Yeast cell factories on the horizon

See allHide authors and affiliations

Science  04 Sep 2015:
Vol. 349, Issue 6252, pp. 1050-1051
DOI: 10.1126/science.aad2081

For thousands of years, yeast has been used for making beer, bread, and wine. In modern times, it has become a commercial workhorse for producing fuels, chemicals, and pharmaceuticals such as insulin, human serum albumin, and vaccines against hepatitis virus and human papillomavirus. Yeast has also been engineered to make chemicals at industrial scale (e.g., succinic acid, lactic acid, resveratrol) and advanced biofuels (e.g., isobutanol) (1). On page 1095 of this issue, Galanie et al. (2) demonstrate that yeast can now be engineered to produce opioids (2), a major class of compounds used for treating severe pain. Their study represents a tour de force in the metabolic engineering of yeast, as it involved the expression of genes for more than 20 enzymatic activities from plants, mammals, bacteria, and yeast itself. It clearly represents a breakthrough advance for making complex natural products in a controlled and sustainable way.

Yeast has previously been recruited for producing complex natural products through the reconstruction of biosynthetic pathways taken from plants or animals (3). This is done by transferring genes that encode enzymes of the metabolic pathway, from the organism that naturally produces the chemical of interest, into yeast. A key requirement is that this heterologous metabolic pathway hooks up to endogenous yeast metabolism. As central metabolism is highly conserved between different organisms, it is generally possible to identify a metabolite in the endogenous yeast metabolism that can be used as a precursor for the heterologous pathway. Indeed, using this approach, yeast was engineered to produce hydrocortisone (a steroid used to synthesize drugs with anti-inflammatory and antiproliferative effects) by recruiting part of the endogenous yeast pathway that generates ergosterol (4). Several endogenous yeast enzymes were put to use, but it was still necessary to express more than 10 mammalian enzymes in combination with overexpressing the yeast genes to enable production of hydrocortisone from glucose. Ergosterol is synthesized via the so-called mevalonate pathway (also used for cholesterol biosynthesis in mammals), and plants use this pathway to make a wide range of isoprenoids (or terpenoids). Members of this family have a broad spectrum of applications, including food products, pharmaceuticals, cosmetics, and biofuels (3), and through recruitment of the mevalonate pathway, it has been possible to engineer yeast to produce compounds normally extracted from plants [such as perfumes (e.g., santalene) (5)]. Several companies are currently working on developing this technology to enable the production of fine chemicals through yeast fermentation, thereby providing a stable, sustainable, and scalable source of the desired compound.

Versatile yeast.

Reconstructing heterologous biosynthesis pathways enables the recruitment of yeast endogenous metabolism to produce complex natural products. By using the mevalonate pathway (Mev), a platform yeast strain can be engineered to synthesize different isoprenoids and sterols; through recruitment of an aromatic amino acid biosynthesis pathway (Aro), such as for tyrosine, a different platform yeast strain can be engineered to produce flavonoids, stilbenes, and opioids.


The biosynthesis of perfumes typically requires the expression of only one to three plant enzymes in yeast. However, it is possible to express a multistep heterologous pathway from plants to produce artemisinic acid, which can be chemically converted to the antimalarial drug artemisinin (6). Again, this relied on the endogenous mevalonate pathway in yeast, and through combined engineering of this pathway and optimization of the heterologous pathway, productivity was improved to the point where it could be taken forward for commercial production of artemisinin (7, 8). This process currently provides up to one-third of the global need for this antimalarial drug. Even though artemisinin can be extracted from plants, the biotech-based production ensures stable production, which is particularly important as the supply has suffered from large fluctuations in the past.

A general observation from past studies is that to ensure sufficient production of the natural products, it was necessary to boost the endogenous yeast pathway for efficient provision of precursors, such as those from the mevalonate pathway. Galanie et al. also relied on up-regulating the endogenous yeast pathway for providing tyrosine, the precursor for opioid biosynthesis. It is likely that the successful production of opioids from glucose rests on the initial generation of “platform” yeast strains—one platform strain for efficient production of tyrosine, and a second platform strain to efficiently produce reticuline, an intermediate in the opiod biosynthetic pathway. A general lesson therefore seems to be that successful metabolic engineering of yeast for producing complex natural products requires platform strains (9). Moreover, a single platform strain can be used to make a range of different natural products (see the figure).

The progress made by Galanie et al. represents an important milestone in metabolic engineering of long and complex biosynthetic pathways in yeast. Even though the titer and productivities are low, the result may ultimately lead to commercial production of opioids. Controlled production of these chemicals also will allow for contained production, with the goal of completely eliminating traditional extraction from plants. This could reduce illicit production. The question of scaling up yeast-based production of opioids is now at hand, but getting there may not be so trivial. With the current output, it would take 4400 gallons of bioengineered yeast to produce a single therapeutic dose for pain relief (10). Scaling up the production of artemisinin, for example, took more than 5 years of investigation and required investments exceeding $50 million. That is not to say that scaling up opioid production in yeast is not inevitable; therefore, the advance by Galanie et al. raises the important question of regulating access to such yeast strains.

The study of Galanie et al. also demonstrates how reconstruction of a plant pathway in yeast allows for enzyme discovery and screening for efficient enzymes for biosynthesis, which is of value for any metabolic engineering project. For opioid synthesis, for example, the authors identified one new enzyme required for the biosynthesis and screened more than 20 variants of a key enzyme of the pathway to increase the flux toward opioids. By demonstrating that a 23-step biosynthetic pathway can be reconstructed in yeast, the study advances the ability to ensure sustainable production of fuels, chemicals, and pharmaceuticals. Even though the 2010 market for renewable chemicals was only about 1% of the total chemical market (about $30 billion out of a total market of $3 trillion), metabolic engineering of cell factories, and especially yeast, has clear potential to increase this proportion in the future.


View Abstract

Navigate This Article