AcCoA is not only a building block of FA
but also of many compound classes that are of
interest in biotechnology, such as isoprenoids
(e.g., artemisinic acid; Paddon et al. 2013),
polyesters (e.g., polyhydroxybutyric acid;
Kocharin et al. 2012), polyketides (e.g., 6methylsalicylic acid; Wattanachaisaereekul
et al. 2008), or flavonoids (e.g., naringenin;
Koopman et al. 2012). To avoid constraints
such as limited metabolic capacity of organelles, availability of cofactors, and transport
of products across organellar membranes, heterologous pathways for production of these
compounds are preferably expressed in the
cytosol. Therefore, significant effort has been
devoted to improving the cytosolic AcCoA supply in S. cerevisiae. An excellent review article
(van Rossum et al. 2016b) provided a systematic analysis of possible pathways to optimize
the supply of cytosolic AcCoA under consideration of reaction stoichiometry, energy conservation, and maximally attainable yields for four
model AcCoA-derived compounds including
FA. Here, we will revisit the current literature
that exploited some of those different possibilities (summarized in Fig. 14.1) to produce FA
and other AcCoA-derived products. Since the
provision of reducing equivalents—either
NADH or NADPH—is equally important for
FA production and intrinsically dependent on
the choice of the AcCoA synthesis route, the
redox cofactor supply will be discussed
throughout this chapter. The synthesis of FA
can be engineered to occur either via FAS that
strictly requires NADPH or by reversing b-oxidation, a FA degradation pathway, which allows
for more flexibility regarding the cofactor
dependence (see Sect. III.B). Therefore, the
choice of the appropriate AcCoA pathway also
depends on the synthesis mode of FA.
A. Engineering the Pyruvate Dehydrogenase
Bypass
In most proof-of-concept studies, the production of AcCoA-derived products in S. cerevisiae
relied on the native PDH bypass or engineered
variants of it, without introducing heterologous
AcCoA routes. Even when various alternative
AcCoA pathways were introduced, the native
PDH bypass was present in the background of
all engineered strains, with one exception (Meadows et al. 2016) known to us (see Sect. II.C).
An increased flux through the PDH bypass
can be achieved by redirecting acetaldehyde
away from reduction to ethanol by alcohol
dehydrogenases (ADHs) toward oxidation to
acetate by ALDs. In a first study (aiming to
increase the production of the isoprenoid
amorphadiene), it could be shown that overexpressing the acetaldehyde dehydrogenase
ALD6 alone substantially increased the accumulation of acetate (Shiba et al. 2007). Since
Ald6 is NADP-specific, this step is also important to provide NADPH required by FAS. Next,
the authors tested different ACS variants,
whereby only the expression of a heterologous
enzyme from Salmonella enterica, mutated to
prevent
inactivation
by
acetylation
(SeACS
L641P ), led to a substantial increase in
amorphadiene production (Shiba et al. 2007).
In the same study, the overexpression of endogenous enzymes had no (for Acs2) or little (for
Acs1) effect on amorphadiene titers, which was
attributed to negative regulatory mechanisms
at the post-translational level, possibly acetylation. In subsequent studies, a positive effect of
combined Ald6/SeACS
L641P overexpression on
production of several AcCoA-derived products,
including FA derivatives such as FA ethyl esters
(FAEEs) and hexadecanol, was confirmed by
different groups (Chen et al. 2013b; Krivoruchko et al. 2013; Jong et al. 2014; Lian et al.
2014; Feng et al. 2015).
As a strategy to utilize the major fermentation product
ethanol for AcCoA synthesis, the ADH isoform II
(Adh2), which is the main enzyme responsible for
oxidation of ethanol to acetaldehyde, was co-expressed
with Ald6/SeACS
L641P in some variants (Chen et al.
2013b; Krivoruchko et al. 2013; Jong et al. 2014; Feng
et al. 2015). Unfortunately, the effect of Adh2 overexpression alone remained unclear, since it was overexpressed only in combination with other enzymes
(e.g., acetoacetyl-CoA thiolase Erg10) that affect the
product yield (Kocharin et al. 2012; Chen et al. 2013b;
Krivoruchko et al. 2013) or together with Ald6 and
SeACS
L641P , without a direct comparison to a control
not overexpressing Adh2 (Jong et al. 2014; Feng et al.
2015). Considering that the interconversion of acetaldehyde and ethanol is rather driven by the chemical equilibrium, it remains yet to be demonstrated that Adh2
overexpression indeed diverts the flux toward AcCoA in
engineered strains in the presence of glucose.
14 Engineering Saccharomyces cerevisiae for Production of Fatty Acids and Their Derivatives
341
but also of many compound classes that are of
interest in biotechnology, such as isoprenoids
(e.g., artemisinic acid; Paddon et al. 2013),
polyesters (e.g., polyhydroxybutyric acid;
Kocharin et al. 2012), polyketides (e.g., 6methylsalicylic acid; Wattanachaisaereekul
et al. 2008), or flavonoids (e.g., naringenin;
Koopman et al. 2012). To avoid constraints
such as limited metabolic capacity of organelles, availability of cofactors, and transport
of products across organellar membranes, heterologous pathways for production of these
compounds are preferably expressed in the
cytosol. Therefore, significant effort has been
devoted to improving the cytosolic AcCoA supply in S. cerevisiae. An excellent review article
(van Rossum et al. 2016b) provided a systematic analysis of possible pathways to optimize
the supply of cytosolic AcCoA under consideration of reaction stoichiometry, energy conservation, and maximally attainable yields for four
model AcCoA-derived compounds including
FA. Here, we will revisit the current literature
that exploited some of those different possibilities (summarized in Fig. 14.1) to produce FA
and other AcCoA-derived products. Since the
provision of reducing equivalents—either
NADH or NADPH—is equally important for
FA production and intrinsically dependent on
the choice of the AcCoA synthesis route, the
redox cofactor supply will be discussed
throughout this chapter. The synthesis of FA
can be engineered to occur either via FAS that
strictly requires NADPH or by reversing b-oxidation, a FA degradation pathway, which allows
for more flexibility regarding the cofactor
dependence (see Sect. III.B). Therefore, the
choice of the appropriate AcCoA pathway also
depends on the synthesis mode of FA.
A. Engineering the Pyruvate Dehydrogenase
Bypass
In most proof-of-concept studies, the production of AcCoA-derived products in S. cerevisiae
relied on the native PDH bypass or engineered
variants of it, without introducing heterologous
AcCoA routes. Even when various alternative
AcCoA pathways were introduced, the native
PDH bypass was present in the background of
all engineered strains, with one exception (Meadows et al. 2016) known to us (see Sect. II.C).
An increased flux through the PDH bypass
can be achieved by redirecting acetaldehyde
away from reduction to ethanol by alcohol
dehydrogenases (ADHs) toward oxidation to
acetate by ALDs. In a first study (aiming to
increase the production of the isoprenoid
amorphadiene), it could be shown that overexpressing the acetaldehyde dehydrogenase
ALD6 alone substantially increased the accumulation of acetate (Shiba et al. 2007). Since
Ald6 is NADP-specific, this step is also important to provide NADPH required by FAS. Next,
the authors tested different ACS variants,
whereby only the expression of a heterologous
enzyme from Salmonella enterica, mutated to
prevent
inactivation
by
acetylation
(SeACS
L641P ), led to a substantial increase in
amorphadiene production (Shiba et al. 2007).
In the same study, the overexpression of endogenous enzymes had no (for Acs2) or little (for
Acs1) effect on amorphadiene titers, which was
attributed to negative regulatory mechanisms
at the post-translational level, possibly acetylation. In subsequent studies, a positive effect of
combined Ald6/SeACS
L641P overexpression on
production of several AcCoA-derived products,
including FA derivatives such as FA ethyl esters
(FAEEs) and hexadecanol, was confirmed by
different groups (Chen et al. 2013b; Krivoruchko et al. 2013; Jong et al. 2014; Lian et al.
2014; Feng et al. 2015).
As a strategy to utilize the major fermentation product
ethanol for AcCoA synthesis, the ADH isoform II
(Adh2), which is the main enzyme responsible for
oxidation of ethanol to acetaldehyde, was co-expressed
with Ald6/SeACS
L641P in some variants (Chen et al.
2013b; Krivoruchko et al. 2013; Jong et al. 2014; Feng
et al. 2015). Unfortunately, the effect of Adh2 overexpression alone remained unclear, since it was overexpressed only in combination with other enzymes
(e.g., acetoacetyl-CoA thiolase Erg10) that affect the
product yield (Kocharin et al. 2012; Chen et al. 2013b;
Krivoruchko et al. 2013) or together with Ald6 and
SeACS
L641P , without a direct comparison to a control
not overexpressing Adh2 (Jong et al. 2014; Feng et al.
2015). Considering that the interconversion of acetaldehyde and ethanol is rather driven by the chemical equilibrium, it remains yet to be demonstrated that Adh2
overexpression indeed diverts the flux toward AcCoA in
engineered strains in the presence of glucose.
14 Engineering Saccharomyces cerevisiae for Production of Fatty Acids and Their Derivatives
341