Microbial production of alka(e)ne biofuels

By Sanket Lunawat

Introduction Due to depleting petroleum supply and environmental issue presented by

fractional distillation of crude oil, many researchers seek to exploit the use of biofuels as a clean energy
source [1,2]. Many governments have invested greatly in the application of biofuels for years [3].
Although various types of biofuels have already been put in everyday use, many of them have several
draw- backs due to their chemical properties. Bioethanol, for example, has considerably lower energy
density than petroleum based fuels; its high hygroscopicity and vola- tility make it harder to preserve [4].
Alka(e)ne, however, as one of the hydrocarbon biofuels and chemical feed-stock derived from microbial
fatty acid metabolic path-way, has no such disadvantages. Key features of product alka(e)ne (chain
length; degree of saturation; straight or branched) are determining factors for their use. According to its
chain length, alka(e)ne constitutes most of the petroleum based fuels (gasoline, aviation fuel, etc.). The
presence of unsaturated and branched alka(e)ne ensures fuels low freezing-point during high altitude [5].
Alkenes, on the other hand, have also been in use for the production of chemical products such as deter-
gents and lubricants. The potential of alka(e)ne generating system is deter- mined by three essential
factors: metabolic pathway engineering, microbial platform and culture condition. In this review, we
discuss these three factors’ impact on alka(e)ne production individually. With Table 1 and Figure 1
summarizing detailed strategies for alka(e)ne biosynthesis in recent years, we elaborate several unique
findings in main text. Among these factors, pathway engineering attract most of the attention. However,
to some extent, we deem it is equally important to investi- gate suitable host and culture method for
alka(e)ne production.

Metabolic pathway engineering for alka(e)ne biosynthesis

Since the confirmation of two main steps responsible for the synthesis of alka(e)nes in cyanobacteria [6],
many researches attempt to construct this alka(e)ne biosynthe-sis pathway heterologously with the
expectation of ele-vated yield and altered alkane properties. As shown in Table 1 and Figure 1, core
alka(e)ne generating pathway often involves a simple binary structure with one enzyme (Table 2)
tapping into the host fatty acid metabolism to provide precursor, while second enzyme (Table 3)
converting this precursor into hydrocarbons [7,8__,9,10,11_,12,13]. The first step of alka(e)ne
biosynthesis usually directs carbon source into three precursors (Figure 1 and Table 2): fatty aldehydes
(in most cases), fatty acids and acyl-CoA/ACP. This step is highly susceptible to native competing
pathways. For instance, when using fatty aldehyde as precursor, many studies detect fatty alcohol
byproduct (Table 1). In some extreme cases, the titer of fatty alcohols even exceeds the target alka(e)nes’
[7,14_,15__]. This is mainly because the presence of multiple fatty aldehyde reductase homologues in
Escher-ichia coli (more than 13) as well as in Saccharomyces cerevisiae [16,17]. Another pathway
competing for alde-hyde precursor is fatty aldehyde dehydrogenation path-way [7,14_,15__,16,18]. Study
shows that, when expressing AAR and ADO in Aspergillus carbonarius, the level of internal fatty acids and
triacylglycerol rise as soon as alkane emerges. This is attributed to the presence of fatty aldehyde
dehydrogenase which allows aldehydes to be redirected into fatty acids [19 _]. Many experiments start
with the deletion of these competing pathways in order to raise the accumulation of fatty aldehyde
precur-sor thus improve the overall performance of alka(e)ne generating system. However, considering
the intensity of competing pathway elimination, Cao et al. show that the total removal of aldehyde
reductase activity might not be the best scenario for alkane production. Only if the alkane/alcohol titer is
within the range of 1.5–3.3, alkane titer is able to reach its peak. The intensive reduction of fatty alcohol
byproduct will somehow lower the titer of alkane [20_]. It is worth mentioning that, in need of supplying
the system with sufficient precursor, research- ers often introduce different types of thioesterase into the
system. These thioesterase successfully improve the concentration of internal free fatty acids by more
than three-fold, which ultimately become product alkane [21–26]. In addition to the enhancement of total
alkane yield, thioesterase also serves as a valuable control point for managing chain length of final alka(e)ne
product. Due to different preferences towards substrate chain length, co- expressing of a thioesterase from
Umbellularia californica (UcFatB) and a fatty acyl-CoA reductase from Acinetobacter sp. M1, both of which
prefer C12 and C14 substrate, Yan et al. successfully obtained shorter chain alkanes (2.21 mg/g C11 and 1.83
mg/g C13) [27]. With the help of C4-specific thioesterase Tes4 from Bacteroides fragilis, Kallio et al. managed
to acquire 32 mg/L of propane from an engineered E. coli [28]. As shown in Figure 1, fatty acid metabolism is
tightly linked with alka(e)ne biosynthesis. Alterations of fatty acid synthesis/degradation pathway often affect
the supply of necessary precursors. With the modification of fatty acid metabolic pathway (DfadE; DfadR;
+fadD; +TesA0), Choi et al. acquired 580.8 mg/L alkane from an engineered E. coli [23]. Combining fatty
aldehyde reductase deletion with the overexpression of fatty acid metabolism regulator protein (FadR), Song
et al. obtained product alka(e)ne with a titer of 255.6 mg/L [29]. In fact, most researchers have attempted to
optimize the fatty acid metabolic pathway in order to boost alka(e)ne biosynthesis. Some studies even
designed a reverse-b-oxidation pathway with enhanced theoretical efficiency to increase short chain alka(e)ne
production [25__]. Most of these measures seem to be effective. The second step of alka(e)ne biosynthesis is
often regarded as rate limiting due to the low conversion rate from precursor to alka(e)ne product [6,16]. The
perfor- mance of several ADOs, CER1 and CYP4G, which con-vert fatty aldehydes into alka(e)nes, have been
compared in a S. cerevisiae fatty acid overproducing strain [14_]. Although CER1 and CYP4G clearly have
activities towards fatty aldehydes [23,30,31], in this work, only ADOs lead to the production of medium chain
alka(e) nes. Due to the reliability of ADOs in different hosts, most researchers tend to incorporate them into
their work as the final step towards alka(e)ne. ADOs obtained from different cyanobacterias often possess
significant differ- ences considering their catalytic abilities [6,12,14_,32]. Restructuring these ADOs may
provide the system with a better performance. For example, ADO-catalase fusion protein could alleviate the
inhibition caused by hydrogen peroxide [33]; point mutations of different ADOs may alter their preferences
towards substrates chain length [21,34–36]; spatial organization of AARs and ADOs could enhance the overall
production by 8.8-fold [37]. Enzymes using fatty acids as substrate (OleTJE/UndA/UndB) are not very popular in
these years’ work. However, due to the fact that their catalytic processes are relatively short when using fatty
acid/triacylglycerol as the substrate, they represent a promising direction for efficient alkene bio-synthesis.
UndA and UndB are fatty acid decarboxylase recently found in Pseudomonas. The difference betweenthem is
that UndA is ubiquitous in Pseudomonas sp. whereas UndB is a membrane-bound fatty acid decarbox- ylase
which only exists in few Pseudomonas. Expression of an UndB homologue (Pmen_4370) in E. coli leads to the
production of 55 mg/L 1-undecene, which is 3-fold higher than OleTJE or UndA [38,39__]. More than 95% of
calcu-lated lauric acid is converted to 1-undecene, indicating its great advantage over ADOs considering the
generation of alkenes.

Microbial platform for alka(e)ne biosynthesis Many organisms, including plant, insect, bacteria,
yeast and fungus, have the ability to synthesize alka(e)nes. Some studies show that, alka(e)nes may provide
the host with additional ability to deal with harsh environ- ment or means of communication (e.g.
pheromones) [30,31,40,41]. Generally speaking, alka(e)ne yields in natural producers are usually low [42]. It is
unpractical to use these natural producers to generate alka(e)nes on industrial scale. However, a strain of
Aureobasidium mel- anogenum recently isolated from mangrove ecosystem in Hainan Province of China
shows an exception. This strain seems to be able to produce more than 28 g/L alkanes when provided with
optimal culture conditions [43_,44]. Given the fact that huge amount of microorgan- isms exist in nature, one
cannot rule out the possibility that some of them may have the potential to be used directly as a microbial
platform for alka(e)ne biosynthe-sis. As mentioned above, fatty acid metabolism is critical to the production of
alka(e)ne. Theoretically speaking, any microorganism with a highly effective fatty acid metabolic pathway has
the potential to become a microbial platform for alka(e)ne biosynthesis, even though they cannot produce
hydrocarbons naturally. Up to now, E. coli is the most extensively used microbialplatform for alka(e)ne
synthesis. The highest amount of alkane produced by microorganism is achieved in a meta-bolic engineered E.
coli (580.8 mg/L). Despite the alde-hyde reductase (>13) redundancy presence in E. coli, some studies show
that, when given fatty acids as sub-strate, E. coli can only assimilate a small amount (_24%) to generate
alka(e)nes and fatty alcohols, indicating its insufficient activity towards fatty acid utilization [45]. Therefore,
researchers tend to use glucose as carbon source for alka(e)ne generation, as well as engineer E. coli towards
high fatty acid accumulation at the same time. Employing yeast as host often results in lower yield than E. coli.
Evidence indicates that some yeast possess at least 12 alk genes which are responsible for alkane degradation
[46]. Alkane chain length between C9–11 also has poison- ous effect towards S. cerevisiae [47]. Moreover,
complex cellular structure of eukaryotes often hinders the meta- bolic process due to the
compartmentalization between cellular organelles [48]. Cofactors required for alka(e)ne generating enzymes
are often synthesized in mitochon- dria and peroxisome [49] whereas artificial routes from substrate to
alka(e)ne usually take place in cytosol. By repositioning the alka(e)ne biosynthesis pathway into the
peroxisome of S. cerevisiae, Zhou et al. enhanced the titer of alka(e)ne and fatty alcohols by seven-fold.
Additional deletion of certain genes increase the number of peroxi-some, leads to another three-fold titer
enhancement [15__]. Reports about alka(e)ne generating cyanobacteria can trace back to 1960’.
Cyanobacterias obtain their energy through photosynthesis and are able to use CO 2 as sole carbon source.
With sunlight and aeration, they can accumulate a lot of biomass in a short period of time [3]. With expression
of multiple copies of AAR and ADO, metabolic engineered cyanobacteria can produce alkanes up to 12.9% of
their cell dry weight [50_,51,52]. Metabolic engineered chemoautotrophic bacteria Cupriavidus neca-tor can
also transform CO2 into alka(e)nes [8__]. Due to the abundance of CO2 in the atmosphere and its special role as
a greenhouse gas, CO2-assimilating microbes like cyanobacteria or C. necator represent a unique type of
microbial alka(e)ne producing platform which has a very promising future.

Conclusions
Among renewable biofuels, the most commonly used in todays market are bioethanol and biodiesel. Wild
type S. cerevisiae KL17 can produce ethanol at a concentration of 96.9 g/L [62]. Although biodiesel is
mainly produced by chemical transesterification of plant oil with alcohol, in situ esterification can also
lead to a biodiesel titer of 1.28 g/ L by coupling microbial fatty acid assimilation with ethanol formation in
metabolic engineered E. coli [63]. However, compared to bioethanol and biodiesel, low yield of microbial
produced alka(e)ne still has a long way to go to fulfill industrial use. Continuing search for efficient
alka(e)ne biosynthesis pathway is an essential task as well as the search for suitable microbial platform.
Most work have taken place in E. coli, which lacks the strong ability in fatty acid accumulation and
assimilation even with intensive genetic modifications. We believe it is beneficial to employ different
kinds of oleaginous, lipolytic and CO2-fixing microbes into the production of alka(e)nes in future. These
microorganisms will bestow the system with higher efficiency, shorter metabolic pathway and cheaper
feedstock. Successful integration of these microbes with efficient alkane generating path-way should
deliver more satisfactory alka(e)ne yield. Under optimized culture conditions, these alka(e)ne gen-erating
platform may one day fulfill industrial purposes. Conflicts of interest None. .

References and recommended

Papers of particular interest, published within the period of review, have been highlighted as: _ of special interest __ of outstanding interest 1.
Stephanopoulos G: Challenges in engineering microbes for biofuels production. Science 2007, 315:801-804. 2. Kerr RA: Climate change. Global
warming is changing the world. Science 2007, 316:188-190. 3. Su YJ, Song KH, Zhang PD, Su YQ, Cheng J, Chen X: Progress of microalgae biofuels
commercialization. Renew Sustain Energy Rev 2017, 74:402-411. 4. Peralta-Yahya PP, Zhang F, del Cardayre SB, Keasling JD: Microbial engineering
for the production of advanced biofuels. Nature 2012, 488:320-328. 5. Kallio P, Pasztor A, Akhtar MK, Jones PR: Renewable jet fuel. Curr Opin
Biotechnol 2014, 26:50-55. 6. Schirmer A, Rude MA, Li X, Popova E, del Cardayre SB: Microbial biosynthesis of alkanes. Science 2010, 329:559-562