JBA-06905; No of Pages 8

Biotechnology Advances xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv

Research review paper

Impact of synthetic biology and metabolic engineering on industrial

production of ne chemicals
David Jullesson a, Florian David a, Brian Peger b, Jens Nielsen a,c,
a
b
c

Department of Biology and Biological Engineering, Chalmers University of Technology, 41296 Gothenburg, Sweden
University of Wisconsin-Madison, Department of Chemical and Biological Engineering, 1415 Engineering Drive, Madison, WI 53706, United States
Novo Nordisk Foundation Center for Biosustainability, DK-2970 Hrsholm, Denmark

a r t i c l e

i n f o

Article history:
Received 16 September 2014
Received in revised form 29 January 2015
Accepted 19 February 2015
Available online xxxx
Keywords:
Synthetic biology
Metabolic engineering
Cell factories
Saccharomyces cerevisiae
Escherichia coli
Industrial production

a b s t r a c t
Industrial bio-processes for ne chemical production are increasingly relying on cell factories developed through
metabolic engineering and synthetic biology. The use of high throughput techniques and automation for the
design of cell factories, and especially platform strains, has played an important role in the transition from laboratory research to industrial production. Model organisms such as Saccharomyces cerevisiae and Escherichia coli
remain widely used host strains for industrial production due to their robust and desirable traits. This review
describes some of the bio-based ne chemicals that have reached the market, key metabolic engineering tools
that have allowed this to happen and some of the companies that are currently utilizing these technologies for
developing industrial production processes.
2015 Elsevier Inc. All rights reserved.

Contents
1.
Introduction . . . . . . . . . . . . . . . . . . . . .
2.
From system understanding to design of function . . . .
3.
Platform strains and their role in ne chemical production
4.
Industrial strain design . . . . . . . . . . . . . . . .
5.
Impact in industrial applications . . . . . . . . . . . .
6.
Discussion . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

1. Introduction
Cell factories created by engineering metabolic pathways are
capable of converting renewable feed-stocks into fuels, chemicals,
food ingredients and pharmaceuticals (Keasling, 2010). With increasing
climate change awareness alternative transportation fuels are needed.
Furthermore, many food, pharmaceuticals and cosmetic ingredients
are extracted from plants where seasonal dependent growth can
cause supply depletion and extraction methods can be expensive.
There is therefore much interest in developing cellular biocatalysts to
Corresponding author at: Department of Biology and Biological Engineering, Chalmers
University of Technology, 41296 Gothenburg, Sweden. Tel.: +46 31 772 3804; fax: +46
31 772 3801.
E-mail address: [email protected] (J. Nielsen).

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

0
0
0
0
0
0
0

produce direct replacements for specic chemicals as well as new

advanced bioproducts that have properties superior to existing
products. Many companies, both biotech and traditional chemical companies, are now translating research successes from both academic and
industrial groups to industrial processes. These ventures are motivated
by consumer demand for chemical products that are environmentally
friendly, less expensive, and possess superior properties compared
with those generated by traditional chemical synthesis. Advances in
industrial biotechnology and bioengineering over the last two decades and several successful implementations of novel industrial
processes have led to signicant growth of the eld of so-called
white biotechnology. Continued advances in DNA synthesis, synthetic
biology, and systems biology will only encourage further interest in
using genetically engineered cell factories for production of many
different chemicals.

http://dx.doi.org/10.1016/j.biotechadv.2015.02.011
0734-9750/ 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

Cell factories are designed and assembled using metabolic engineering and synthetic biology. Tools from these two elds can be used to redirect uxes towards desired metabolites. The two elds share the same
basis in bioengineering with the use of quantitative, model-driven
methods for predicting cellular phenotypes (Nielsen et al., 2014), and
a common in silico to in vivo approach. Metabolic engineering involves
enhancing or redirection of ux through metabolic pathways by making
genetic modications that alter the activity of enzymatic reactions.
Genetic modications include deletion of genes, replacement of gene
expression signals, and/or introduction of recombinant DNA cassettes
encoding foreign enzymes. Metabolic engineering also includes a
detailed analysis of the metabolic pathways to identify targets for manipulation (Nielsen and Jewett, 2008; Ostergaard et al., 2000). These
strategies often comprise elimination of unwanted activities, increasing
activity at ux controlling steps, and introduction of irreversible reactions to drive the ux in desired directions. Synthetic Biology is the
study of how to perform these manipulations in a quantitatively predictive way and how engineering principles can be applied to the design
and construction of biological systems. A pillar of synthetic biology is
the use of assembly standards for assembling genetic materials. This approach is adapted from other engineering disciplines such as electrical
engineering wherein complex systems can be built by combining separate, well-characterized parts. The scope of synthetic biology has grown
from construction of codon optimized genes to assembling a complete
synthetic genomes (e.g. Mycoplasma Synthia (Gibson et al., 2010)), assembling a designer eukaryotic chromosome (Annaluru et al., 2014) and
incorporation of new synthetic nucleotides (Malyshev et al., 2014) to increase the information content. The synergy between the two elds will
be further exploited to advance research in pathway engineering. However, the two elds can be distinguished as metabolic engineering
being a top-down approach, i.e. retrotting of the metabolism of a cell
factory, and synthetic biology being a bottom-up approach, i.e. the reconstruction of a new synthetic cell, is considered (Nielsen et al., 2014).
The performance of cell factories is typically evaluated through three
chemical production metrics: titer, rate (a.k.a. productivity), and yield
(TRY). In addition, a high performance cell factory should have the
qualities needed to thrive in an industrial setting including high osmotic
tolerance, broad pH tolerance, and minimal impact on the environment
(i.e. generally regarded as safeGRAS). The yeast Saccharomyces

cerevisiae meets most of these criteria, and it is therefore widely used

as a cell factory for production of food and beverages. Yeast has been
in the service of humans since the neolithic period. S. cerevisiae has
thus been accepted for industrial chemical production and has been
assigned to GRAS status. S. cerevisiae has also a low risk of contamination due to its low pH tolerance. Homologous recombination, which
also occurs to a high degree in yeast, allows for incorporation of genetic
fragments into the genome resulting in a more stable strain. Escherichia
coli is another well-studied cell factory. This model Gram-negative prokaryote is easy to genetically manipulate, an enormous knowledgebase
is available, and has been used to develop many of the underlying principles of synthetic biology and metabolic engineering. E. coli has been
used to produce chemicals commercially and remains a highly studied
system in the academic community. Success with model organisms
and the difculties encountered in conferring complex traits to them
has motivated interest to develop cell factories from non-model microorganisms that possess desired abilities. For example, photosynthetic
and methanotrophic organisms are being developed as cell factories to
gain the advantage of using CO2 or natural gas as carbon source.
Adaptation of synthetic biology tools and metabolic engineering
strategies for these and other organisms will enable the deployment
of future cell factories.
2. From system understanding to design of function
Over the past decades our knowledge about biological systems
has increased dramatically. To a large extent this is due to the many
techniques that have arisen, not only from biology, but from physics
and mathematics and that have proven useful in the service of biology.
This is illustrated in Fig. 1, summarizing major technologies and tools in
a historical context for the past 20 years combined with upcoming
trends. During the past 20 years, technologies have enhanced the scale
of bioengineering efforts from individual genes and gene products,
through pathways and traits, to complex systems encoded by genome
scale DNAincluding organisms. Technologies targeting small scales
have matured, but much remains to be learned when it comes to designing, constructing, and modifying chromosomes and entire genomes.
The technique that has inuenced molecular biology the most is the invention of Polymerase Chain Reaction (PCR) in the 1980s (Mullis et al.,

Fig. 1. The graph of knowledge and complexity. At present, we have reached a level of understanding, compared to 20 years ago, that allows us to design parts in a system but we lack a
systematic overview to fully understand what makes a system function appropriately.

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

1987). A decade earlier, recombinant DNA technology (Lobban and

Kaiser, 1973) was invented as a tool to fuse multiple sources of DNA
fragments into one, these techniques allowed for recombinant protein
production in a host cell. In the 1980s solid-phase phosphoramidite
chemistry was invented and is now the most prominent way of creating
synthetic DNA (Kosuri and Church, 2014). DNA synthesis has also been
optimized by the use of automated instruments, which have revolutionized the gene design eld. This automation allowed a reduction in
the purchase prize of completely synthesized genes from a price of
50 $/bp 20 years ago to b 0.2 $/bp today. This has enabled a design-offunction or bottom-up-approach rather than the trial and error approach, which was common in the rst attempts to create new genes.
The eld of physics provided X-ray crystallography (Drenth, 2007)
and nuclear magnetic resonance (Wishart et al., 1991), which have
been instrumental in studies of proteins and enzyme functions. These
two technologies have given rise to over 100,000 enzyme structures
and enzyme compositions (Bank, 2014). Earlier it was extremely computationally demanding to calculate the 3D structure and active sites
of enzymes as the algorithmic landscape for optimizing the lowest energy is too great for a computer to calculate. Today, computer tools like
FoldIt (Khatib et al., 2011) have enabled fast structure prediction as
the algorithm learns in ingenuities from the users how to fold the peptide chains in a more natural and correct manner. Directed evolution
(Crameri et al., 1998), which is also used for enzyme maturing, is a
high throughput technique where diversity is created by inserting random mutations into the genetic sequence encoding the enzyme. These
different variants are then analyzed to select one with desired properties (Voigt et al., 2000). Tools have been developed for altering enzyme
activity by controlling gene expression and therefore the amount of a
particular enzyme. Synthetic biology tools for controlling transcription
(Redden et al., 2014; Rhodius et al., 2012), translation (Na and Lee,
2010; Salis et al., 2009), and RNA turnover (Peger et al., 2006) have
been instrumental in balancing gene expression in metabolic pathways.
Most of these tools optimize expression from genes in cis, but recent
tools such as small RNAs, CRISPR/Cas, and TAL effectors offer the ability
to regulate genes in trans (Copeland et al., 2014; Yoo et al., 2013). Cloning methods such as Ligase Chain Reaction (LCR) (Kok et al., 2014) or
Gibson assembly (Gibson et al., 2009), have greatly enhanced the ability
to assemble libraries of DNA constructs and enabled high throughput
exploration of possible sequence space. Unfortunately, our ability to
predict the optimal level of gene expression and our understanding of
the pathways in a metabolic process remains limited. As a metabolic
ux of interest is likely to be controlled by several enzymes, altering
one specic enzyme might not be enough. To identify ux control the
concept of metabolic control analysis (MCA) is useful, even though it requires extensive information about the kinetics of the individual enzymes. Other modeling approaches like genome metabolic models
(GEM) combined with ux balance analysis (FBA) and analysis of elementary ux modes are easier to implement as these models require
only a few parameters, but they cannot provide specic information
about ux control (Kerkhoven et al., 2014). C13-labeled ux analysis is
a technique that allows a more accurate model based estimation of metabolic uxes and is used to detect ux changes and potential metabolic
bottlenecks (Blank et al., 2005; Soons et al., 2013). Through combination
with elementary ux mode analysis it is even possible to identify
ux control at branch points in the metabolic network (Bordel and
Nielsen, 2010). In the future, genome scale kinetic models may give a
systemic view that allows for precise identication of bottlenecks and
targets for manipulation.
With the Human Genome Project (HUGO) project and complete genome sequences of several microorganisms hopes were high to nd the
formula to create complete synthetic in silico organisms. However, this
turned out to be harder to be realized than anticipated, particular as our
knowledge about molecular interactions, both in terms of connectivity
and in terms of kinetics, are limited. Thus, even though it has been possible to create synthetic chromosomes (Annaluru et al., 2014) a

milestone of full understanding of the complete function of minimal genomes has yet not been achieved. However, many approaches to generate a minimal genome have been employed and this is likely to reveal a
set of key genes required for cellular function (Kumagai et al., 2014;
Moran and Bennett, 2014). These techniques, and in the future even
better and more reliable techniques, will aid us in reaching a deeper systematic understanding that will allow for a design-offunction, not only
of single genes and proteins but rather of whole microorganisms with
desired functionalities.
3. Platform strains and their role in ne chemical production
One of the most applied concepts when it comes to ne chemical
production is the usage of platform strains that have desired traits
(Fig. 2). This approach allows easy insertion of different product formation pathways and thereby signicantly reduces the development time,
which is of great importance for commercial success. Feedstocks that
are utilized by platform strains can vary, but platform strains have
been evolved to utilize several different raw materials. The platform
strain is engineered to have robust properties such as stress tolerance,
fermentation performance and substrate utilization. This enables the
strain to be used under industrial conditions, where the environment
can be rather harsh for the organism to exclude contaminations and
reach high cell densities. Products produced by platform strains can
range from high volume chemicals, such as biofuels and commodity
chemicals, to high value chemicals such as ne chemicals and proteins.
At the center of cellular metabolic networks a set of twelve chemicals,
so-called precursor metabolites, resides from which all cellular building
blocks and chemical products can be derived (Neidhardt et al., 1992;
Nielsen, 2003). Metabolic reactions can therefore be organized into one
of three categories: 1.) catabolic reactions, 2.) anabolic reactions, and 3.)
central metabolic reactions. Catabolic reactions comprise pathways that
convert feedstock (e.g. carbon source) into precursor metabolites, reducing power, and energy. Anabolic reactions comprise pathways that

Fig. 2. The concept of a platform strain as a plug-and-play solution for industrial production. A platform strain is robust and possesses desirable traits thus enabling easy upstream development of new processes. Platform strains possessing desired abilities can
be selected to process a desired feedstock and convert it into one of several different products. In the chemical market, market volume is typically inversely proportional to the
product selling price. Cell factories have long been used for high value products such as
proteins and are increasingly being used for molecules at the other end of the spectrum.

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

Fig. 3. The bow tie structure of metabolism with the main metabolic precursor metabolites residing in the center of the three super-pathways: catabolic reactions, central metabolism and
anabolic reactions. Key precursor metabolites are: G3P (glyceraldehyde 3-phosphate), pyruvate, acetyl-CoA (acetyl-Coenzyme A), oxaloacetate, a-ketoglutarate, succinate, fatty acids,
prenyl-pyrophosphates, and acyl-thioesters.

consume reducing power and energy to produce cellular components

(e.g. lipids, nucleic acids, cell wall) or desired chemical products. Central
metabolic reactions are those that enable the cell to interconvert
between the twelve precursor metabolites thereby permitting production of all cellular components from a single catabolic pathway.
This structure is akin to the unit operation framework for chemical process design. In this way, a biocatalyst can be designed by
combining technologies for consuming a desired feedstock with
an anabolic pathway for producing a desired chemical product.
From this perspective, metabolism is shaped like a bow tie with a large
number of pathways funneling into a small number of central metabolites
that then branch out into a wide range of anabolic pathways. Given this
structure, the strongest intellectual property position is obtained by controlling the branch points. For this reason, groups seek to develop platform strains for producing metabolites at the branch points of pathways
within specic chemical families. For example, isoprenoids are a large
family of compounds (N 50,000) made from two, 5-carbon building blocks
(isopentylpyrophosphate (IPP) and dimetylallylpyrophosphate (DMAPP)
(Kirby and Keasling, 2008)), and further from acetyl-CoA (Chen et al.,
2012, 2013). Technologies that enable high volume production of these
two precursors can then be used to produce any or all of the highervalue downstream products. Amyris has used this strategy to develop
platform strains for producing isoprenoid precursors and applied the
technology to a wide range of markets including fuels (e.g. farnesene),
drugs (e.g. artemisinin), avors and fragrances (e.g. patchouli oil)
and other chemicals (e.g. squalene). Similar efforts by other groups have
targeted branch metabolites such as malonyl-CoA, acetyl-CoA, pyruvate,
succinate, with the goal of ultimately producing oleochemicals,
polyketides, fusel alcohols, amino acids, diols, avonoids, or other ne
chemicals. Other groups have developed platform technologies for consuming specic feed-stocks such as pentose sugars, cellulose, and/or
CO2 that they plan to use in combination with various anabolic pathways. Lastly, groups have screened, evolved, and engineered platform
strains that possess specic traits such as pH tolerance, solvent tolerance, thermal stability (Caspeta et al., 2014) with the hypothesis that
superior (cheaper, more stable, more efcient) processes can be run
under these extreme conditions. For example, the production of lactic
acid was greatly enhanced by development of biocatalysts based on
yeasts that could tolerate low pH thereby circumventing the costs
associated with neutralization and purication of lactic acid at neutral
pH (Ilmn et al., 2007) (Fig. 3).

4. Industrial strain design

Cell factories are developed using an iterative design, build, test
cycle. Projects often start by selecting a platform strain possessing a
large portion of required traits. The design phase adds genes that confer
the missing traits and enables production of a target compound from a
target feedstock. The design phase can be accomplished with different
strategies. These range from establishing entirely new metabolic pathways, to optimizing gene expression of existing pathways, to replacing
specic genes. Designs must optimize the essential metabolic pathways,
but must also consider strain physiology (e.g. product tolerance, energy
management, robustness) (Steensels et al., 2014). Here, techniques for
whole genome engineering come into play comprising rational sitedirected and more untargeted techniques (David and Siewers, 2014).
The most classical tool for strain improvement is by inducing random
mutations through UV light or chemicals and selecting for a desired
phenotype. In many cases a genetic basis for desired traits or optimal
pathway function is unknown. Therefore high throughput techniques
capable of screening large libraries are used to process thousands of
new strains each week. Tools for creating these libraries include combinatorial genetics where tens to hundreds of genes derived from various
species are randomly combined and tested. Standardized tools like genetic circuits (Brophy and Voigt, 2014) with well characterized properties enable rational approaches and fast developments. Using a directed
evolution approach specic enzymes of the pathway are typically in
focus of the optimization. Here, diversity is created by gene mutations
combined with an appropriate screening approach to nd a benecial
variant. The nal functionality of a component of interest sometimes
has to be enhanced by specic modications. Here, decoration technologies like glycosylation and transfer of molecular oxygen to \CH, \NH
or \SH bonds (Renault et al., 2014; Weitzel and Simonsen, 2013) are of
great importance. In this context, libraries with different enzyme
variants for modifying glycosylation patterns and cytochrome P450
enzymes are used.
For high throughput screenings it is essential to have a suitable readout signal to speed up the optimization cycle. These can be based on so
called function-led screens were the main parameter is e.g. cell survival
or GFP expression using transcription factors or riboswitch based biosensors (Schallmey et al., 2014; Yang et al., 2013). These function-led screens
can have as much as 1 billion screening events per day. For structure-led
screens LC-MS and NMR can be used for evaluating the performance of

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

Fig. 4. The development of industrial strains for chemical production undergoes several iterative cycles where desired traits and scale-up are guided by the many available engineering
tools and analysis techniques.

pathways leading to the desired product, but the throughput of these

screens is much lower. When evaluating the engineered strain, after establishing the pathway and the design, there are some main goals that
the strains must achieve to be ready for industrial production. These
goals are: reduced by-product formation, variety and efciency of feedstock conversion (yield), speed of production (rate), a high nal titer
and robustness of the cell factory. When the strain is evolved to an intermediate strain, the fermentation development and scale-up are of outmost importance as the chosen strain often dictates the design of the
full scale production process. Scale-up typically emerges late in the process and any problems occurring at this stage with the chosen organism
may have serious impact on the development time and can signicantly
increase costs. Following screening the chosen strain is often analyzed
in detail through systems analysis and omics techniques (Kim et al.,
2012), and based on this new engineering strategies are implemented.
This cycling optimization of the cell factory forms the so-called iterative
loop for strain development, or the metabolic engineering cycle
(Nielsen, 2001), that following many cycles nally results in a cell factory
that meets the TRY requirement for implementing a commercially viable
production process. (Fig. 4)
5. Impact in industrial applications
As previously mentioned, metabolic engineering and synthetic
biology have an impact on multiple industrial sectors ranging from production of chemicals, biofuels, food ingredients and supplements, and
pharmaceuticals. Fig. 5 provides a representative overview of the current market situation related to companies, their target products and
the particular current market status. These grouped products range
from low-value, high market volume products like biofuels and commodity chemicals to high-value small market volume products like
food additives and pharmaceuticals. Companies generally aim for producing chemical building blocks, or strategic intermediates that can be
further used by chemical processes to create even higher value-added
products. Three stages of development will be considered in the following discussion to highlight the stage of current development projects:
small-scale laboratory development (1), pilot-scale evaluation (2) and

commercial scale production (3). Here we primarily focus on metabolite

based products produced by yeast, bacteria or algae, and will discuss
some products that have made it to commercial scale.
Hydrocortisone was one of the rst complex products demonstrated
to be produced in yeast through metabolic engineering and synthetic biology. The recombinant pathway involves 13 engineered genes including a P450 system (Szczebara et al., 2003). Sano is currently working
on an industrial production based on this system (Brocard-Masson
et al., 2012). Another commercial scale product that has been launched
in recent yeasts is D-lactic acid for use in production of polylactic acid
(PLA), a bioplastic. A main advantage of the bioprocess is that the
production of only the D-lactic acid is permitted to ensure proper PLA
formation and to prevent expensive separation of the two forms. Cargill
(Carlson and Peters, 2002) was the rst to develop a novel process for
lactic acid production using a low pH tolerant yeast strain, and they
have been the key driver of introducing PLA on the market. Today
there are, however, several industrial producers of lactic acid, e.g.
Myriant that produces lactic acid from non-food cellulosic feedstock
since 2008. In the biofuel space a good example of a commercial product
is Gevo's production of isobutanol from renewable feedstock for the
chemical and fuel market. Existing ethanol-producing plants were
successfully retrotted with a production capacity of 38 million gallons
of isobutanol per year. Another example of biofuel production is
Solazyme's production of biodiesel based on engineered microalgae.
This biofuel is fully compliant with Fatty Acid Methyl Ester (FAME)
standards. There is much interest in advanced biofuels that resemble
traditional fuels, and Amyris has established a successful production of
farnesene (Keasling et al., 2005). Farnesene can be dehydrated to
farnesane that can be used as a diesel fuel, but also be in surfactants,
lubricants or personal care products. Another commercial scale compound based on the farnesene platform strain is squalene, which is a
high value ingredient, used in cosmetics and personal care products
and produced on a commercial basis since 2011. Succinic acid is another
important product in the chemical eld and Bioamber produces this
chemical using a genetically engineered E. coli or a low pH tolerant
yeast strain licensed from Cargill. According to the Environmental Protection Agency (EPA) Bioamber produces succinic acid at costs 40%

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

Fig. 5. Companies in the eld of microbial engineering for ne chemical production (Allylix, 2014; Amyris, 2014; BioAmber, 2014; Bioenergies, 2014; Biofuels, 2014; Biotechnologies, 2014;
Butamax, 2014; Cargill, 2014; Codexis, 2014; Digest, 2014; DSM, 2014; DuPont, 2014; Energy, 2014; Evolva, 2014; Explorer, 2014; Genomatica, 2014; Gevo, 2014; International Flavors and
Fragrances, 2014; Intrexon, 2014; Isobionics, 2014; Mascoma, 2014; Merck, 2014; Metabolix, 2014; Modular Genetics, 2014; Myriant, 2014; Natureworks, 2014; Project, 2014; REG, 2014;
Sano, 2014a; Solazyme, 2014; Synthetic Genomics, 2014; Unlimited, 2014; Verdezyne, 2014; Ziopharma Oncology, 2014), data also generated from Biofuels Digest (2014) and Synthetic
Biology Project (2014). Four major areas are visualized: Food, Pharma Chemicals, and Biofuels. The number corresponds to the stage in progress that the company has reached: 1) smallscale laboratory development, 2) over pilot-scale, and 3) commercial scale.

less than petroleum based production. It can be widely applied as a

chemical building block for production of various chemicals and polymers as well as food, drugs and cosmetic ingredients. Another example
of commercial production of a chemical building block is Genomatica's
commercially feasible production of 1,4 butandiol (Burk et al., 2011),
an important precursor compound for plastics and the textiles industry
with a value of 2000$/ton and global market sales of $4 billion per year.
In ve years Genomatica successfully optimized the production strain
and the associated bioprocess and succeeded in licensing it to
Novamont and BASF, the largest BDO producer in the world. Similar
type chemicals are produced by Lanzatech, that is producing ethanol
and 2,3-BDO from syngas by using microbial conversion. A rst commercial facility is operating at a steel mill in China, producing 30 million
gallons of ethanol per year based on waste industrial ue gases.
Having a closer look at the eld of food additives, companies like
Isobionics (Sonke and de Jong, 2012) and Allylix (recently acquired by
Evolva) found a niche in producing the citrus avors nalencene and
nootkatone based on engineered terpene producing yeast cells. Another
product that recently entered commercial scale production is vanillin,
which was launched by Evolva in collaboration with International Flavor & Fragences Inc. (Hansen et al., 2013). A major product candidate
from the pharma sector is artemisinin, an anti-malaria drug that is traditionally derived from plants. Here, Sano/Amyris recently launched
large scale industrial artemisinin production and entered the market
based on original development of the cell factory by Amyris (Sano,
2014b).

6. Discussion
Here we provide an overview of the current impact of synthetic biology and metabolic engineering in the industrial area where numerous
companies have developed commercially viable cell factories. Also
established big players in the chemical sector like BASF, Cargill, Dupont,
DSM and Sano are increasingly involved in the eld of biochemical
production and typically enter collaborations with smaller biotech companies, through establishing partnerships, formation of joint-ventures
or by licensing their technologies. We have seen in the last years how
synthetic biology and metabolic engineering have delivered novel and
innovative solutions to ensure process robustness and utilization of
different substrates at the same time reaching targets on high titer,
rate, yield and purity of the cell factory that can be used for industrial
production. All this has been possible through many years of tool and
technique development where non-automated and labor intense
techniques have been replaced by fully automated systems for high
throughput screening. DNA sequencing, DNA synthesis and modeling
have all gone through rapid changes as computer power has increased.
Could our knowledge in in silico genome re-engineering be boosted by
articial intelligence and neural networks or is it our ideas and lack of
understanding that is the bottleneck in the nal frontier of creating
cell factories?
In addition, when it comes to industrial production and developing
new platform strains; the host strain matters. Shall one rather go for
a few well-studied main workhorses or rather go for multiple hosts

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

with different special production properties? Through bio-prospecting

novel strains can be found that are more suitable for specic compound
production. However, these strains need to be evolved to strains that are
applicable for industrial fermentation and if needed go through GMP assessment. Furthermore, lack of suitable tools for genetic manipulation
for such strains often hinders their use for metabolic engineering.
However, with further research novel host strains can be adapted to industrial fermentation by altering just a few key regulators in the strains'
metabolic machinery (Damkir et al., 2013).
Looking at the product spectrum of the current bio-based industry
we see a clear trend focusing on chemicals rather than biofuels but
also food additives and pharmaceuticals are of increasing interest. One
has to keep in mind that these products still have to compete with the
current petroleum or plant based production methods, meaning that
besides the argument of being more sustainable the process also has
to be cost efcient. This has been a main driver for companies solely
dedicated to biofuels to shift towards a broader spectrum of more
valuable products, but still rely on platform strains as an opportunity
to produce biofuels.
When it comes to process economics the feedstock is of great importance. Most bioprocesses still use glucose as carbon source, which
makes the discussion of food versus fuel still relevant. Currently,
many efforts are made to enable 2nd generation production of fuels
and chemicals. Here one aims to use biomass derived cellulosic material
as a direct carbon source. So far established processes still rely on
chemical and enzymatic pretreatment but the nal aim is to engineer
microorganisms to break down cellulosic polymers. The ultimate goal
is to use photosynthetic organisms as production host only relying on
CO2 and light for growth. Though there are much interest in this space
there are still no industrial processes established, and an in depth
system understanding is required, in particular to reach a sufcient
high cell density to ensure high productivities, but also novel scale-up
technologies are limiting industrial implementation.
The biggest bottleneck for industrial implementation of novel
bioprocesses is often in the scale-up step, where the host strain has to
be chosen denitely as it will generally be too expensive to change at
a later stage. From an industrial perspective S. cerevisiae is besides
others benets as mentioned above, a robust host strain with many
desired traits such as high osmo-tolerance, low pH-tolerance and a permissive insertion of recombinant DNA through homologous recombination. S. cerevisiae is widely used both in academia and in industry as
illustrated by several of the examples discussed here. The wide use of
S. cerevisiae in academic research is important, as academia and industry
have in many cases joined forces where we are starting to see the fruits
of these joint ventures in products that have reached the market.
References
Allylix. http://www.allylix.com/, 2014.
Amyris I. http://www.amyris.com/, 2014.
Annaluru N, Muller H, Mitchell LA, Ramalingam S, Stracquadanio G, Richardson SM, et al.
Total synthesis of a functional designer eukaryotic chromosome. Science 2014;344:
558.
Bank PD. Yearly growth of total structures; 2014.
BioAmber. http://www.bio-amber.com/, 2014.
Bioenergies G. http://www.global-bioenergies.com, 2014.
Biofuels A. http://www.algenol.com/, 2014.
Biofuels Digest. http://www.biofuelsdigest.com/, 2014.
Biotechnologies O. http://www.opxbio.com/, 2014.
Blank L, Kuepfer L, Sauer U. Large-scale 13C-ux analysis reveals mechanistic principles of
metabolic network robustness to null mutations in yeast. Genome Biol 2005;6:R49.
Bordel S, Nielsen J. Identication of ux control in metabolic networks using nonequilibrium thermodynamics. Metab Eng 2010;12:36977.
Brocard-Masson C, Bonnin I, Dumas B. Process for preparing genetically transformed
yeasts capable of producing a molecule of interest at a high titre; 2012.
Brophy JAN, Voigt CA. Principles of genetic circuit design. Nat Methods 2014;11:50820.
Burk MJ, Van Dien SJ, Burgard AP, Niu W. Compositions and methods for the biosynthesis
of 1,4-butanediol and its precursors; 2011.
Butamax. http://www.butamax.com/, 2014.
Cargill. http://www.cargill.com/, 2014.
Carlson TL, Peters EM. Low PH lactic acid fermentation; 2002.

Caspeta L, Chen Y, Ghiaci P, Feizi A, Buskov S, Hallstrm BM, et al. Altered sterol
composition renders yeast thermotolerant. Science 2014;346:758.
Chen Y, Siewers V, Nielsen J. Proling of cytosolic and peroxisomal acetyl-CoA
metabolism in Saccharomyces cerevisiae. PLoS One 2012;7:e42475.
Chen Y, Daviet L, Schalk M, Siewers V, Nielsen J. Establishing a platform cell factory
through engineering of yeast acetyl-CoA metabolism. Metab Eng 2013;15:4854.
Codexis. http://www.codexis.com/, 2014.
Copeland MF, Politz MC, Peger BF. Application of TALEs, CRISPR/Cas and sRNAs as transacting regulators in prokaryotes. Curr Opin Biotechnol 2014;29:4654.
Crameri A, Raillard S-A, Bermudez E, Stemmer WPC. DNA shufing of a family of genes
from diverse species accelerates directed evolution. Nature 1998;391:28891.
Damkir S, Yang L, Molin S, Jelsbak L. Evolutionary remodeling of global regulatory
networks during long-term bacterial adaptation to human hosts. Proc Natl Acad Sci
2013;110:776671.
David F, Siewers V. Advances in yeast genome engineering. FEMS Yeast Res 2014;15:
114.
Drenth J. Principles of protein X-ray crystallography. Springer; 2007.
DSM. http://www.dsm.com/corporate/home.htm, 2014.
DuPont. http://biosciences.dupont.com/, 2014.
Energy S. http://www.sapphireenergy.com/, 2014.
Evolva. http://www.evolva.com/, 2014.
Explorer M. http://www.metabolic-explorer.com/, 2014.
Genomatica. http://www.genomatica.com/, 2014.
Gevo. http://www.gevo.com/, 2014.
Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO. Enzymatic assembly
of DNA molecules up to several hundred kilobases. Nat Methods 2009;6:3435.
Gibson DG, Glass JI, Lartigue C, Noskov VN, Chuang RY, Algire MA, et al. Creation of a
bacterial cell controlled by a chemically synthesized genome. Science 2010;329:526.
Hansen J, Hansen EH, Sompalli HP, Sheridan JM, Heal JR, Hamilton WDO. Compositions
and methods for the biosynthesis of vanillin or vanillin beta-d-glucoside; 2013.
Ilmn M, Koivuranta K, Ruohonen L, Suominen P, Penttil M. Efcient production of l-lactic
acid from xylose by Pichia stipitis. Appl Environ Microbiol 2007;73:11723.
International Flavors and Fragrances I. http://www.iff.com/, 2014.
Intrexon. http://www.dna.com/, 2014.
Isobionics. http://www.isobionics.com/, 2014.
Keasling JD. Manufacturing molecules through metabolic engineering. Science 2010;330:
13558.
Keasling JD, Martin VJJ, Newman JD, Pitera DJ, Reiling KK, Withers IST. Methods for
identifying a biosynthetic pathway gene product; 2005.
Kerkhoven EJ, Lahtvee P-J, Nielsen J. Applications of computational modeling in metabolic
engineering of yeast. FEMS Yeast Res 2014;15:113.
Khatib F, DiMaio F, Cooper S, Kazmierczyk M, Gilski M, Krzywda S, et al. Crystal structure
of a monomeric retroviral protease solved by protein folding game players. Nat Struct
Mol Biol 2011;18:11757.
Kim I-K, Roldo A, Siewers V, Nielsen J. A systems-level approach for metabolic engineering of yeast cell factories. FEMS Yeast Res 2012;12:22848.
Kirby J, Keasling JD. Metabolic engineering of microorganisms for isoprenoid production.
Nat Prod Rep 2008;25:65661.
Kok Sd, Stanton LH, Slaby T, Durot M, Holmes VF, Patel KG, et al. Rapid and reliable DNA
assembly via ligase cycling reaction. ACS Synth Biol 2014;3:97106.
Kosuri S, Church GM. Large-scale de novo DNA synthesis: technologies and applications.
Nat Methods 2014;11:499507.
Kumagai H, Sasaki M, Idiris A, Tohda H. Minimum genome factories in Schizosaccharomyces
pombe. In: Anazawa H, Shimizu S, editors. Microbial production. Japan: Springer; 2014.
p. 1724.
Lobban PE, Kaiser AD. Enzymatic end-to-end joining of DNA molecules. J Mol Biol 1973;
78:45371.
Malyshev DA, Dhami K, Lavergne T, Chen T, Dai N, Foster JM, et al. A semi-synthetic
organism with an expanded genetic alphabet. Nature 2014;509:3858.
Mascoma. http://www.mascoma.com/, 2014.
Merck. http://www.merck.com/index.html, 2014.
Metabolix. http://www.metabolix.com/, 2014.
Modular Genetics I. http://www.modulargenetics.com/, 2014.
Moran NA, Bennett GM. The tiniest tiny genomes. Annu Rev Microbiol 2014;68. [null].
Mullis KB, Erlich HA, Arnheim N, Horn GT, Saiki RK, Scharf SJ. One of the rst Polymerase
Chain Reaction (PCR) patents; 1987.
Myriant. http://www.myriant.com/, 2014.
Na D, Lee D. RBSDesigner: software for designing synthetic ribosome binding sites that
yields a desired level of protein expression. Bioinformatics 2010;26:26334.
Natureworks L. http://www.natureworksllc.com/, 2014.
Neidhardt FC, Ingraham JL, Schaechter M. Physiology of the bacterial cell. A molecular
approach. Biochem Educ 1992;20:1245.
Nielsen J. Metabolic engineering. Appl Microbiol Biotechnol 2001;55:26383.
Nielsen J. It is all about metabolic uxes. J Bacteriol 2003;185:70315.
Nielsen J, Jewett MC. Impact of systems biology on metabolic engineering of Saccharomyces
cerevisiae. FEMS Yeast Res 2008;8:12231.
Nielsen J, Fussenegger M, Keasling JD, Lee SY, Liao JC, Prather K, et al. Engineering synergy
in biotechnology. Nat Chem Biol 2014;10:4.
Ostergaard S, Olsson L, Nielsen J. Metabolic engineering of Saccharomyces cerevisiae.
Microbiol Mol Biol Rev 2000:3450.
Peger BF, Pitera DJ, Smolke CD, Keasling JD. Combinatorial engineering of intergenic
regions in operons tunes expression of multiple genes. Nat Biotechnol 2006;24:
102732.
Redden H, Morse N, Alper HS. The synthetic biology toolbox for tuning gene expression in
yeast. FEMS Yeast Res 2014;15:110.
REG I. http://www.regi.com/, 2014.

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011

D. Jullesson et al. / Biotechnology Advances xxx (2015) xxxxxx

Renault H, Bassard J-E, Hamberger B, Werck-Reichhart D. Cytochrome P450-mediated

metabolic engineering: current progress and future challenges. Curr Opin Plant Biol
2014;19:2734.
Rhodius VA, Mutalik VK, Gross CA. Predicting the strength of UP-elements and full-length
E. coli (E) promoters. Nucleic Acids Res 2012;40:290724.
Salis HM, Mirsky EA, Voigt CA. Automated design of synthetic ribosome binding sites to
control protein expression. Nat Biotechnol 2009;27:94650.
Sano. http://en.sano.com/, 2014.
Sano. Press release: rst antimalarial treatments produced with semisynthetic
artemisinin enter market; 2014b.
Schallmey M, Frunzke J, Eggeling L, Marienhagen J. Looking for the pick of the bunch:
high-throughput screening of producing microorganisms with biosensors. Curr
Opin Biotechnol 2014;26:14854.
Solazyme. http://solazyme.com/, 2014.
Sonke T, de Jong RM. Valencene synthase; 2012.
Soons ZITA, Ferreira EC, Patil KR, Rocha I. Identication of metabolic engineering targets
through analysis of optimal and sub-optimal routes. PLoS One 2013;8:e61648.
Steensels J, Snoek T, Meersman E, Nicolino MP, Voordeckers K, Verstrepen KJ. Improving
industrial yeast strains: exploiting natural and articial diversity. FEMS Microbiol Rev
2014;38:94795.

Synthetic Genomics I. http://www.syntheticgenomics.com/, 2014.

Synthetic Biology Project. http://www.synbioproject.org/, 2014.
Szczebara FM, Chandelier C, Villeret C, Masurel A, Bourot S, Duport C, et al. Total biosynthesis of hydrocortisone from a simple carbon source in yeast. Nat Biotechnol 2003;
21:1439.
Unlimited J. http://www.jouleunlimited.com/, 2014.
Verdezyne I. http://verdezyne.com/, 2014.
Voigt CA, Kauffman S, Wang ZG. Rational evolutionary design: the theory of in vitro
protein evolution. Adv Protein Chem 2000;55:79160.
Weitzel C, Simonsen H. Cytochrome P450-enzymes involved in the biosynthesis of monoand sesquiterpenes. Phytochem Rev 2013;118.
Wishart DS, Sykes BD, Richards FM. Relationship between nuclear magnetic resonance chemical shift and protein secondary structure. J Mol Biol 1991;222:
31133.
Yang J, Seo SW, Jang S, Shin S-I, Lim CH, Roh T-Y, et al. Synthetic RNA devices to expedite the evolution of metabolite-producing microbes. Nat Commun 2013;4:
1413.
Yoo SM, Na D, Lee SY. Design and use of synthetic regulatory small RNAs to control gene
expression in Escherichia coli. Nat Protoc 2013;8:1694707.
Ziopharma Oncology I. http://www.ziopharm.com/, 2014.

Please cite this article as: Jullesson D, et al, Impact of synthetic biology and metabolic engineering on industrial production of ne chemicals,
Biotechnol Adv (2015), http://dx.doi.org/10.1016/j.biotechadv.2015.02.011