Towards biobased ethyl acetate and beyond : Identification and application of the elusive Eat1 enzyme

Ester production by microorganisms is of great importance to biotechnological processes. These volatile compounds impart pleasant, fruity aromas to beer, wine, and other fermented products. Yeast in particular are well known for their ability to produce volatile esters. The genetic and enzymatic intricacies of ester production have been the focus of many studies, mainly in <em>S. cerevisiae</em>. Despite this, ester synthesis in yeast is not fully understood. This holds particularly true for yeasts that produce high amounts of ethyl acetate from sugars, such as <em>Kluyveromyces marxianus</em> and <em>Wickerhamomyces</em> <em>anomalus</em>. As was introduced in <strong>Chapter 1</strong>, this ability has been described more than a century ago, but the enzymatic mechanisms behind the synthesis were unclear. Circumstantial evidence suggested that an alcohol acetyltransferase (AAT) was responsible for the ethyl acetate formation. Ethyl acetate is a versatile commodity chemical that is currently produced in unsustainable processes. Understanding the enzymes responsible for bulk ethyl acetate synthesis in yeast could enable rational design of novel production strains. This, in turn could facilitate the development of biobased ethyl acetate production processes. This thesis has made several important breakthroughs in the field of ester production in yeast, particularly ethyl acetate. The most significant of these was the discovery of the elusive enzyme responsible for bulk ethyl acetate synthesis in yeast. The identification of the ethyl acetate-producing enzyme in <em>W. anomalus </em>is described in <strong>Chapter 2</strong>. The purified enzyme showed AAT activity with ethanol and acetyl-CoA and was therefore named Ethanol acetyltransferase 1 (Eat1). Production of Eat1 in <em>Escherichia coli</em> enabled efficient ethyl acetate production in <em>E. coli</em>. The enzyme could thus potentially be used to develop new biobased ethyl acetate production processes. However, Eat1 was also able to function as a thioesterase and esterase active against acetyl-CoA and ethyl acetate, respectively. It was observed that the presence of ethanol was able to repress the hydrolytic activities, at which point the AAT activity was dominant. How ethanol can control the activities of Eat1 is unclear. It highlights that sufficient ethanol concentrations must be present to produce ethyl acetate with Eat1. It is also shown that Eat1 homologs are present in other bulk ethyl acetate-producing yeasts. These homologs are only distantly related to known AATs. Eat1 is therefore proposed to compose a novel alcohol acetyltransferase family. The discovery of this novel enzyme family was the cornerstone of the research presented in this thesis. The identification enabled further studies on the physiology of Eat1 and bulk ethyl acetate production in the native yeasts in <strong>Chapter 3</strong>. The cellular location of Eat1 in <em>Kluyveromyces lactis</em> was determined. The enzyme localised to the mitochondria of the yeast. This observation opposed the literature consensus which assumed that bulk ethyl acetate synthesis occurred in the yeast cytosol. Cytosolic acetyl-CoA flux in yeast is low and would presumably not support the synthesis of high amounts of ethyl acetate. The localisation of Eat1 in the mitochondria could better explain how bulk ethyl acetate synthesis occurs. The current hypothesis suggests that bulk ethyl acetate is an overflow product of yeast under iron limited conditions. Under these conditions, the entry of acetyl-CoA into the TCA cycle is impaired and acetyl-CoA accumulates. Eat1 is then proposed to relieve the accumulation by forming ethyl acetate. The TCA cycle and the major flux of acetyl-CoA in yeast are located in the mitochondria, where Eat1 is also located. It is thereby established that bulk ethyl acetate is a mitochondrial product of certain yeasts. <strong>Chapter 4</strong> describes the engineering of efficient ethyl acetate production in <em>E. coli</em> using Eat1 as the catalyst<em>.</em> Unlike yeast, the metabolism of <em>E. coli</em> can support the synthesis of ethyl acetate under anaerobic conditions. This removes the need for aeration, which is costly on a large scale. Establishing the anaerobic ethyl acetate pathway was faced with several bottlenecks, and removing them was the main theme of this chapter. The pathway towards ethyl acetate could be improved by disrupting by-product formation and optimising the expression levels of <em>eat1</em>. Further improvements could be made by removing the N-terminal mitochondrial localisation sequence of Eat1. These sequences typically destabilise proteins unless they are removed. These approaches did improve ethyl acetate formation by Eat1 to the point where ethanol levels were no longer sufficient to repress the hydrolytic activities of Eat1. To prevent ester hydrolysis the volatility of ethyl acetate was used to strip it from the fermenter. The result was ethyl acetate production at 63.4 % of the pathway maximum. We then looked beyond ethyl acetate as a bulk chemical in <strong>Chapter 5</strong> and investigated the role Eat1 has in general ester formation by yeasts, particularly <em>S. cerevisiae</em>. The formation of esters is industrially relevant also in<em> S. cerevisiae</em> as these compounds contribute to the aroma of fermented foods. <em>S. cerevisiae</em> naturally produces a variety of alcohols and thus provided a convenient platform to compare Eat1 homologs from different yeast. Expression of various <em>eat1</em> genes resulted in an increase of various in acetate and propionate ester levels. By disrupting the <em>S. cerevisiae</em> <em>eat1</em> gene the inverse effect was observed. Eat1 therefore seems to contribute to acetate ester synthesis in <em>S. cerevisiae</em> as well. In this chapter, a <em>S. cerevisiae</em> strain was generated where all known AAT genes were disrupted. Opposite to the expectations, ester production persisted in this strain, showing that even more ester-producing mechanisms exist. <strong>Chapter 6 </strong>focuses on the complex field of ester production as bulk chemicals, from the fundamentals of microbial ester production. Much research has been devoted to understanding the ester-producing processes in microorganisms. This is a daunting task as the structures and functions of esters in microorganisms are in many cases unrelated. Most esters are produced via the AAT reaction. With the exception of Eat1, which was identified in this thesis, other AATs have been studied for a long time. Still, relatively little is known about their structure and catalytic mechanisms, likely because there are no crystal structures available yet. Nevertheless, they have been applied extensively in metabolic engineering of ester production. The AAT reaction is simple. As long as a suitable alcohol and acyl-CoA are provided, the AAT will catalyse the ester formation. The real challenge of engineering ester formation is in the efficient supply of alcohols and acyl-CoA substrates. Some remarkable success has been made in recent years which is outlined in the review. In many cases, esters are less toxic and more readily extractable than the alcohols and acids they are composed of. They could therefore serve as a platform compound. In summary, this thesis contributed significantly to the knowledge of ester synthesis in yeast. The progress made on the production of ethyl acetate specifically can be used to develop new biotechnological, more sustainable production processes for this versatile compound. At the same time, much remains to be discovered.