Spotlight 1: Development of a Tandem ACP Polyketide Synthase

Access the article here:

Polyketides are a diverse class of molecule composed of multiple ketide groups (adjacent methylene and carbonyl groups) or their derivatives and are biologically synthesized via polyketide synthases, which are derived from polyunsaturated fatty acid synthases (Staunton and Weissman 2001). In fatty acid synthesis, addition of acetyl groups to the growing fatty acid is followed by full reduction of the acetyl carbonyl to an alkyl group. In contrast, during polyketide synthesis carbonyls can be fully reduced, partially reduced to a hydroxy group, or not reduced at all, which results in a huge variety of polyketides. Polyketide synthases frequently contain more than one “module” that add and modify acetyl monomers, and multiple polyketide synthases often work in conjuction to create large polyketides. Intended as chemical weapons against other organisms by the organisms that produce them, many polyketides have found applications as therapeutic drugs, including antibiotic, antifungal, and anti-cancer drugs. Fatty acid synthases and polyketide synthases are composed of several enzymes that perform the various reactions involved in the biosynthesis. One of these is an acyl carrier protein (ACP) that carries the growing molecule from one enzymatic active site to the next. Most Type I polyketide synthases and fatty acid synthases have just one ACP. However, some fatty acid synthases have several ACP domains (together called a tandem ACP domain), which has been correlated with increased production of fatty acids (Jiang et al. 2008). This indicates that one of the rate-limiting factors of polyketide biosynthesis is the availability of substrates made readily accessible by being bound to ACP domains.

Research into polyketides is of great interest but is hampered by the low levels at which they tend to be produced by organisms, inhibiting the ability of researchers to obtain the large quantities necessary for their work. Wang et al. (2018) approached this problem by developing a method to create polyketide synthases that have increased production capabilities by integrating multiple ACP domains into a single polyketide synthase. The authors expressed both module 6-TE, a simple polyketide synthase developed from the well-characterized 6-deoxyerythronolide B synthase (DEBS), responsible for synthesizing the macrolide ring of erythromycin, and moldule 6-3ACP-TE, a version of module 6-TE containing three ACP domains rather than one. Liquid chromatography-mass spectrometry was used to assess the polyketide production capacities of the isolated module 6-TE and module 6-3ACP-TE proteins.

Polyketide production after 7 hours by native polyketide synthase DEBS module 6-TE and the engineered DEBS module 6-3ACP-TE showing the approximate 2.5-fold increase in polyketide biosynthesis gained by increasing the number of acyl carrier proteins present by two. The figure also displays the lack of triketide lactone production by module 6-3ACP-TE, the cause of which the authors did not determine. Understanding the reason behind this and utilizing it could lead to greater specificity in creating potential therapeutics.

Module 6-TE synthesizes both triketide lactone and triketide ketolactone under normal conditions. The creation of triketide lactone is a result of a ketoreductase domain in the synthase that uses NADPH as a cofactor to reduce one of the carbonyls to a hydroxl group. The major significant finding of this study is that Wang et al. determined that their incorporation of two additional ACP domains increased polyketide production by a factor of about 2.5. This finding provides the opportunity to overcome a major challenge in the efforts to create polyketide-based therapeutics, which could open the door to a faster development process.

Avenues for further work are determining the impact of larger tandem ACP domains on polyketide biosynthesis rates and developing more complex multi-module polyketide synthases to experiment with increasing the production of more complex and therapeutically interesting polketides. Future applications of engineered polyketide synthases are hinted at by the most surprising result discovered by the author: while module 6-TE produced a mixture of triketide lactone and triketide ketolactone, module 6-3ACP-TE solely synthesized triketide ketolactone despite the presence of NADPH. As of yet the authors have not determined an explanation for this result. This is an avenue for further exploration, since determining what caused the specificity in products could allow for control of the mechanism governing it. Being able to increase the rate of polyketide production is not as beneficial as it could be if increasing the rate of biosynthesis also decreases the diversity of compounds that can be produced. Figuring out issues such as why module 6-3ACP-TE does not display any ketoreductase activity would not only restore the benefits of increased production rate, but would also provide the benefit of greater control over the polyketides being synthesized.

Natural products or their derivatives are often developed into therapeutic drugs. However, creating total syntheses, which are complete syntheses from commercially available precursors, for natural products is a notorious challenge. For instance, the synthesis of erythromycin, a widely used polyketide antibiotic that is on the World Health Organization’s list of the most safe and effective medicines necessary for a basic health-care system, was worked on for more than a decade by over a dozen different research groups (“WHO Model List of Essential Medicines,” Mulzer 1991). A popular potential alternative to performing difficult laboratory syntheses is biosynthesis using bacteria, where bacteria are engineered to produce large quantities of pharmaceutical drugs. The work presented by Wang and colleagues is a step toward the large-scale aspect of commercially biosynthesizing complex drugs. Work on why module 6-3ACP-TE exclusively produces triketide ketolactone would contribute to the capacity to readily produce complex drug candidates that differ at specific locations in order to develop compounds with the desired biological properties.


Jiang, H., Zirkle, R., Metz, J. G., Braun, L., Richter, L., Van Lanen, S. G., and Shen, B. 2008. “The role of tandem acyl carrier protein domains in polyunsaturated fatty acid biosynthesis.” J. Am. Chem. Soc. 130, 6336-6337.

Khosla, C., Tang, Y., Chen, A. Y., Scnarr, N. A., and Cane, D. E. 2007. “Structure and mechanism of the 6-deoxyerythronolide B synthase.” Annu. Rev. Biochem. 76, 195-221.

Mulzer, J. 1991. “Erythromycin synthesis—a never-ending story?” Angew. Chem. 30, 1452-1454.

Staunton, J., and Weissman, K. J. 2001. “Polyketide biosynthesis: a millennium review.” Nat. Prod. Rep. 18, 380-416.

Wang, Z., Bagde, S. R., Zavala, G., Matsui, T., Chen, X., and Kim, C. 2018. “De novo design and implementation of a tandem acyl carrier protein domain in a type I modular polyketide synthase.” ACS Chem. Biol. 13, 3072-3077.

WHO Model List of Essential Medicines, 20th List. 2017. World Health Organization.

7 Replies to “Spotlight 1: Development of a Tandem ACP Polyketide Synthase”

  1. The authors specifically state they selected DEBS3, module 6: “we chose this PKS because (i) its small size allows for facile genetic manipulation and (ii) the presence of the TE domain makes direct product detection possible” (3073). I know that they said in their conclusion a possible solution to regenerate KR’s activity is to change the ACP domain location, however I wonder if maybe targeting a different module or even protein (e.g. DEBS3, module 5) would make a difference. I know they are concerned about product detection but wouldn’t the product be detected later on after it is released from module 6? It sounds like it may be difficult to move the tandem ACP domain to upstream of KR.

    1. Absolutely, targeting a different module or protein could made a difference. My initial thought on why they’re sticking with module 6-TE is that they’ve already invested work into it and trying a different module or protein would be like starting over. Additionally, in the first methods subsection, the authors say that the plasmids they used for expression were pre-made & given to them by Dr. Chaitan Khosla. Based on this, another reason for staying with module 6-TE could be that switching to something else would also necessitate doing the groundwork to made new plasmids. I think the product should be detectable as long as they have the thioesterase (TE) domain in the module.

  2. Very interesting to say the least, and extremely useful medically speaking if the production of polyketides is brought up to a massive scale. The problem where 6-3ACP-TE only produces triketide ketolactone is challenging; however, as stated polyketide synthesis can the fully, partially, or not perform a reduction of the carbonyl group. So the limited product results may be from an inability to achieve a partial reduction between the three active sites. Also, if the rate-limiting step is the availability of ACP domains could forcing an organism into overproduction of polyketide synthesis also achieve the increase in production factors. Finding the genes that code for the protein unit(s) of polyketide synthesis, or exposure to toxins/ hostiles could be viable alternatives to creating a new enzyme with multiple domains.

    1. Inducing overexpression could increase the production of polyketides, but I think it may only increase it to a certain point. The increase would be helpful in obtaining greater quantities for research, but as I understand it, engineering polyketide synthases with multiple ACP domains could enable polyketide production to even go beyond that increase. As for finding the genes for different polyketide synthases, I agree that it could be a viable alternative. However, I think that creating new multidomain enzymes would enable researchers to develop highly specific and completely new polyketides rather than being constrained to those produced by naturally evolved proteins. Additionally, developing this technology may make it easier to simply custom design polyketide synthases to produce the desired molecules rather than trying to find previously existing proteins.

  3. Very interesting paper and very clear review! I’m surprised that the way to increase the production rate of polyketide is adding three ACP to the synthases, a simple method but creative thought. I have a question that may be dumb, but why did they add three ACP domains to the synthases rather than other numbers of ACP? If three ACP-domains means more ACP, more productive, why didn’t they add 4 or 5 domians?

    1. I agree, I think this solution is surprisingly simple, which makes it interesting in itself. I had not thought to ask your question, I just went along with the authors deciding to use three ACP domains. I suspect the number of domains used had to do with making the research easier on themselves. Each domain added would involve more work and potentially decrease the chances of success at such an early stage in the overall research, while two domains may not have been enough to clearly demonstrate improved activity.

  4. The article says that “in nature, polyketides confer competitive advantage to the host organism such as antimicrobial activity and pigmentation.” Could polyketide activity also be a source for the development of antibiotic resistance mechanisms? Additionally, how else could or has tandem ACP engineering been used? The authros conclude by saying that hypothetically creating an artificial tandem ACP containing more than three ACPs could lead to increased product turnover. It would be interesting to know how this research continues and the progress they make in modifying this mechanism.

Leave a Reply

Your email address will not be published. Required fields are marked *