6. stacchiotti a, g favero, lf rodella (2020) impact of melatonin on skeletal muscle and exercise. Cells 9: 288.

  7. Gunata M, H Parlakpinar, HA Acet (2020) Melatonin: A review of its potential functions and effects on neurological diseases. Rev. Neurol. (Paris) 176: 148–165.

  8. (2022) Data from “Melatonin Market Size And Forecast.” https://www.verifiedmarketresearch.com/.

  9. Brusco LI, et al. (2021) Efficacy of melatonin in non-intensive care unit patients with COVID-19 pneumonia and sleep dysregulation. Melatonin Res. 4: 173–188.

  10. Gurunathan S, M-H Kang, Y Choi, RJ Reiter, J-H Kim (2021) Melatonin: A potential therapeutic agent against COVID-19. Melatonin Res. 4: 30–69.

  11. Juhnevica-Radenkova K, DA Moreno, L Ikase, I Drudze, V Radenkovs (2020)  Naturally occurring melatonin: Sources and possible ways of its biosynthesis. Compr. Rev. Food Sci. Food Saf. 19: 4008–4030.

  12. (2022) Data from “Melatonin Market by Product and Geography - Forecast and Analysis 2022-2026.” https://www.technavio.com/report/.

  13. Lin Y, X Sun, Q Yuan, Y Yan (2014) Engineering bacterial phenylalanine 4-hydroxylase for microbial synthesis of human neurotransmitter precursor 5-hydroxytryptophan. ACS Synth. Biol. 3: 497–505.

  14. Wang H, et al. (2018) Metabolic pathway engineering for high-level production of 5-hydroxytryptophan in Escherichia coli. Metab. Eng. 48: 279–287.

  15. Mora-Villalobos J-A, A-P Zeng (2018) Synthetic pathways and processes for effective production of 5-hydroxytryptophan and serotonin from glucose in Escherichia coli. J. Biol. Eng. 12: 3.

  16. Germann SM, et al. (2016) Glucose‐based microbial production of the hormone melatonin in yeast Saccharomyces cerevisiae. Biotechnol. J. 11: 717–724.

  17. Luo H, et al. (2020) Microbial synthesis of human-hormone melatonin at gram scales. ACS Synth. Biol. 9, 1240–1245.

  18. Luo H, et al. (2019) Coupling S-adenosylmethionine–dependent methylation to growth: Design and uses. PLOS Biol. 17: e2007050.

  19. Luo H, et al. (2020) Directed Metabolic Pathway Evolution Enables Functional Pterin-Dependent Aromatic-Amino-Acid Hydroxylation in Escherichia coli. ACS Synth. Biol. 9: 494–499.

  20. Kunjapur AM, JC Hyun, KLJ Prather (2016) Deregulation of S-adenosylmethionine biosynthesis and regeneration improves methylation in the E. coli de novo vanillin biosynthesis pathway. Microb. Cell Factories 15: 61.

  21. Lopes JG, V Sourjik (2018) Chemotaxis of Escherichia coli to major hormones and polyamines present in human gut. ISME J. 12: 2736–2747.

  22. Pereira R, et al. (2019) Adaptive laboratory evolution of tolerance to dicarboxylic acids in Saccharomyces cerevisiae. Metab. Eng. 56: 130–141.

  23. Malla S, et al. (2022) A novel efficient l-lysine exporter identified by functional metagenomics. Front. Microbiol. 13: 855736.

  24. van der Hoek SA, I Borodina (2020) Transporter engineering in microbial cell factories: the ins, the outs, and the in-betweens. Curr. Opin. Biotechnol. 66: 186–194.

  25. Genee HJ, et al. (2016) Functional mining of transporters using synthetic selections. Nat. Chem. Biol. 12: 1015–1022.

  26. yang l, et al. (2022) identification and engineering of transporters for efficient melatonin production in Escherichia coli. Front. Microbiol. 13: 880847.

  27. zhao d, et al. (2019) melatonin synthesis and function: evolutionary history in animals and plants. Front. Endocrinol. 10: 249.

  28. O’Donnell A, Y Bai, Z Bai, B McNeil, LM Harvey (2007) Introduction to bioreactors of shake-flask inocula leads to development of oxidative stress in Aspergillus niger. Biotechnol. Lett. 29: 895–900.

  29. Gulcin İ, ME Buyukokuroglu, OI Kufrevioglu (2003) Metal chelating and hydrogen peroxide scavenging effects of melatonin: Metal chelating and H2O2 scavenging of melatonin. J. Pineal Res. 34: 278–281.

  30. He F, et al. (2022) Melatonin inhibits Gram-negative pathogens by targeting citrate synthase. Sci. China Life Sci. 65: 1430–1444.

  31. Moreno AD, C González-Fernández, M Ballesteros, E Tomás-Pejó (2019) Insoluble solids at high concentrations repress yeast’s response against stress and increase intracellular ROS levels. Sci. Rep. 9: 12236.

  32. Zautsen RRM, et al. (2009) Liquid-liquid extraction of fermentation inhibiting compounds in lignocellulose hydrolysate. Biotechnol. Bioeng. 102: 1354–1360.

​CCBY.png

This work is licensed under a Creative Commons Attribution 4.0 International License