2015 - Productionofbiofuelsandbiochemicalsbyinvitrosynthe (Retrieved-2016!01!25)

Biotechnology Advances 33 (2015) 14671483
Contents lists available at ScienceDirect
Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv
Research review paper
Production of biofuels and biochemicals by in vitro synthetic biosystems:

Opportunities and challenges
Yi-Heng Percival Zhang
Biological Systems Engineering Department, Virginia Tech, 304 Seitz Hall, Blacksburg, VA 24061, USA
Cell Free Bioinnovations Inc., Suite 1200B, 2200 Kraft Drive, Blacksburg, VA 24060, USA
Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences, 32 West 7th Avenue, Tianjin Airport Economic Area, Tianjin 300308, China
a r t i c l e
i n f o
Available online 27 October 2014

Keywords:
Biochemicals
Biofuels
Economic analysis
In vitro synthetic biosystem
In vitro synthetic biology
Innovative biomanufacturing
Paradigm shift
a b s t r a c t
The largest obstacle to the cost-competitive production of low-value and high-impact biofuels and biochemicals
(called biocommodities) is high production costs catalyzed by microbes due to their inherent weaknesses, such
as low product yield, slow reaction rate, high separation cost, intolerance to toxic products, and so on. This predominant whole-cell platform suffers from a mismatch between the primary goal of living microbes cell proliferation and the desired biomanufacturing goal desired products (not cell mass most times). In vitro synthetic
biosystems consist of numerous enzymes as building bricks, enzyme complexes as building modules, and/or
(biomimetic) coenzymes, which are assembled into synthetic enzymatic pathways for implementing complicated bioreactions. They emerge as an alternative solution for accomplishing a desired biotransformation without
concerns of cell proliferation, complicated cellular regulation, and side-product formation. In addition to the
most important advantage high product yield, in vitro synthetic biosystems feature several other
biomanufacturing advantages, such as fast reaction rate, easy product separation, open process control, broad reaction condition, tolerance to toxic substrates or products, and so on. In this perspective review, the general design rules of in vitro synthetic pathways are presented with eight supporting examples: hydrogen, n-butanol,
isobutanol, electricity, starch, lactate,1,3-propanediol, and poly-3-hydroxylbutyrate. Also, a detailed economic
analysis for enzymatic hydrogen production from carbohydrates is presented to illustrate some advantages of
this system and the remaining challenges. Great market potentials will motivate worldwide efforts from multiple
disciplines (i.e., chemistry, biology and engineering) to address the remaining obstacles pertaining to cost and
stability of enzymes and coenzymes, standardized building parts and modules, biomimetic coenzymes,
biosystem optimization, and scale-up, soon.
2014 Elsevier Inc. All rights reserved.
Contents
1.
2.
3.
4.
Introduction . . . . . . . . . . . . . . . .
Opportunities of in vitro synthetic biosystems .
2.1.
General design principles . . . . . .
2.2.
Examples of biofuels and bioelectricity
2.2.1.
Hydrogen . . . . . . . . .
2.2.2.
n-Butanol . . . . . . . . .
2.2.3.
Isobutanol . . . . . . . . .
2.2.4.
Bioelectricity . . . . . . . .
2.3.
Examples of biochemicals . . . . . .
2.3.1.
Synthetic amylose . . . . .
2.3.2.
Lactate
. . . . . . . . . .
2.3.3.
1,3-Propanediol . . . . . .
2.3.4.
Poly-3-hydroxylbutyrate . .
Economic analysis of the hydrogen case . . . .
Challenges and solutions . . . . . . . . . .
Corresponding author. Tel.: +1 540 231 7414.

E-mail addresses: ypzhang@vt.edu, profCH2O@outlook.com.
http://dx.doi.org/10.1016/j.biotechadv.2014.10.009
0734-9750/ 2014 Elsevier Inc. All rights reserved.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1468
1470
1470
1471
1471
1472
1473
1474
1474
1475
1475
1476
1476
1476
1478
1468
Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483
4.1.
Stable enzymes as standardized building blocks . . . . . . . . . .
4.2.
Synthetic multiple-enzyme complexes (metabolons) as building modules
4.3.
Bulk enzyme production and separation . . . . . . . . . . . . . .
4.4.
NAD(P)-dependent enzyme engineering . . . . . . . . . . . . . .
4.5.
System optimization and modeling . . . . . . . . . . . . . . . .
4.6.
Stabilization of thermolabile metabolites at elevated temperatures . .
4.7.
Scale-up by integrative innovation . . . . . . . . . . . . . . . .
4.8.
Other issues . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conclusions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.Introduction
The sustainability revolution is taking place in this century mainly
due to concerns of depleting fossil fuels and climate changes. At the
same time, growing world population and increasing living standards
require more natural resource consumption. Biomass produced from
terrestrial plants is the most abundant renewable bioresource, approximately ve times global energy consumption (Zhang, 2013). Biomass
and its carbohydrates are the sole cost-competitive energy and carbon
sources that will be converted to produce biofuels and biochemicals
instead of fossil fuels (Wyman, 1999). Compared to low energy concentration (i.e., nonpoint) solar energy, biomass is a concentrated chemical
energy, which could be harvested, stored, transported, and converted to
other chemical energy forms relatively easily (Zhang, 2011c).
Human beings had utilized living whole-cell microorganisms for
making numerous fermentative products for thousands of years. Living
whole-cell fermentation is the predominant biomanufacturing platform
(Fig. 1a). However, the primary goal of microorganisms is their proliferation while bioconversions are the side effects. Recent advances in synthetic biology and systems biology present numerous breakthroughs,
such as the production of non-natural products (e.g., 1,4-butanediol
(Yim et al., 2011), and isobutanol (Atsumi et al., 2008)). However,
some inherent constraints of living microorganisms (e.g., net ATP generation, intact cellular membrane) prevent them from implementing
some important chemical reactions, for example, 12 H2 production
from one glucose and water.
Whole cell lysates (Fig. 1b) have been used as an important scientic
tool to investigate complicated biological reactions for more than
100 years. For example, Eduard Buchner discovered that the yeast lysate converted glucose to ethanol (Buchner, 1897). Due to his
paradigm-shifting discovery, he won the Nobel Chemistry Prize in
1907. Later, numerous scientists applied this tool to discover and
. .
.
. .
. .
. .
. .
. .
. .
. .
. .
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1478
1479
1479
1479
1480
1480
1480
1480
1481
1481
1481
study natural metabolic pathways. For example, Harden et al. discovered key enzymes in glycolysis (Nobel Chemistry Prize 1929), Calvin
elucidated the CO2 assimilation in plants (Nobel Chemistry Prize
1961), and Nirenberg and Matthaei interpreted the genetic code and
its function in protein synthesis (Nobel Physiology Prize 1968)
(Matthaei et al., 1962). Recently, cell-free protein synthesis has been
suggested to be the fastest way to make recombinant proteins, even
for membrane or complicated proteins (Carlson et al., 2012; Swartz,
2013). Cell-free protein synthesis has been scaled up to 100 l levels recently (Hodgman and Jewett, 2012).
In vitro synthetic biosystems emerge as a manufacturing platform by
the assembly of numerous enzymes and enzyme complexes from different sources and/or (biomimetic) coenzymes (Fig. 1c). Such systems
could surpass the constraints of whole-cells and cell lysates for
implementing some biological reactions that microbe cannot do, for example, high yield production of hydrogen (Martn del Campo et al.,
2013), or enzymatic transformation of cellulose to starch (You et al.,
2013). Although in vitro synthetic biosystems are on their early stage,
they have a great potential to becoming a disruptive biomanufacturing
platform, especially for low-cost production of biofuels and biochemicals.
Table 1 presents the comparison of biomanufacturing advantages between whole cell-based fermentations and in vitro (cell-free) synthetic
systems, from product yield, volumetric productivity, reaction condition,
product separation, product titer, product purity to process control and
optimization. Among them, three key criteria of biomanufacturing are
(i) product yield mainly related to feedstock cost, (ii) volumetric
productivity (or reaction rate) mainly related to capital investment,
and (iii) product separation (or product titer) mainly related to processing costs.
The history of biomanufacturing platforms accompanied with key
milestones is presented in Fig. 2. Before Louis Pasteur clearly proposed
the theory of biogenesis (1864), human beings had used microorganisms
Fig. 1. Schemes of biotransformation catalyzed by whole cell (a), cell extract (b), and in vitro synthetic biosystem (c).
1469
Table 1
Comparison of biomanufacturing features by microbial cells and in vitro synthetic biosystems.
Feature
Microbial cells
In vitro synthetic
biosystems
Comparative examples
Ref.
Product yield
Low-modest
Very high
Volumetric productivity
Low-modest
Modest-high
Reaction conditions
Narrow
Broad
Hydrogen
1,3-propanodiol
Electricity
Lactate
Poly-3-hydroxylbutyrate
Synthetic starcha
Electricity
Recombinant proteins b
Toxic products (e.g., isobutanol)
Toxic compounds (e.g., biomass hydrolysate)
Presence of organic solvents
Martn del Campo et al. (2013); Myung et al. (2014)

Rieckenberg et al. (2014)
Zhu et al. (2014b)
Ye et al. (2012a)
Opgenorth et al. (2014)
You et al. (2013)
Sakai et al. (2009); Zhu et al. (2014b)
Carlson et al. (2012)
Guterl et al. (2012)
Wang et al. (2011)
Chen and Arnold (1993); Serdakowski and Dordick
(2008)
Arnold (1998); Shaw et al. (1999)
Kragl et al. (2002)
Martn del Campo et al. (2013); Myung et al.
(2014)
Opgenorth et al. (2014)
You et al. (2013)
Carlson et al. (2012)
Guterl et al. (2012)
Goerke and Swartz (2008); Kanter et al. (2007)
Qi et al., 2014; You et al. (2013)
Satoh et al. (2003)
Zhu et al. (2014b)
Martn del Campo et al. (2013); Myung et al. (2014)
Goerke and Swartz (2008); Kanter et al. (2007);
Wang and Zhang (2009a)
Ardao and Zeng (2013); Rollin et al. (2014); Ye et al.
(2009)
Qi et al. (2014); You et al. (2013)
Krutsakorn et al. (2013a)
Product separation
Cellular membrane
No membrane
Product titer
Product purity/quality
Low-high
Low-modest
Low high
Very high
Reaction control and

optimization
Hard
Easy
Extreme pH or temperature
Presence of ionic liquids
Hydrogen
Starch
Isobutanol
Recombinant antibodies b
Synthetic starch
Electricity
Hydrogen
Hydrogen
Synthetic starch
n-butanol
a
b
High product yield due to co-production of ethanol, resulting in no sugar losses.

Cell-free protein synthesis.
for making wine, beer, bread, cheese and so on by solid state fermentation for thousands of years. Weizmann developed anaerobic liquid fermentation by using a mono-culture to produce acetone in World War I.
In World War II, the submerged aerobic fermentation technology was
invented to produce penicillin at low costs. The next major breakthroughs were recombinant DNA technology and mammalian cell
cultures for the production of erythropoietin (EPO), insulin, vaccines, antibodies, and so on in the 1980s. In this century, new biomanufacturing
Fig. 2. History of biomanufacturing catalyzed by whole cells and in vitro (cell-free) biosystems associated with key milestones.
1470
breakthroughs will revolutionize industrial bioprocess by producing a

myriad of new or old products through more cost-effective approaches.
The development of in vitro (cell-free) technologies lagged far
behind the whole cell-based technologies (Bornscheuer et al.,
2012; Vasic-Racki, 2006). Although Eduard Buchner discovered the
phenomenon of cell-free ethanol fermentation in the 1890s, the
use of one puried enzyme along with enzyme immobilization for
industrial production began in the 1960s1970s, for example, high
fructose corn syrup (i.e., more than 10 billion US dollars yearly)
and semi-synthetic antibiotics (e.g., cephalosporin) (Demain,
2004). Such cell-free systems evolved to more complicated ones
multi-enzyme one pot for enhancing volumetric productivity, decreasing product inhibition, and shifting reaction equilibrium
(Santacoloma et al., 2010; Schoffelen and van Hest, 2012). For example, the pharmaceutical and ne chemistry industries adopt this
platform to produce high-value chiral alcohols, -hydroxy acids,
and -amino acids, such as, (S)-2-butanol, L-tert-leucine, (S)-ethyl4-chloro-3-hydroxybutyrate, atorvastatin, and so on (Bornscheuer
et al., 2012; Huisman et al., 2010; Wildeman et al., 2007). In the organic chemistry eld, the synthesis of monosaccharides, activated
monosaccharides, oligosaccharides, and glycopeptides by using
multi-enzyme one pot has been intensively investigated, such as, Lfructose, 5-deoxy-5-ethyl- D -xylulose, amylose, and so on (Endo
and Koizumi, 2000; Fessner, 1998; Fessner and Helaine, 2001;
Franke et al., 2003; Huang et al., 2006; Qi et al., 2014; Schoevaart
et al., 2000; Zhang et al., 2005). In this century, a few researchers
propose to put more than four biocatalytic components in one vessel
to implement very complicated reactions comparable to microbial
cell factories. Because in vitro synthetic biosystems for future
biomanufacturing are on their early stage, different names have
been suggested in the literature, such as, synthetic pathway biotransformation (SyPaB) (Zhang, 2010a, 2011a), synthetic cascade
biomanufacturing (Stefer and Sieber, 2013), cell-free bioprocessing
(Swartz, 2013), synthetic biochemistry (Korman et al., 2014;
Opgenorth et al., 2014), cell-free biosystems for biomanufacturing
or cell-free biomanufacturing (You and Zhang, 2013), in vitro metabolic engineering (Honda et al., 2004), cascade enzyme factories
(Zhang and Huang, 2012), and so on.
Here I present pathway design principles of in vitro synthetic
biosystems for biomanufacturing of biofuels and biochemicals (called
biocommodities) with eight examples, provide a detailed economic
analysis for enzymatic hydrogen production for identifying the remaining challenges, and suggest respective solutions. The other applications
of in vitro synthetic systems or cell extracts are not covered here, for example, the synthesis of ne chemicals (Bruggink et al., 2003; Bujara
et al., 2011), of radiolabelled products (Rodriguez and Leyh, 2014;
Schultheisz et al., 2008, 2009), of special compounds (Schultheisz
et al., 2008; Scism and Bachmann, 2010), of proteins (Carlson et al.,
2012; Hodgman and Jewett, 2012; Swartz, 2013), and of drugs (Boltje
et al., 2009; Xu et al., 2011, 2014), in vitro genetic or protein circuits
and computation (Hockenberry and Jewett, 2012; Katz and Privman,
2010; Kim and Winfree, 2011; Noireaux, 2003; Yordanov et al., 2014),
in-depth understanding of in vivo biosystems (Bogorad et al., 2013;
Keller et al., 2013), and so on.
2. Opportunities of in vitro synthetic biosystems
For the production of low-value and high-impact biocommodities,
the major cost factor is carbohydrate costs, accounting for at least half
or even more than 80% product price (Lynd et al., 1999; Zhang,
2010a). In vitro synthetic biosystems could lower feedstock costs than
whole-cell systems because of high product yields (that is, neither the
synthesis of cell mass nor the formation of side-products) (Table 1). Furthermore, because balanced coenzymes are designed in in vitro synthetic pathways (General design principles), in vitro synthetic
biosystems could have very low processing costs (that is, no aeration,
low cooling energy) and low capital investment, especially important

for large-scale production of biocommodities. As shown in Table 1,
in vitro synthetic biosystems could have numerous biomanufacturing
advantages over traditional whole-cell fermentation, leading to great
opportunities to the production of biocommodities. Here we present
the design principles of in vitro synthetic enzymatic pathways with
eight supporting examples four for biofuels and bioelectricity and
four for biochemicals.
2.1. General design principles
The design principles of in vitro synthetic biosystems for manufacturing include (i) in vitro pathway reconstruction, (ii) enzyme selection,
(iii) enzyme engineering, (iv) enzyme production, and (v) process
engineering (Zhang, 2010a). Whole processes can be improved in an
iterative manner by inputs of different specialists (e.g., enzyme engineers, bioprocess engineers, biochemists, bulk enzyme producers,
bioinformaticians, organic chemists, and so on), gradually leading to an
efcient industrial process (Zhang, 2010a). The design of an in vitro synthetic enzymatic pathway is central, while the others (i.e., enzyme selection, enzyme engineering, enzyme production, and process engineering)
are pretty mature. Synthetic enzymatic pathways are usually designed
based on natural metabolic pathways with some modications. Because
the same biochemical reactions can be designed by different pathways,
synthetic pathways need to be designed carefully by considering ATP balance, NAD(P) balance, NAD and NADP matching, coenzyme involvement,
enzyme chosen, thermodynamics, reaction equilibrium, product separation, and so on (Rollin et al., 2013; Zhang, 2010a).
In vitro synthetic enzymatic pathways must have a balance in ATP
production and consumption while in vivo synthetic pathways in
whole cells can obtain or depose extra ATP from or to cellular metabolisms. If net ATP is generated for the case of in vitro ethanol fermentation via the glycolysis pathway, the accumulation of ATPs stops
systems from running for a long time (Welch and Scopes, 1985). If
ATP input is needed, it is important to have low-cost ATP regeneration
systems. The best solution is careful design of pathways without ATP involvement or with ATP balance. Different from the typical glycolytic
pathway from glucose to two pyruvate that generates two moles of
ATP, in vitro pathways can produce pyruvate without ATP involvement
(Guterl et al., 2012) or produce zero net ATP (Krutsakorn et al., 2013a;
Ye et al., 2012a). If net ATP is accumulated in the overall process, the addition of ATPase or arsenate or phosphatase could dissipate some ATP
(Welch and Scopes, 1985) or hydrolyze high-energy phosphate bondcontaining metabolites (Moradian and Benner, 1992). If a small amount
of ATP is required as an input for the whole system, several ways can be
selected, such as the generation of glucose 1-phosphate by using glucan
phosphorylases (Wang and Zhang, 2009a; Ye et al., 2009; Zhang et al.,
2007), the generation of sugar phosphates from polyphosphate (Liao
et al., 2012; Martin del Campo et al., 2013), or the regeneration of ATP
from the secondary energy sources, such as, polyphosphate, creatine
phosphate, phosphoenolpyruvate, and so on, through substrate phosphorylation (Resnick and Zehnder, 2000; Wang and Zhang, 2009a). It
is noted that the use of costly second energy sources prohibits their application for biocommodity production. If a large amount of ATP is needed as an input, cell-free oxidative phosphorylation (Jewett et al., 2008)
or novel ATP synthase-based regeneration systems may be a choice
(Rupp, 2013). But the long-term stability of synthetic membranes
should be addressed.
Similar to ATP balance, it is essentially important to design
NAD(P)H-balanced pathways. First, the type of reduced NAD(P)H generated from substrates should match that of NAD(P)H consumption
for the production of desired products. If not, it can be addressed to
nd out new enzymes using the same type of coenzymes from nature
(Krutsakorn et al., 2013a), engineer coenzyme preferences by rational
protein design (Opgenorth et al., 2014; Rollin et al., 2013) or add another enzyme transhydrogenase. Second, when in vitro pathways generate
extra reduced coenzymes, such reduced coenzymes should be removed

by using hydrogenase under anaerobic conditions (Wichmann and
Vasic-Racki, 2005), by using NAD(H) oxidase under aerobic conditions
(Opgenorth et al., 2014), or via electron mediators in enzymatic fuel
cells (Zhu et al., 2012). When pathways need extra reduced coenzymes
inputs, NAD(P)H can be usually generated by using a hydrogen-donor
substrate and one of the following: a single enzyme, cascade enzymes,
and electro-enzymes. Single-enzyme systems include alcohol/alcohol
dehydrogenase (Wichmann and Vasic-Racki, 2005), formate/formate
dehydrogenase (Bozic et al., 2010), glucose/glucose dehydrogenase
(Xu et al., 2007), glucose 6-phosphate/glucose 6-phosphate dehydrogenase (Wong and Whitesides, 1981), dihydrogen/hydrogenase (Mertens
and Liese, 2004; Wong et al., 1981), and phosphite/phosphite dehydrogenase (Johannes et al., 2007). Single-enzyme NAD(P)H regeneration
systems are widely used in the synthesis of high-value of chiral compounds in the pharmaceutical industry. A 12-enzyme system is used
to produce nearly 12 NADPH from one glucose unit of cellobiose
(Wang et al., 2011). Among all hydrogen-donor compounds, renewable
sugars have the lowest substrate costs, but this 12-enzyme system requires more enzymes and increases system complexity (Wang et al.,
2011). The utilization of electrochemistry to generate reduced coenzymes is low-cost and clean, but the instability of NAD(P)H under
high over-potential must be solved before this technique becomes industrially feasible (Kohlmann et al., 2008; Paul et al., 2014; Zheng
et al., 2009).
Thermodynamics should be calculated between substrates and desired products, especially for the cases of non-natural biochemical reactions designed. An online calculator for biochemical thermodynamics
eQuilibrator is available (http://equilibrator.weizmann.ac.il/) (Jankowski
et al., 2008). Also, it is important to identify reduced power sources as
substrates, which may be carbohydrates, glycerol, H2, CO, methane, methanol, electricity and so on. For example, butanol production based on glucose is spontaneous (i.e., delta G b0) mainly due to a negative enthalpy
(Li et al., 2010). In contrast, methane generation from acetate occurs
spontaneously due to a gain of entropy but a positive enthalpy (Thauer
et al., 1977; Zhang, 2011a). Synthetic pathways or special enzymatic reactions cannot be accomplished against thermodynamics laws
(Bornscheuer et al., 2012). To ensure driving the reaction towards the desired direction, Gibbs free energy under standard or custom conditions
needs to be checked. To drive the reactions towards desired products,
we could apply a pushpull strategy sometimes. In situ product removal
(for example, hydrogen removal by vacuum or gas ush (Martin del
Campo et al., 2013; Myung et al., 2014), butanol stripping (Smith et al.,
2010)) provides more pull forces, while increasing the pressure or concentration of gaseous substrates could generate more push forces for desired reactions (Keller et al., 2013; Rieckenberg et al., 2014; Zhang and
Huang, 2012).
Generally speaking, enzymes used in vitro are highly exchangeable
regardless of their resources (Zhang et al., 2010). However, a choice of
right enzymes is important sometimes. In the case of enzymatic transformation of cellulose to synthetic starch, only alpha-glucan phosphorylase from potato can synthesize amylose from cellobiose while the
other two bacterial alpha-glucan enzymes having the same enzyme catalog number cannot do (You et al., 2013) because potato alpha-glucan
phosphorylase equipped with a special cap on its catalytic domain has
a lower activation energy for the amylose synthesis reaction than bacterial enzymes (You et al., 2013).
2.2. Examples of biofuels and bioelectricity

The combined markets of hydrogen for hydrogen fuel cell vehicles,
liquid biofuels for jet planes, and bioelectricity for portable electronics
are expected to be trillions of US dollars annually. Production of biofuels
and bioelectricity from renewable carbohydrates could be economically
viable because carbohydrates are less costly in term of energy content
1471
compared to most fossil fuels (Zhang, 2009) and are expected to decline
annually.
2.2.1. Hydrogen
Natural and genetically-modied microorganisms cannot produce
hydrogen with a yield of more than 4 H2 per glucose, that is, the Thauer
limit (Thauer et al., 2008; Zhang, 2011a), although a theoretical yield is
12 H2 per glucose. Nature cannot evolve such high-yield hydrogen generation pathways due to two reasons. First, theoretical yield of hydrogen
production is an endothermic reaction so that it cannot co-generate
ATP. Second, if a small fraction of reduced NAD(P)H was used to generate ATP via oxidative phosphorylation (Swartz, 2013), the presence of
oxygen would inhibit oxygen-sensitive hydrogenase activity greatly.
Woodward and his coworkers (Woodward et al., 2000) produced
nearly 12 H2 from costly glucose 6-phosphate (G-6-P). This pathway
comprised three modules: (i) two NADPH generation from G-6-P mediated by two dehydrogenases (Eq. (1)), (ii) hydrogen generation from
NADPH mediated by hydrogenase (Eq. (2)), and (iii) re-generation of
G-6-P from ribulose 5-phosphate (Eq. (3)). However, costly substrate
G-6-P prevents its potential application so that Woodward did not le
a patent for this in vitro synthetic pathway.
G6P 2NADP
H2 O ribulose 5phosphate 2NADPH 2H CO2
2NADPH 2H 2H2 2NADP
6 Ribulose 5phosphate H2 O 5 G6P Pi
Furthermore, I proposed to generate glucose 1-phosphate (G-1-P)

from starch and phosphate catalyzed by alpha-glucan phosphorylase
(Eq. (4)) and then converted to G-6-P by phosphoglucomutase
(Eq. (5)).
C6 H10 O5 n Pi C6 H10 O5 n1 G1P
G1P G6P
As a result, the whole stoichiometric reaction can be written as

Eq. (6), that is, neither costly G-6-P nor ATP is needed for high-yield hydrogen production, suggesting a great potential of low-cost green hydrogen production from polysaccharides.
C6 H10 O5 aq 7H2 Ol 12H2 g 6CO2 g
The proof-of-concept experiment was conducted by using 13 enzymes from ve different sources (e.g., bacterium, yeast, plant, animal,
and archaea) (Zhang et al., 2007). Furthermore, oligosaccharides
(e.g., cellobiose, cellopentaose and sucrose) along with their respective
phosphorylases (e.g., cellobiose phosphorylase, cellodextrin phosphorylase, and sucrose phosphorlyase) can be used to produce high-yield hydrogen (Myung et al., 2014; Ye et al., 2009). However, one glucose unit
per mole of polysaccharide or oligosaccharide cannot be converted to G1-P. To completely utilize all hexose units, G-6-P can be produced from
low-cost polyphosphate instead of ATP mediated by polyphosphate glucokinase (Liao et al., 2012). Therefore, high-yield hydrogen can be produced from all hexose units of sucrose (Myung et al., 2014) and
glucose (Rollin et al., 2014). To utilize xylose, the most abundant pentose for the production of low-cost hydrogen, a novel polyphosphate
xylulokinase was discovered (Martin del Campo et al., 2013). By using
this enzyme plus polyphosphate, nearly 10 mol of hydrogen was produced from one xylose and water (Martn del Campo et al., 2013).
1472
Fig. 3. Scheme of in vitro synthetic enzymatic pathways for the production of high-yield hydrogen from a variety of carbohydrates starch, cellodextrin, sucrose, glucose, fructose and
xylose as well as water. The pathways are compiled and modied from References (Martn del Campo et al. (2013); Myung et al. (2014); Rollin et al. (2014); Ye et al. (2009); Zhang
et al. (2007)). The enzymes are GP, alpha-glucan phosphorylase; CDP, cellodextrin phosphorylase; CBP, cellobiose phosphorylase; SP, sucrose phosphorlyase; GI, glucose isomerase;
XI, xylose isomerase; PPGK, polyphosphate glucokinase; PPXK, polyphosphate xylulokinase; PGM, phosphoglucomutase; G6PDH, glucose 6-phosphate dehydrogenase; 6PGDH, 6phosphogluconate dehydrogenase; RPI, ribose 5-phosphate isomerase; RPE, ribulose-5-phosphate 3-epimerase; TK, transketolase; TAL, transaldolase; TIM, triose phosphate isomerase;
ALD, (fructose-bisphosphate) aldolase; FBP, fructose bisphosphatase; PGI, phosphoglucose isomerase; and H2ase. Pi and (Pi)n are inorganic phosphate and polyphosphate with a degree
of polymerization of n. The metabolites are: g1p, glucose-1-phosphate; g6p, glucose-6-phosphate; ru5p, ribulose 5-phosphate; x5p, xylulose 5-phosphate; r5p, ribose 5-phosphate; s7p,
sedoheptulose 7-phosphate; g3p, glyceraldehyde 3-phosphate; e4p, erythrose 4-phosphate; dhap, dihydroxacetone phosphate; fdp, fructose-1,6-diphosphate; f6p, fructose 6-phosphate.
The overall reaction (Eq. (6)) is spontaneous and endothermic

(i.e., Go = ~ 50 kJ/mol and Ho = + 598 kJ/mol) (Zhang et al.,
2007). This entropy-driven reaction can generate the chemical energy
output in the form of hydrogen more than the chemical energy of polysaccharides by absorbing low temperature waste heat, which is widely
available from refrigerators, air conditioners, cooling towers, and fuel
cells. These overall reactions are spontaneous due to the phase changes
from more orderly liquid to less orderly gas, resulting in a large positive
S (Ye et al., 2009; Zhang et al., 2007).
High-yield enzymatic hydrogen production from a variety of carbohydrates from starch, cellodextrins, sucrose, glucose, fructose, and xylose (Fig. 3) brings multiple implications: (i) low-cost green hydrogen
production for hydrogen fuel cell vehicles, (ii) a new hydrogen carrier
carbohydrate for on-demand hydrogen production, and (iii) an outside solution to the costly infrastructure of the hydrogen economy
(Zhang, 2009; Zhang, 2010b; Zhang et al., 2013).
2.2.2. n-Butanol
n-Butanol, a primary 4-carbon alcohol, is a better liquid biofuel than
ethanol due to higher energy density and less water adsorption. It used
to be produced by acetonebutanolethanol (ABE) fermentation using
Clostridium acetobutylicum. However, its fermentation involves a complicated transition from acidogenesis to solvogenesis and suffers from
low product yields and severe product inhibition, resulting in low product titers (Ezeji et al., 2007).
Honda and his coworkers designed an in vitro non-natural, ATPbalanced pathway for n-butanol production from glucose (Krutsakorn
et al., 2013a). This pathway comprised 16 thermostable enzymes
(Fig. 4). The pathway has three modules: (i) generation of two pyruvate
and two NADH from one glucose without ATP accumulation (Eq. (7),
different from glycolysis), (ii) generation of acetyl-CoA from pyruvate
(Eq. (8)); and (iii) n-butanol production from 2 acetyl-CoA (Eq. (9)).
As a result, one glucose can produce one n-butanol, two CO2 and one
water (Eq. (10)).
C6 H12 O6 glucose 2NAD
2 pyruvate 2 NADH 2H
2 pyruvate 2 CoA 2 acetylCoA 2 CO2
2 acetylCoA 2 NADH C4 H10 O nbutanol H2 O 2 CoA 2NAD
C6 H12 O6 glucose C4 H10 O nbutanol 2 CO2 H2 O
10
1473
et al., 2008), Bacillus subtilis (Li et al., 2011), Corynebacterium

glutamicum (Smith et al., 2010), Saccharomyces cerevisiae (Chen et al.,
2011), and C. cellulolyticum (Higashide et al., 2011). However, its hydrophobic effect is toxic to cellular membranes, resulting in low isobutanol
titers (e.g., 12%) (Atsumi et al., 2010).
Sieber and his coworkers designed an in vitro non-natural, ATP- and
CoA-free pathway for isobutanol production from glucose (Fig. 5)
(Guterl et al., 2012). This synthetic pathway contains two modules:
(i) generation of two pyruvate from glucose without involvement of
ATP and CoA mediated by four enzymes (Eq. (11)), and (ii) production
of one isobutanol from two pyruvate (Eq. (12)). As a result, one glucose
can produce one isobutanol, two CO2 and one water (Eq. (13)).
C6 H12 O6 glucose 2NAD
11
12
C4 H10 O isobutanol 2 CO2 H2 O 2NAD
C6 H12 O6 glucose C4 H10 O isobutanol 2 CO2 H2 O
13
Fig. 4. Scheme of the in vitro synthetic enzymatic pathway for the production of n-butanol
from glucose, modied from Ref. (Krutsakorn et al., 2013a). The enzymes are HK,
hexokinase; PGI, phosphoglucose isomerase; PFK, 6-phosphofructokinase; FBA, fructosebisphosphate aldolase; TIM, triosephosphate isomerase; GAPN, non-phosphorylating
GAP dehydrogenase; PGM, cofactor-independent phosphoglycerate mutase; ENO,
enolase; PK, pyruvate kinase; PDC, pyruvate decarboxylase; ADDH, CoA-acylating
aldehyde dehydrogenase; HBD, hydroxybutyryl-CoA dehydrogenase; HPD, 3hydroxypropionyl-CoA dehydratase; NFO, NADH-dependent avinoxidoreductase;
and HAD, 3-hydroxyacyl-CoA dehydrogenase. Metabolites are g6p, glucose-6-phosphate; f6p, fructose 6-phosphate; fdp, fructose-1,6-diphosphate; g3p, glyceraldehyde 3-phosphate; dhap, dihydroxacetone phosphate; 3pg, 3-phosphoglycerate;
2pg, 2-phosphoglycerate; and pep, phosphoenolpyruvate.
This synthetic pathway has three features: (i) ATP balance, where
the ATP consumption during the conversion of glucose to fructose-1,6bisphosphate matches the ATP regeneration from phosphoenolpyruvate to pyruvate mediated by pyruvate kinase; (ii) NADH balance,
where NADH regeneration by non-phosphorylating GAP dehydrogenase and CoA-acylating aldehyde dehydrogenase (ADDH) matches its
consumption by 3-hydroxyacyl-CoA dehydrogenase, ADDH, NADHdependent avinoxidoreductase, and hydroxybutyryl-CoA dehydrogenase; and (iii) CoA balance, where CoA is needed by ADDH and is released by acetyl-CoA acetyltransferase and ADDH. The overall reaction
is an enthalpy-driven reaction, meaning a slight chemical energy loss
(Huang and Zhang, 2011). By compromising reaction temperatures
and optimizing enzyme loading, the maximum n-butanol yield was
82% (Krutsakorn et al., 2013a), comparable with the best product yields
in ABE fermentation (Ezeji et al., 2007).
2.2.3. Isobutanol
Isobutanol, a branched C4 alcohol, has a similar energy density to nbutanol, limited miscibility with water, and is completely miscible with
gasoline (Peralta-Yahya et al., 2012). Isobutanol's branching gives it a
better octane number than n-butanol. Isobutanol can be produced
through the Ehrlich pathway, also called the 2-keto-acid pathway
(Hazelwood et al., 2008; Peralta-Yahya et al., 2012). Heterologous pathway for isobutanol production from carbohydrates have been introduced to a number of microorganisms, such as Escherichia coli (Atsumi
Fig. 5. Scheme of the in vitro ATP-free synthetic enzymatic pathway for the production of
isobutanol from glucose, modied from Ref. (Guterl et al., 2012). The enzymes are GDH,
glucose dehydrogenase; DHAD, dihydroxy acid dehydratase; AIDH, glyceraldehyde dehydrogenase; KDGA, 2-keto-3-desoxygluconate aldolase; ALS, acetolactate synthase; KARI,
ketolacid reductoisomerase; KDC, 2-ketoacid decarboxylase; and ALD, alcohol
dehydrogenase.
1474
As compared to the in vitro n-butanol pathway, this pathway requires eight enzymes without ATP and CoA. The proof-of-concept
one-pot experiment achieved a 53% isobutanol yield (Guterl et al.,
2012). What is more important, this in vitro system can produce
isobutanol even in the presence of 4% isobutanol (Guterl et al.,
2012), suggesting that enzyme cocktails can tolerate organic solvents far better than cellular membranes even at elevated temperatures (Zhang, 2010a).
2.2.4. Bioelectricity
Enzymatic fuel cells (EFCs) are an emerging electrobiochemical device that directly converts chemical energy of a variety of fuels to electricity through enzymes (Calabrese Barton et al., 2004; Moehlenbrock
and Minteer, 2008; Zhu et al., 2011). Inspired by living cells that can utilize complex organic compounds (e.g., starch, glycogen) as energy
sources, sugar-powered EFCs could be the next generation of biodegradable, highly safe, metal-free biobatteries. Compared to microbial
fuel cells, EFCs usually generate much higher power density in terms
of mW cm2 of membrane or electrode, suggesting their great potentials in powering portable electronics (Cooney et al., 2008; Sakai et al.,
2009; Zebda et al., 2011).
To release all electron potentials (i.e., 24) from each glucose unit,
a non-natural enzymatic pathway containing 13 enzymes is designed (Fig. 6). This synthetic pathway contains four modules:
(i) G-6-P generation from maltodextrin mediated by alpha-glucan
phosphorylase and phosphoglucomutase (Eq. (14)); (ii) 2 NADH
generated from G-6-P mediated by two NAD-dependent G-6-P
dehydrogenase (G6PDH) and 6-phosphogluconate dehydrogenase
(6PGDH) (Eq. (15)); (iii) NADH electro-oxidation through diaphorase (DI) to VK3 that can generate 2 electrons per NADH (Eq. (16));
and (iv) 5/6 mol of G-6-P regeneration from one mol of ribulose-5phosphate through a hybrid pathway comprised enzymes in the
pentose phosphate, glycolysis, and gluconeogenesis pathways
(Eq. (17)). As a result, each glucose unit of maltodextrin can generate
24 electrons on the anode through this de novo pathway (Eq. (18)).
This pathway utilizes two NAD-dependent G6PDH and 6PGDH that
can generate NADH in a different way from that of natural NADPdependent enzymes in the typical pentose phosphate pathway
used in anabolism. The above pathway does require neither ATP
nor CoA, which are too costly and not stable for EFCs (Sokic-Lazic
and Minteer, 2008).
C6 H10 O5 n Pi G6P C6 H10 O5 n1
G6P H2 O 2NAD
14
ribulose5phosphate CO2 2NADH 2H
15
NADH H 2H 2e anode
16
6 ribulose5phosphate H2 O 5 G6P Pi
17
C6 H10 O5 7H2 O 24e 6CO2 24H anode compartment

18
Nearly 24 electrons are produced per glucose unit of maltodextrin
via this pathway in an air-breathing EFC (Zhu et al., 2014b) and its Coulombic efciency is higher than that of the best microbial fuel cell
(Chaudhuri and Lovley, 2003). This EFC also exhibited a maximum
power output of 0.8 mW cm 2 and a maximum current density of
6 mA cm2 (Zhu et al., 2014b), faster than those of the best microbial
fuel cells (Logan, 2009). EFCs containing a 15% (wt./v.) maltodextrin solution have an energy storage density of 596 Ah kg 1, which is one
order of magnitude higher than that of lithium-ion batteries. Sugarpowered biobatteries could serve as next-generation green power
sources, particularly for portable electronics (Zhu et al., 2014b).
2.3. Examples of biochemicals
The production of biochemicals from renewable carbohydrates is of
importance to not only decrease reliance on fossil fuels but also increase
economics of next-generation cellulosic bioreneries, although the
market size of biochemicals is at least one order of magnitude lower
than that of biofuels.
Fig. 6. Scheme of the in vitro ATP-free synthetic enzymatic pathway for the production of electricity from maltodextrin, modied from Ref. (Zhu et al., 2014b). Most enzymes and metabolites are the same as them in Fig. 2, except the enzyme DI, diaphorase.
2.3.1. Synthetic amylose

Synthetic amylose, a linear glucan linked by alpha-1,4-glycosidic
bonds, could be a very important compound as a precursor of lowoxygen diffusion biodegradable plastic lms, a promising health food
additive, and a potential high-density hydrogen carrier (Qi et al.,
2014; You et al., 2013). Cellulose, a linear glucan linked by beta-1,4-glycosidic bonds, is the supporting material of plant cell walls. The annual
cellulose resource is approximately 40 times the starch produced by cultivated crops as food and feed (You et al., 2013). In addition, (perennial)
cellulosic plants and dedicated bioenergy crops can grow on low-quality
or marginal land, and require less input such as fertilizers, herbicides,
pesticides, and water (Zhang, 2013).
An enzymatic pathway is designed to convert cellulose to amylose
without involvement of any coenzymes (You et al., 2013) (Fig. 7). This
pathway includes three modules: (i) partial hydrolysis of cellulose to
cellobiose by endoglucanase and cellobiohydrolase (Eq. (19)), (ii) generation of amylose and glucose from cellobiose catalyzed by cellobiose
phosphorylase (CBP) and potato alpha-glucan phosphorylase (PGP)
(Eq. (20)), and (iii) selective glucose removal by a yeast (Eq. (21)). As
a result, half of a cellulose could be used to produce synthetic amylose
and the other could be used to produce ethanol (Eq. (22)). Thanks to
relatively low yields of cellobiose from cellulose, it leads to a lower amylose yield less than 0.5 g amylose per g of cellulose (You et al., 2013).
C6 H10 O5 n cellulose n=2 H2 O n=2 C12 H22 O11 cellobiose 19
n C12 H22 O11 cellobiose n glucose C6 H10 O5 n amylose
20
glucose 2 C2 H6 O ethanol 2 CO2
21
C6 H10 O5 n cellulose C6 H10 O5 n=2 amylose n C2 H6 O ethanol n CO2
1475
fermentation in second generation cellulosic bioreneries, where

beta-glucosidase is replaced with CBP/PGP. In this process, all glucan
units in cellulose can be utilized to produce amylose, ethanol, and
yeast without sugar losses. Therefore, this process has a great potential
for future commercialization after solving enzyme costs and enzyme
stability. The cost-effective transformation of non-food cellulose to
synthetic starch could revolutionize agriculture and reshape the
bioeconomy, while maintaining biodiversity, minimizing agriculture's
environmental footprint, and conserving fresh water (Somerville et al.,
2010; Zhang, 2013). This transformation would not only promote the
cultivation of plants chosen for rapid growth rather than those optimized for starch-rich seed production, but it would also efciently utilize marginal land for the production of the biomass required to meet
the increasing needs of biofuels and biochemicals (Casillas and
Kammen, 2010; Sheppard et al., 2011; Zhang, 2013).
2.3.2. Lactate
Lactate is the most naturally-occurring hydroxycarboxylic acid
(Datta and Henry, 2006). It has an exploding application as the precursor
of biodegradable plastic polylactic acid (PLA) (Datta and Henry, 2006;
Joglekar et al., 2006; Zhang et al., 2011a). High yield and high-titer lactate is usually fermented from carbohydrates by anaerobic fermentation,
where one glucose can generate two lactate along with two ATP, which
supports cell duplication and basic metabolism of whole cells. However,
its separation is relatively costly because of impurities, such as residual
sugars, nutrients, and other organic acids in broths, the consumption
of sulfuric acid and alkali, and the generation of a side-product calcium
sulfate (Joglekar et al., 2006; Patnaik et al., 2002).
Honda and his coworkers demonstrated an alternative in vitro ATPbalanced enzymatic pathway that can convert glucose to lactate (Ye
et al., 2012a) (Fig. 8). This pathway comprised ten enzymes, containing
two modules: (i) generation of two pyruvate and two NADH from one
glucose (Eq. (7)), and (ii) production of two lactate from two pyruvate
22
Simultaneous enzymatic biotransformation and microbial fermentation can be regarded as modied simultaneous saccharication and
Fig. 7. Scheme of the in vitro ATP-free synthetic enzymatic pathway for the production of
synthetic amylose and ethanol from cellulose, modied from Ref. (You et al., 2013). The
enzymes are EG, endoglucanase; CBH, cellobiohydrolase; CBP, cellobiose phosphorylase;
and PGP, potato alpha-glucan phosphorylase.
Fig. 8. Scheme of the in vitro ATP-neutral synthetic enzymatic pathway for the production
of lactate from glucose, modied from Ref. (Ye et al., 2012a). Enzymes are HK, hexokinase;
PGI, g6p isomerase; PFK, 6-phosphofructokinase; FBA, fructose-bisphosphate aldolase;
TIM, triosephosphate isomerase; GAPN, non-phosphorylating GAP dehydrogenase; PGM,
cofactor-independent phosphoglycerate mutase; ENO, enolase; PK, pyruvate kinase; and
MLDH, malate/lactate dehydrogenase.
1476
catalyzed by lactate dehydrogenase (Eq. (23)). As a result, one glucose

can produce two lactates only (Eq. (24)).
2 pyruvate 2 NADH 2H 2 C3 H6 O3 lactate 2 NAD
23
C6 H12 O6 glucose 2 C3 H6 O3 lactate
24
By compromising reaction temperature and optimizing enzyme

loading, the maximum lactate yield reached 100% (Ye et al., 2012a).
This pathway could be further improved. For example, it is possible to
decrease the number of enzymes used from glucose to pyruvate in the
isobutanol pathway (Fig. 5) instead of nine enzymes here (Fig. 8).
Also, this change can avoid using ATP/ADP. Furthermore, in vitro enzymatic lactate production could be conducted at low pHs where enzymes
could tolerate low pHs better than whole cells to decrease separation
cost of lactate (Patnaik et al., 2002). Low pH fermentation or biotransformation will bring several benets: facilitating lactic acid precipitation
at its isoelectric point, decreasing the consumption of an alkali for neutralization of lactate and of sulfuric acid for regeneration of lactic acid,
and lessen the generation of gypsum waste.
2.3.3. 1,3-Propanediol
1,3-Propanediol (1,3-PDO) is a colorless viscous liquid, which can be
used as monomer in the production of polymers such as polytrimethylene
terephthalate.1,3-PDO can be produced from glucose (Nakamura and
Whited, 2003) or glycerol (Dietz and Zeng, 2014) by whole-cell fermentation. However, its yields based on glycerol are approximately 0.50
0.76 mol 1,3-PDO/mol glycerol (Celiska, 2010; Dietz and Zeng, 2014;
Zeng and Sabra, 2011) due to the synthesis of cell mass, the generation
of extra reducing power by branching glycerol utilization pathways
under micro-aerobic conditions, and the production of undesired sideproducts.
Zeng and his coworkers developed a straightforward in vitro pathway for converting glycerol to 1,3-PDO under strictly anaerobic condition (Fig. 9) (Rieckenberg et al., 2014). This pathway includes two
modules: (i) generation of 1,3-PDO from glycerol by consuming one
NADPH catalyzed by glycerol dehydratase and NADPH-dependent
propanediol oidoreductase-isoenzyme (Eq. (25)); and (ii) NADPH regeneration from H2 by hydrogenase (Eq. (26)). As a result, one mole
of 1,3-PDO was produced from one mole of glycerol and one mole of
H2 (Eq. (27)). Via this pathway, a very high 1,3-PDO yield of about
0.95 mol/mol has been accomplished (Rieckenberg et al., 2014).
glycerol NADPH H 1; 3PDO NADP H2 O
NADP H2 NADPH H
25
26
Glycerol H2 1; 3PDO H2 O
27
2.3.4. Poly-3-hydroxylbutyrate
Poly-3-hydroxylbutyrate (P3HB) consisting of 1000 to 30,000 hydroxy fatty acid monomers is the simplest and most commonly occurring form of polyhydroxyalkanoates, which can be either thermoplastic
or elastomeric biodegradable plastics. P3HAB is usually produced by microbial fermentation. Its synthesis is usually caused by certain deciency
conditions (e.g., a lack of phosphorous, nitrogen, trace elements, or oxygen) and the excess supply of carbon sources. The intracellular accumulated polyhydroxyalkanoates can be as high as 80% of the dry cell weight.
However, its purication is costly and its intracellular polymerization
synthesis is difcult to control (Gerngross and Slater, 2003).
Munekata and his coworkers (Satoh et al., 2003) demonstrate an
in vitro pathway to produce P3HB from acetate, where one extra
NADPH input is generated from glucose mediated by glucose dehydrogenase. The weight-average molecular weight and polydispersity of
in vitro synthesized P3HAB are 6.64 106 and 1.36, respectively
(Satoh et al., 2003). Recently, Bowie and his coworkers (Opgenorth
et al., 2014) design an in vitro pathway for the production of P3HB
from pyruvate without an external NAD(P)H input (Fig. 10). This
in vitro pathway contains three modules: (i) acetyl-CoA generation
from pyruvate, co-producing NADH and NADPH by using two NADPdependent and NAD-dependent pyruvate dehydrogenase, respectively
(Eq. (28)); (ii) addition of one unit of P3HB3 from 2 acetyl-CoA, accompanies with the consumption of NADPH (Eq. (29)); and (iii) oxidation of
NADH by using NADH oxidase (NOX). As a result, the whole system has
a reaction of Eq. (31) with balanced coenzymes. This delicate purge
valve system based on PDH-1/PDH-2/NOX relieves extra NADPH accumulation and allows the carbon ux to drive forward the desired product. In the optimized system, the maximum yield of P3HB was
approximately 94% (Opgenorth et al., 2014).
2 Pyruvate 2 CoA NAD NADP
28
2 acetylCoA NADPH NADH H 2CO2
2 acetylCoA NADPH H
P3HB 2 CoA NAPD
29
NADH H O2 NAD H2 O2
30
2 Pyruvate O2 P3HB 2 CO2 H2 O2
31
3. Economic analysis of the hydrogen case
Fig. 9. Scheme of the in vitro synthetic enzymatic pathway for the production of 1,3propanediol from glycerol, modied from Ref. (Rieckenberg et al., 2014). Enzymes are
GDH, glycerol dehydratase; PDORI, NADPH-dependent propanediol xoidoreductase-isoenzyme; and H2ase, hydrogenase. VB12, which is a key coenzyme for GDH, may be
repaired in vitro for a long time running.
The most appealing biomanufacturing feature of in vitro synthetic

biosystems may be nearly theoretical yield of a desired product. Here
we conduct a detailed economic analysis of hydrogen production from
carbohydrates. Hydrogen, one of the most important industrial compounds (i.e., annual market size of more than 100 billion US dollars),
is mainly produced from natural gas and coal. If hydrogen is costeffectively produced from carbohydrates by in vitro synthetic
biosystems, it is likely that others, more valuable biofuels and
Fig. 10. Scheme of the in vitro synthetic enzymatic pathway for the production of poly(R)-3-hydroxybutyrate from pyruvate, modied from Ref. (Opgenorth et al., 2014). Enzymes are PDH, pyruvate dehydrogenase; ACC, acetyl-CoA acetyltransferase; PhaB, 3hydroxybutryl-CoA reductase; PhaC, polyhydroxybutyrate synthase; and NOX, NADH
oxidase.
biochemicals, will also be amenable to cost-competitive production via

in vitro synthetic biosystems.
In vitro synthetic biosystems for biomanufacturing can be regarded
as a two-step biotransformation: (i) the production of numerous enzymes by whole-cell fermentation followed by their purication and
(ii) in vitro biotransformation mediated by enzyme cocktails (Zhang,
2011a). Therefore, it is very important to estimate a carbohydrate fraction of hydrogen production to overall sugar consumption for both production of enzymes and hydrogen. All enzymes except hydrogenase
(Fig. 3) can be produced in E. coli under aerobic conditions (Myung
et al., 2014), where recombinant enzyme yields (YrE) could range
from 0.05 to 0.1 kg protein/kg glucose (Huang and Zhang, 2011;
Wang et al., 2011; Zhang et al., 2010). In contrast, recombinant active
hydrogenase must be produced under anaerobic conditions, and its
yields (YH2ase) could range from 0.001 to 0.01 kg hydrogenase/kg glucose (Chandrayan et al., 2012; Sun et al., 2010).
Fig. 11a shows the effects of enzyme stability (i.e., total turn-over
number, TTN) on the carbohydrate fraction of hydrogen production to
overall sugar consumption. When enzyme TTN values are low
(e.g., 1035), most carbohydrate is used to produce enzymes instead of
the desired product hydrogen. When enzyme TTN values are more
than 107 regardless of recombinant protein yields, more than 99% carbohydrate is used to produce hydrogen. This analysis suggests that enhancing enzyme TTN values is the rst priority for high-yield
production of hydrogen and high-product yields are possible even considering carbohydrate consumption for enzyme production.
The overall production cost of hydrogen from carbohydrates before
tax include (i) feedstock (mainly, carbohydrates) used for enzyme
1477
synthesis and hydrogen production, (ii) biocatalysts enzymes, (iii) coenzymes, and (iv) capital investment and processing costs. Three key
inputs/assumptions are (i) carbohydrate price = $0.30/kg glucose
equivalent (or 6 * CH2O), (ii) coenzyme costs are $4500/kg NADP,
$1500/kg NAD, and $50/kg benzyl nicotinamide (BNA); and (iii) enzyme
costs are $20/kg recombinant protein and $100/kg hydrogenase (Rollin
et al., 2013).
Fig. 11b shows the effects of enzyme stability on hydrogen production
costs, when TTN of coenzyme is 106. Clearly, hydrogen production
costs decrease exponentially from more than $100,000 to several US
dollars/kg H2 where TTN of all enzymes increases from 103 to 108, suggesting that enhancing enzyme stability is the most important factor
for low-cost biomanufacturing. When TTN values of enzymes are more
than 109, further improvement of enzyme stability has nearly no impact
on lowering hydrogen production costs. Similar to this case, no R&D efforts are being made to improve the stability of industrial glucose isomerase. When enzymes are stable enough (e.g., TTN = 109), the cost of
coenzyme greatly determines nal hydrogen production cost
(e.g., $8.43/kg H2 using NADP, $4.64/kg H2 using NAD, and $2.61/kg H2
using BNA), suggesting the importance of the replacement of high-cost
and labile natural coenzymes with low-cost and stable biomimetic coenzymes (Paul et al., 2014; Rollin et al., 2013).
Fig. 11c shows the effects of coenzymes' TTN and costs on hydrogen
production costs, when TTN of all enzymes is 109. Clearly, increasing
TTN of coenzymes exponentially decreases hydrogen production costs,
when TTNs are less than 105. When TTNs of coenzymes are larger
than 108, hydrogen production cost could be as low as $2.38/kg H2.
However, the highest TTN for NAD reported is 106 (Kazandjian and
Klibanov, 1985), it is essentially important to replace natural coenzymes
with biomimics (Rollin et al., 2013) by considering an increasing stability order of coenzymes from NADP to NAD to BNA (Paul et al., 2014).
Fig. 11d shows the effects of carbohydrate costs from $0.05/kg to
$0.5/kg on the overall hydrogen production cost, where the sum of
capital expenses and operation expense is $0.35/kg H2, overall enzyme cost is $0.13/kg H2 (i.e., TTN of enzymes = 109), and coenzyme
cost is $0.01/kg H2 (i.e., TTN of BNA = 108). It is noted that hydrogen
separation costs from the aqueous enzymatic solution are very low;
hydrogen and CO2 can be separated by membrane technology, pressure swing adsorption, or a hybrid of both (Zhang, 2009); highpurity hydrogen without CO can be used by proton exchange membrane fuel cells easily (Huang and Zhang, 2011). Fig. 11d shows
that the overall hydrogen production costs are proportional to carbohydrate cost. When carbohydrate cost is $0.2/kg and $0.3/kg, the
hydrogen production costs could be $1.99/kg and $2.74/kg of H2, respectively. These data are very cost-competitive with hydrogen produced from natural gas without considering two extra benets:
carbon-neutral hydrogen production and distributed hydrogen production systems based on scattered biomass resources. Also, the
fraction of carbohydrate cost to overall hydrogen production cost
ranges from 0.435 to 0.885 when carbohydrate costs increase from
$0.05/kg to $0.5/kg (Fig. 11d). These data are similar to those ratios
of feedstock to nal product in mature industrial processes, such as
vegetable oil to biodiesel, crude oil to gasoline and diesel, corn kernels to ethanol (Lynd et al., 1999; Zhang, 2008).
Distributed hydrogen production from local biomass carbohydrates
has multiple advantages over centered hydrogen production from fossil
fuels. High capital costs of fossil fuel-to-hydrogen result from its technologically complex, multistep process involving high-pressure and hightemperature gasication, and steam shifting and separation of hydrogen
from other products. Critical challenges associated with a lack of high
density hydrogen storage approaches, a lack of infrastructure for compressed hydrogen distribution, and the high cost of hydrogen compression lead to serious doubt about the feasibility of the hydrogen
economy. Beyond the low-cost and green hydrogen production, the
production of high-yield and high-purity hydrogen from local carbohydrate resources via in vitro synthetic biosystems is an out-of-the-box
1478
Fig. 11. Economic analysis of enzymatic hydrogen production from carbohydrates. (a) The carbohydrate allocation to hydrogen production relative to overall carbohydrate consumption in
terms of enzyme TTN. Y(rE) is the protein yield based on carbohydrate consumption by aerobic fermentation in terms of kg recombinant protein per kg of carbohydrate, Y(H2ase) is the
hydrogenase yield based on carbohydrate consumption by anaerobic fermentation in terms of kg recombinant protein per kg of carbohydrate. (b) Effects of enzyme TTN and coenzyme
costs. (c) Effects of coenzyme TTN and coenzyme type. (d) Effects of carbohydrate costs on nal hydrogen production costs.
solution to address numerous challenges of the hydrogen economy

(Zhang, 2009, 2010b, 2013; Zhang et al., 2013).
The above analysis suggests that it could be economically feasible to
produce cost-competitively hydrogen from carbohydrates and water
via in vitro synthetic biosystems. Similar economic analyses may be
conducted for the production of other biocommodities. In short,
in vitro assembly of puried, stabilized enzymes plus biomimetic coenzymes could have much lower production costs than microbial fermentations, because the former can work much longer, achieve higher
product yields, implement faster volumetric productivity, and facilitate
easier product separation, and so on (Zhang, 2011a).
4. Challenges and solutions
Large-scale implementation of in vitro synthetic biosystems for
biomanufacturing requires synergistic efforts in enzyme engineering
and discovery, bulk enzyme production and purication, bioinformatics,
synthetic biology, bioprocess engineering, organic chemistry, etc.
4.1. Stable enzymes as standardized building blocks
As discussed in the Economic analysis of the hydrogen case section,
extending the stability of enzymes as building blocks of in vitro synthetic
biosystems is essentially important to increase the carbohydrate allocation to the desired product and decrease production costs (Fig. 11a&b).
In general, three strategies can be conducted to meet the challenge of
enzyme stability: immobilization, use of enzymes from thermophilic
hosts (thermoenzymes), and enzyme engineering. Whenever possible,
a good starting point is to use thermoenzymes. Many examples exist

of enzyme discovery from hosts like Thermotoga maritima, Thermus
thermophilus, Clostridium thermocellum, Geobacillus spp. and so on.
Now several websites have provided good collections for putative enzyme sources, for example, the Kyoto Encyclopedia of Genes and
Genomes (KEGG) (http://www.genome.jp/kegg/). When a special enzyme is only available from a mesophilic source, or activity modication
is desired, the next tool often used is enzyme engineering, involving rational design and directed evolution or their combination. For example,
a large size enzyme cellobiose phosphorylase has been enhanced greatly
through rational design and directed evolution combined (Ye et al.,
2012b). All recombinant thermoenzymes used in the enzymatic hydrogen production have been produced using E. coli (Martn del Campo
et al., 2013; Myung et al., 2014).
Enzyme immobilization technologies have been developed for more
than half a century. A variety of techniques include physical adsorption,
covalent binding to support structures, cross-linked enzyme aggregates,
and so on (Zhang et al., 2011b). For example, a combination of thermostable enzyme and enzyme immobilization leads to TTN of phosphoglucose
isomerase more than 109 at 60 C (Myung et al., 2011). Another
immobilized thermostable glucose isomerase has a working lifetime
from half a year to two years at 55 C (Vasic-Racki, 2006). Recent advances in enzyme immobilization on nanomaterials (e.g., polymeric
nanogels, nanocrystals, nanomagetic particles) may preserve enzyme activity to a large degree, while simultaneously increasing stability and facilitating great mass transfer (Ge et al., 2012; Myung et al., 2013). Also,
immobilized enzymes can be easily recycled and separated from soluble
products/substrates if necessary.
4.2. Synthetic multiple-enzyme complexes (metabolons) as building modules

Natural biological systems have evolved numerous enzyme complexes or create microcompartments containing several cascade enzymes together (Jandt et al., 2013). Such structures promote specic
metabolic processes by encapsulating and colocalizing cascade enzymes
with their substrates and cofactors, by facilitating substrate channeling,
protecting vulnerable enzymes in dened microenvironments, and by
sequestering toxic, labile, or volatile intermediates (Kerfeld et al.,
2010; You et al., 2012; Zhang, 2011b).
The simplest way to obtain multi-function enzymes is the creation of
chimeric proteins, in which two or more cascade enzymes are combined
by a linker to form a multi-functional single polypeptide (Bulow et al.,
1985; Conrado et al., 2008). For example, Bulow et al. constructed a
bifunctional enzyme containing cascade reactions mediated by E. coli
-galactosidase and galactokinase (Bulow et al., 1985). This synthetic
fusion enzyme displayed kinetic advantages (1.52.4-fold) over free enzyme mixtures. Similar effects have been reported in other fusion enzymatic systems (Agapakis et al., 2010; Meynial-Salles et al., 2007; Orita
et al., 2007; Riedel and Bronnenmeier, 1998). However, the misfolding
of large multi-domain proteins often happens and decreases their apparent activities (Chang et al., 2005), resulting in a large uncertainty of
this strategy.
Scaffolding molecules, includes proteins and nucleic acids, can recruit enzymes to form multienzyme complexes (Aldaye et al., 2008;
You et al., 2012) (Wei et al., 2012). Inspired by cellulosomes (Bayer
et al., 1994), a synthetic scaffold containing different cohesins from different microorganisms is used for the assembly of three metabolic enzymes (You et al., 2012). Triosephosphate isomerase (TIM), aldolase
(ALD), and fructose 1,6-bisphosphatase (FBP) are engineered to have
a dockerin at their C-terminals. These three dockerin-containing enzymes can be self-assembled into a static trifunctional enzyme complex
through the interaction with a mini-scaffold protein consisting of three
different matching cohesins. The synthetic metabolon showed more
than one order of magnitude enhancements in reaction rates compared
to the non-complexed TIM, ALD, and FBP mixture (You et al., 2012). Furthermore, it is found that the annexation of more high-activity ALD in
the synthetic enzyme complexes drastically decreases the degree of
substrate channeling by ve times (You and Zhang, 2014). Either DNA
or RNA can be designed to fold into various structures in vitro, forming
simple structures such as sheets to very more complicated structures
such as tubes and capsules (Aldaye et al., 2008; Wei et al., 2012). It is
more convenient to in vitro synthesize oligonucleotide as DNA or RNA
scaffolds by using automate oligonucleotide synthesis machines than
protein scaffolds. However, the costs of DNA and RNA is far more expensive than that of protein scaffolds produced by whole cells, which may
impair their applications on a large scale, especially for the production
of biocommodities.
Co-immobilization of multiple enzymes is another very practical
choice. Enzyme components can be randomly distributed (Betancor
et al., 2006; El-Zahab et al., 2004), positionally assembled (El-Zahab
et al., 2004), and even the active site of an enzyme face to the one of another enzyme (Mansson et al., 1983) on solid supports. The enhanced
reaction rates among co-immobilization of cascade enzymes have
been observed for several systems (Mateo et al., 2006; Myung et al.,
2013; Zhang, 2011b), but direct cross linking could lead to loss of enzyme activity (Sheldon and van Pelt, 2013).
Regardless of fusion proteins, the use of scaffolds, and coimmobilization, the general rules of constructing such synthetic enzyme
complexes need further investigation for their optimal ratios and their
orientation.
4.3. Bulk enzyme production and separation
A wide range of bulk enzyme production methods have been explored in recent years (Demain and Vaishnav, 2009). Desired enzymes
1479
are produced in recombinant forms, using bacteria, fungal, yeast and

plant platforms, preferred due to in-depth knowledge of the organisms,
ease of genetic manipulation, and low production cost (Demain and
Vaishnav, 2009). For example, protease is produced at a cost of ~$10
per kg by B. subtilis, cellulase is produced at a cost of ~$10 per kg by
Trichoderma spp. and A. niger spp., and recombinant proteins produced
at costs of tens of dollars per kg in E. coli (Taylor et al., 2008; Zhang
and Zhang, 2010; Zhang et al., 2006). The DOE Biomass Program even
anticipated industrial enzyme production costs as low as $0.7 per kg of
dry enzyme in the future, equaling that of soybean protein (Rollin
et al., 2013). To increase recombinant protein expression levels, systematic efforts have been conducted by optimization of codon usage, expression plasmid and host, inducer type, concentration, addition time,
and so on. It resulted in ~500-fold enhancement of a hyperthermophilic
T. maritima 6-phosphogluconate dehydrogenase in E. coli (Wang and
Zhang, 2009b). The other option is overexpression of the target protein
in its native organism, such as hydrogenase by Pyroccus furiosus
(Chandrayan et al., 2012).
Instead of costly chromatographic separation techniques, several
low-cost, scalable enzyme purication approaches have been developed, for example, simple centrifugation for secretory enzymes
(e.g., cellulase, amylase, protease), one-step cellulose binding moduletagged enzyme purication and immobilization on low-cost cellulosic
materials (Liao et al., 2012; Myung et al., 2011), heat precipitation for
thermostable enzymes (Sun et al., 2012; Ye et al., 2012a), and ammonia
precipitation (Myung and Zhang, 2013). It is noted that the protein of
thermostable enzymes in mesophilic hosts can decrease the enzyme
purication cost because enzymes cloned from thermophilic microbes
can be stable at high temperature (6080 C). Therefore, heat treatment
can deactivate E. coli endogenous proteins; after centrifugation, only
soluble thermostable proteins remain in the supernatant. This heat precipitation provides a simple and less-costly purication method for numerous thermostable enzymes (Krutsakorn et al., 2013a,b; Sun et al.,
2012; Wang and Zhang, 2009b; Ye et al., 2012a). Honda and his coworkers attempted to coexpress nine thermoenzymes in one E. coli
cell (Ninh et al., 2014). After heat precipitation, they harvested the
thermoenzyme cocktail suitable for one-pot biotransformation (Ninh
et al., 2014).
Therefore, there are no clear technical obstacles to producing lowcost bulk enzymes suitable for in vitro synthetic biosystems for
biomanufacturing.
4.4. NAD(P)-dependent enzyme engineering
NAD and NADP are the predominant energy carriers for cellular metabolisms but they cannot be self-regenerated or self-repaired in vitro.
These coenzymes are too costly and easily degraded under some conditions, such as oxygen presence, low or high pH, high temperature, and
so on. Their costs allow their use for the synthesis of high-value chiral
compounds (Wildeman et al., 2007) but prevent them from the production of biocommodities (Fig. 11c). The best solution is the replacement
of NAD(P) with low-cost and high stability biomimetic cofactors
(mNADs) (Paul et al., 2014; Rollin et al., 2013). However, most wildtype NAD(P)-dependent redox enzymes cannot work on mNADs, except avin-containing monooxygenase (Lutz et al., 2004), enoate reductase (Paul et al., 2013), and diaphorase.
Coenzyme engineering from natural to biomimetic coenzymes remains in the early stage (Campbell et al., 2012; Paul et al., 2014; Rollin
et al., 2013). The three approaches are rational design, directed evolution, or swapping modules (Rollin et al., 2013). Clark and Fish demonstrate that an engineered P450 mutant with two amino acid changes
can utilize BNA (Ryan et al., 2008). Because of NAD(P)-dependent
redox enzymes that have highly conserved cofactor-binding motifs
Rossmann motif, the module swap strategy could be useful for coenzyme
specicity switch. For example, Yaoi et al. demonstrate the replacement
of the NADP-binding domain of an isocitrate dehydrogenase by an NAD-
1480
binding domain, resulting in a change of coenzyme preference (Yaoi

et al., 1996). The success of the use of biomimetic coenzymes along
with engineered NAD(P)-dependent redox enzymes could greatly expand enzyme-based biocatalysis, whose applications are not restricted
by reactions catalyzed by hydrolases and isomerases.
4.5. System optimization and modeling
In vitro synthetic biosystems can be modeled at multiple levels from
molecules to modules to systems (see a review (Jandt et al., 2013)). The
relative simplicity of in vitro biosystems compared to whole cells makes
them far easier to simulate processes and predict optimal enzyme ratios
for maximizing product yield and accelerating volumetric productivity.
Different from great challenges in assays of thousands of intracellular metabolites (Bennett et al., 2009), it is relatively easy to measure tens of metabolites in in vitro biosystems. Also, protein chips allow better pathway
construction and optimization in vitro (Jung and Stephanopoulos, 2004).
A nonlinear kinetic model was developed to simulate enzymatic hydrogen production and predict rate-limiting steps (Ye et al., 2009). This
model suggests that hydrogenase and two dehydrogenases were responsible for catalyzing rate-limiting steps when the same unit of enzymes was used (Ye et al., 2009). Later, Ardao and Zeng (Ardao and
Zeng, 2013) further improved this in silico model by using a genetic algorithm to solve a multi-objective optimization. As a result, when using
yield and rate as objective functions in this model, one set of solutions
predicted the ability for this system to retain 90% yield while increasing
the reaction rate to 355 mmol/L/h (Ardao and Zeng, 2013). Furthermore, parameter setting of this model was t with experimental data
under an elevated temperature of 50 C by using a genetic algorithm,
and then a global sensitivity analysis was used to identify the most important enzymes to determine reaction rate and yield improvements (in
revision for publication). Via an optimal set of enzyme concentrations
predicted by the model, volumetric productivity of hydrogen was increased to 32 mmol H2/L/h by more than 10-fold. The maximum productivity was further enhanced to 54 mmol H2/L/h by elevating
reaction temperature to 60 C and increasing substrate level. The reaction rate of 54 mmol of H2 per hour per liter equals 3.7 W/L, approximately 21-time the maximum algal photohydrogen rate. This
productivity is comparative to industrial biogas and hydrogen production (Argun and Kargi, 2011). In the past decade, enzymatic hydrogen
generation has been enhanced from 0.21 (Woodward et al., 2000) to
54 mmol of H2/L/h by a factor of more than 250. Further increases in volumetric productivity could decrease capital investment for bioreactors.
The most ambitious goal is the production of on-demand hydrogen
from carbohydrates and water in a small-size hydrogen fuel cell vehicle
(Zhang, 2009). It means that we would increase hydrogen productivity
to 510 mol H2/L/h, by another 100200-fold.
4.6. Stabilization of thermolabile metabolites at elevated temperatures
Elevating reaction temperature usually increases reaction rates but
may cause the degradation of substrates, products or intermediates.
High product yields of in vitro synthetic biosystems are usually accomplished at modest temperatures, when the degradation of metabolites
are minimal. High reaction temperatures may decrease product yields.
For example, high yield of lactate was obtained at a temperature of
50 C, while 60 C leads to the degradation of glyceraldehyde 3phosphate, dihydroxacetone phosphate, phosphoenolpyruvate, NADH,
etc. (Ye et al., 2012a). Similarly, nearly theoretical yields of hydrogen
were obtained at 3050 C (Martin del Campo et al., 2013; Myung
et al., 2014; Ye et al., 2009), but the yields were lower at 60 C (Rollin
et al., 2014).
Hyperthermophiles, whose optimal temperatures are more than
80 C, have evolved several mechanisms for protecting thermo-labile
small molecules metabolites and coenzymes: rapid turnover, substrate
channeling, and local stabilization (Daniel, 2000). For example, labile
NADPH must be recycled efciently for an extreme thermoacidophile

Picrophilus torridus though its halftime is 1.7 min at pH 4.5 and 65 C
(Angelov et al., 2005). To mimic the rapid turnover strategy in vitro, it
is suggested that the enzymes responsible for converting easilydegraded metabolites is over-added (Myung et al., 2010) or high catalytic efciency enzymes are used. Triosephosphate isomerase which is
responsible for interconversion of two thermolabile metabolites glyceraldehyde 3-phosphate and dihydroxacetone phosphate evolves to be
kinetically-perfect and keep their levels very low (Zhang, 2011b). Inspired to natural enzyme complexes tryptophan synthase (Hyde
et al., 1988), carbamoyl-phosphate synthetase (Thoden et al., 1997),
fatty acid synthase (Maier et al., 2008), the construction of synthetic enzyme complexes that channels labile metabolites among cascade enzymes is a powerful strategy (Cheng et al., 2008; Zhang, 2011b).
Because it is known that decreasing water activity or dehydration can
stabilize a large number of biomolecules, creating desolvating environments for labile metabolites, or binding them around respective enzymes could help decrease their degradation odds. For example, the
highest TTN of NAD was achieved in an organic solvent (Kazandjian
and Klibanov, 1985) and the stability of NAD can be enhanced by its immobilization or binding with its enzymes (Liu and Wang, 2007).
Hyperthermophiles generally accumulate very unusual compatible
solutes, such as di-myo-inositol-phosphate, di-glycerol-phosphate,
ectoines, which have not been identied in mesophiles (Lentzen and
Schwarz, 2006; Santos and da Costa, 2002). These solutes can protect
thermal degradation of cell components and enzymes (Lentzen and
Schwarz, 2006; Santos and da Costa, 2002). Therefore, it is speculated
that such compatible solutes may protect metabolites by decreasing
water activity. But this hypothesis needs further in vitro testing.
4.7. Scale-up by integrative innovation
The implementation of in vitro synthetic biosystems for
biomanufacturing is like making of iPhone by integrating numerous
well-known parts and creating a seemingly-new system. Recent advances in DNA sequencing, bioinformatics, recombinant DNA technology,
enzyme engineering and immobilization, as well as bulk enzyme production and purication allow developing highly efcient, tunable enzymes
tailored for large-scale industrial production (Bornscheuer et al., 2012).
The integration of numerous stable enzymes in one bioreactor that can
last a very long reaction time (e.g., several months or even years)
would become a disruptive technology for low-cost biomanufacturing.
The successful implementation of this game-changing platform requires a teamwork of academic researchers, industrial developers, and
even policy makers. To our limited knowledge, the only successful example is 100 L cell-free protein synthesis (Hodgman and Jewett,
2012). This case is possible because of relatively easy preparation of
the E. coli cell lysate instead of their purication and high-value products. It is believed that the success of the rst in vitro synthetic
biosystem for large-scale biomanufacturing in next several years will
erase doubt of its industrial feasibility.
4.8. Other issues
A few in vitro synthetic pathways have been investigated to help
understand in-depth natural pathways that may be used to produce
advanced biofuels and biochemicals in vivo. For example, Khosla and
his coworkers investigated the detailed kinetics of the fatty acid synthesis using the reconstitution of the puried E. coli fatty acid
synthase components (Yu et al., 2011). Liu and his co-workers investigated the synthesis of farnesene (a precursor of new jet fuel)
through the mevalonate pathway by using puried enzymes (Zhu
et al., 2014a). Adams and his coworkers demonstrated the feasibility
of production of 3-hydroxypropionic acid, one of the top 12 industrial chemical building blocks, from hydrogen and carbon dioxide by
using P. furiosus cell-free extract (Keller et al., 2013). However,
these in vitro synthetic biosystems have more challenges in future

biomanufacturing, such as NAD(P)H imbalance, ATP imbalance, or
CoA involvement. For example, when fatty acid ethyl ester is produced from glucose, this pathway always generates extra NAD(P)H
(Huang and Zhang, 2011). For microbial cells, they can get rid of
extra NADPH easily under aerobic conditions (Steen et al., 2010),
but leading to decreased energy conversion efciencies (Huang and
Zhang, 2011). To balance extra NADPH in vitro, the possible solutions
may co-produce hydrogen from NADPH via hydrogenase under anaerobic conditions, like the case of enzymatic hydrogen production
(Martin del Campo et al., 2013; Myung et al., 2014), or oxidize
NADPH oxidization via NADPH oxidase under aerobic conditions,
like the case of P3HB synthesis (Opgenorth et al., 2014). Another
challenge is the use of and stability of CoA. Similar to the case of
NAD(P), small-size biomimetic CoA molecules may be used to replace CoA. If possible, the design of CoA-free pathways may be the
rst choice as occurred in the synthetic isobutanol pathway (Fig. 5).
Conclusions
Different from whole cell biocatalysis that has been exploited for
thousands of years, in vitro synthetic biosystems is on their early stage.
The primary goal of any living entities is to duplicate themselves instead
of the production of desired products, while the primary goal of in vitro
synthetic biosystems is to produce desired products. Although in vitro
biosystems could not be used to produce a few mature industrial
products, such as ethanol, penicillin, bulk enzymes, living products
(e.g., stem cells and tissues), they could be a disruptive biomanufacturing
platform for the cost-competitive production of biocommodities due to
their unique biomanufacturing advantages (Table 1). Great potential
markets of biocommodities (e.g., up to trillions of dollars for biofuels, at
least tens of billions dollars for biochemicals) will motivate to address remaining obstacles pertaining to cost and stability of enzymes and coenzymes, low-cost production of numerous bulk enzymes, standardization
of building parts and modules, biomimetic coenzymes, biosystem optimization and scale-up, with next several decades.
Acknowledgments
This work was supported by the Shell GameChanger Program and
DOE ARPA-E PETRO to PZ as well as NSF STTR I (IIP - 1321528) to Cell
Free Bioinnovations Inc. In addition, funding for this work was provided
in part, by the Virginia Agricultural Experiment Station and the Hatch
Program of the National Institute of Food and Agriculture, U.S. Department of Agriculture.
References
Agapakis C, Ducat D, Boyle P, Wintermute E, Way J, Silver P. Insulation of a synthetic hydrogen metabolism circuit in bacteria. J Biol Eng 2010;4(1):3.
Aldaye FA, Palmer AL, Sleiman HF. Assembling materials with DNA as the guide. Science
2008;321(5897):17959.
Angelov A, Ftterer O, Valerius O, Braus GH, Liebl W. Properties of the recombinant glucose/galactose dehydrogenase from the extreme thermoacidophile, Picrophilus
torridus. FEBS J 2005;272(4):105462.
Ardao I, Zeng A-P. In silico evaluation of a complex multi-enzymatic system using one-pot
and modular approaches: application to the high-yield production of hydrogen from
a synthetic metabolic pathway. Chem Eng Sci 2013;87:18393.
Argun H, Kargi F. Bio-hydrogen production by different operational modes of dark and
photo-fermentation: an overview. Int J Hydrog Energy 2011;36(13):744359.
Arnold FH. Enzyme engineering reaches the boiling point. Proc Natl Acad Sci U S A 1998;
95:20356.
Atsumi S, Hanai T, Liao JC. Non-fermentative pathways for synthesis of branched-chain
higher alcohols as biofuels. Nature 2008;451(7174):869.
Atsumi S, Wu T-Y, Machado IMP, Huang W-C, Chen P-Y, Pellegrini M, et al. Evolution, genomic analysis, and reconstruction of isobutanol tolerance in Escherichia coli. Mol Syst
Biol 2010;6(1):449.
Bayer EA, Morag E, Lamed R. The cellulosomea treasure-trove for biotechnology. Trends
Biotechnol 1994;12:37986.
1481
Bennett BD, Kimball EH, Gao M, Osterhout R, Van Dien SJ, Rabinowitz JD. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat
Chem Biol 2009;5(8):5939.
Betancor L, Berne C, Luckarift HR, Spain JC. Coimmobilization of a redox enzyme and a cofactor regeneration system. Chem Commun 2006;34:36402.
Bogorad IW, Lin T-S, Liao JC. Synthetic non-oxidative glycolysis enables complete carbon
conservation. Nature 2013;502:6937.
Boltje TJ, Buskas T, Boons G-J. Opportunities and challenges in synthetic oligosaccharide
and glycoconjugate research. Nat Chem 2009;1(8):61122.
Bornscheuer UT, Huisman GW, Kazlauskas RJ, Lutz S, Moore JC, Robins K. Engineering the
third wave of biocatalysis. Nature 2012;485(7397):18594.
Bozic M, Pricelius S, Guebitz GM, Kokol V. Enzymatic reduction of complex redox dyes
using NADH-dependent reductase from Bacillus subtilis coupled with cofactor regeneration. Appl Microbiol Biotechnol 2010;85:56371.
Bruggink A, Schoevaart R, Kieboom T. Concepts of nature in organic synthesis: cascade catalysis and multistep conversions in concert. Org Process Res Dev 2003;
7(5):62240.
Buchner E. Alkoholische Grung ohne Hefezellen (Vorluge Mitteilung). Ber Dtsch Chem
Ges 1897;30:11724.
Bujara M, Schmperli M, Pellaux R, Heinemann M, Panke S. Optimization of a blueprint for
in vitro glycolysis by metabolic real-time analysis. Nat Chem Biol 2011;7(5):2717.
Bulow L, Ljungcrantz P, Mosbach K. Preparation of a soluble bifunctional enzyme by gene
fusion. Nat Biotechnol 1985;3(9):8213.
Calabrese Barton S, Gallaway J, Atanassov P. Enzymatic biofuel bells for implantable and
microscale devices. Chem Rev 2004;104(10):486786.
Campbell E, Meredith M, Minteer SD, Banta S. Enzymatic biofuel cells utilizing a biomimetic cofactor. Chem Commun 2012;48(13):1898900.
Carlson ED, Gan R, Hodgman CE, Jewett MC. Cell-free protein synthesis: applications
come of age. Biotechnol Adv 2012;30:118594.
Casillas CE, Kammen DM. The energypovertyclimate nexus. Science 2010;330(6008):
11812.
Celiska E. Debottlenecking the 1,3-propanediol pathway by metabolic engineering.
Biotechnol Adv 2010;28(4):51930.
Chandrayan SK, McTernan PM, Hopkins RC, Sun JS, Jenney FE, Adams MWW. Engineering
hyperthermophilic archaeon Pyrococcus furiosus to overproduce its cytoplasmic NiFehydrogenase. J Biol Chem 2012;287(5):325764.
Chang H-C, Kaiser CM, Hartl FU, Barral JM. De novo folding of GFP fusion proteins: high
efciency in eukaryotes but not in bacteria. J Mol Biol 2005;353(2):397409.
Chaudhuri SK, Lovley DR. Electricity generation by direct oxidation of glucose in
mediatorless microbial fuel cells. Nat Biotechnol 2003;21(10):122932.
Chen K, Arnold FH. Turning the activity of an enzyme for unusual environments: sequential random mutagenesis of subtilisin E for catalysis in dimethylformamide. Proc Natl
Acad Sci U S A 1993;90:561822.
Chen X, Nielsen K, Borodina I, Kielland-Brandt M, Karhumaa K. Increased isobutanol production in Saccharomyces cerevisiae by overexpression of genes in valine metabolism.
Biotechnol Biofuels 2011;4(1):21.
Cheng Y, Chang C-eA YuZ, Zhang Y, Sun M, Leyh TS, Holst MJ, et al. Diffusional channeling
in the sulfate-activating complex: combined continuum modeling and coarsegrained Brownian dynamics studies. Biophys J 2008;95(10):465967.
Conrado RJ, Varner JD, DeLisa MP. Engineering the spatial organization of metabolic enzymes: mimicking nature's synergy. Curr Opin Biotechnol 2008;19(5):4929.
Cooney MJ, Svoboda V, Lau C, Martin G, Minteer SD. Enzyme catalysed biofuel cells. Energy Environ Sci 2008;1:32037.
Daniel RM. Biomolecular stability and life at high temperatures. Cell Mol Life Sci 2000;
57(2):25064.
Datta R, Henry M. Lactic acid: recent advances in products, processes and technologies
a review. J Chem Technol Biotechnol 2006;81(7):111929.
Demain AL. Pickles, pectin, and penicillin. Annu Rev Microbiol 2004;58:142.
Demain AL, Vaishnav P. Production of recombinant proteins by microbes and higher organisms. Biotechnol Adv 2009;27(3):297306.
Dietz D, Zeng A-P. Efcient production of 1,3-propanediol from fermentation of crude
glycerol with mixed cultures in a simple medium. Bioprocess Biosyst Eng 2014;
37(2):22533.
El-Zahab B, Jia H, Wang P. Enabling multienzyme biocatalysis using nanoporous materials.
Biotechnol Bioeng 2004;87(2):17883.
Endo T, Koizumi S. Large-scale production of oligosaccharides using engineered bacteria.
Curr Opin Struct Biol 2000;10(5):53641.
Ezeji TC, Qureshi N, Blaschek HP. Bioproduction of butanol from biomass: from genes to
bioreactors. Curr Opin Biotechnol 2007;18(3):2207.
Fessner W-D. Enzyme mediated CC bond formation. Curr Opin Chem Biol 1998;2(1):8597.
Fessner W-D, Helaine V. Biocatalytic synthesis of hydroxylated natural products using
aldolases and related enzymes. Curr Opin Biotechnol 2001;12(6):57486.
Franke D, Machajewski T, Hsu C-C, Wong C-H. One-pot synthesis of L-fructose using
coupled multienzyme systems based on rhamnulose-1-phosphate aldolase. J Org
Chem 2003;68(17):682831.
Ge J, Lei J, Zare RN. Protein-inorganic hybrid nanoowers. Nat Nanotechnol 2012;7(7):
42832.
Gerngross T, Slater S. Biopolymers and the environment. Science 2003;299:8225.
Goerke AR, Swartz JR. Development of cell-free protein synthesis platforms for disulde
bonded proteins. Biotechnol Bioeng 2008;99(2):35167.
Guterl J-K, Garbe D, Carsten J, Stefer F, Sommer B, Reie S, et al. Cell-free metabolic engineering production of chemicals via minimized reaction cascades. ChemSusChem
2012;5:216572.
Hazelwood LA, Daran J-M, van Maris AJA, Pronk JT, Dickinson JR. The Ehrlich pathway for
fusel alcohol production: a century of research on Saccharomyces cerevisiae metabolism. Appl Environ Microbiol 2008;74(8):225966.
1482
Higashide W, Li Y, Yang Y, Liao JC. Metabolic engineering of Clostridium cellulolyticum for

production of isobutanol from cellulose. Appl Environ Microbiol 2011;77(8):272733.
Hockenberry A, Jewett MC. Synthetic in vitro circuits. Curr Opin Chem Biol 2012;16:2539.
Hodgman CE, Jewett MC. Cell-free synthetic biology: thinking outside the cell. Metab Eng
2012;14:2619.
Honda Y, Kitaoka M, Hayashi K. Reaction mechanism of chitobiose phosphorylase from
Vibrio proteolyticus: identication of family 36 glycosyltransferase in Vibrio. Biochem
J 2004;377(1):22532.
Huang WD, Zhang Y-HP. Analysis of biofuels production from sugar based on three
criteria: thermodynamics, bioenergetics, and product separation. Energy Environ
Sci 2011;4:78492.
Huang K-T, Wu B-C, Lin C-C, Luo S-C, Chen C, Wong C-H, et al. Multi-enzyme one-pot
strategy for the synthesis of sialyl Lewis X-containing PSGL-1 glycopeptide.
Carbohydr Res 2006;341(12):21515.
Huisman GW, Liang J, Krebber A. Practical chiral alcohol manufacture using
ketoreductases. Curr Opin Chem Biol 2010;14(2):1229.
Hyde CC, Ahmed SA, Padlan EA, Miles EW, Davies DR. Three-dimensional structure of the
tryptophan synthase alpha 2 beta 2 multienzyme complex from Salmonella
typhimurium. J Biol Chem 1988;263(33):1785771.
Jandt U, You C, Zhang Y-HP, Zeng AP. Compartmentalization and metabolic channeling for
multienzymatic biosynthesis: practical strategies and modeling approaches. Adv
Biochem Eng Biotechnol 2013;137:125.
Jankowski MD, Henry CS, Broadbelt LJ, Hatzimanikatis V. Group contribution method for
thermodynamic analysis of complex metabolic networks. Biophys J 2008;95(3):
148799.
Jewett MC, Calhoun KA, Voloshin A, Wuu JJ, Swartz JR. An integrated cell-free metabolic
platform for protein production and synthetic biology. Mol Syst Biol 2008;4:57.
Joglekar HG, Rahman I, Babu S, Kulkarni BD, Joshi A. Comparative assessment of downstream processing options for lactic acid. Sep Purif Technol 2006;52(1):117.
Johannes TW, Woodyer RD, Zhao H. Efcient regeneration of NADPH using an engineered
phosphite dehydrogenase. Biotechnol Bioeng 2007;96(1):1826.
Jung GY, Stephanopoulos G. A functional protein chip for pathway optimization and
in vitro metabolic engineering. Science 2004;304(5669):42831.
Kanter G, Yang J, Voloshin A, Levy S, Swartz JR, Levy R. Cell-free production of scFv fusion
proteins: an efcient approach for personalized lymphoma vaccines. Blood 2007;
109(8):33939.
Katz E, Privman V. Enzyme-based logic systems for information processing. Chem Soc Rev
2010;39(5):183557.
Kazandjian R, Klibanov A. Regioselective oxidation of phenols catalyzed by polyphenol
oxidase in chloroform. J Am Chem Soc 1985;107:544850.
Keller MW, Schut GJ, Lipscomb GL, Menon AL, Iwuchukwu IJ, Leuko TT, et al. Exploiting
microbial hyperthermophilicity to produce an industrial chemical, using hydrogen
and carbon dioxide. Proc Natl Acad Sci U S A 2013;110:58405.
Kerfeld CA, Heinhorst S, Cannon GC. Bacterial microcompartments. Annu Rev Microbiol
2010;64(1):391408.
Kim J, Winfree E. Synthetic in vitro transcriptional oscillators. Mol Syst Biol 2011;7:115.
Kohlmann C, Mrkle W, Ltz S. Electroenzymatic synthesis. J Mol Catal B: Enzym 2008;
51(34):5772.
Korman TP, Sahachartsiri B, Li D, Vinokur JM, Eisenberg D, Bowie JU. A synthetic biochemistry system for the in vitro production of isoprene from glycolysis intermediates.
Protein Sci 2014;25:57685.
Kragl U, Eckstein M, Kaftzik N. Enzyme catalysis in ionic liquids. Curr Opin Biotechnol
2002;13(6):56571.
Krutsakorn B, Honda K, Ye X, Imagawa T, Bei X, Okano K, et al. In vitro production of nbutanol from glucose. Metab Eng 2013a;20:8491.
Krutsakorn B, Imagawa T, Honda K, Okano K, Ohtake H. Construction of an in vitro
bypassed pyruvate decarboxylation pathway using thermostable enzyme modules and its application to N-acetylglutamate production. Microb Cell Fact
2013b;12(1):91.
Lentzen G, Schwarz T. Extremolytes: natural compounds from extremophiles for versatile
applications. Appl Microbiol Biotechnol 2006;72(4):62334.
Li H, Cann AF, Liao JC. Biofuels: biomolecular engineering fundamentals and advances.
Annu Rev Chem Biomol Eng 2010;1(1):1936.
Li S, Wen J, Jia X. Engineering Bacillus subtilis for isobutanol production by heterologous
Ehrlich pathway construction and the biosynthetic 2-ketoisovalerate precursor pathway overexpression. Appl Microbiol Biotechnol 2011:113.
Liao HH, Myung S, Zhang Y-HP. One-step purication and immobilization of thermophilic
polyphosphate glucokinase from Thermobida fusca YX: glucose-6-phosphate generation without ATP. Appl Microbiol Biotechnol 2012;93:110917.
Liu W, Wang P. Cofactor regeneration for sustainable enzymatic biosynthesis. Biotechnol
Adv 2007;25(4):36984.
Logan BE. Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev Microbiol
2009;7(5):37581.
Lutz J, Hollmann F, Ho TV, Schnyder A, Fish RH, Schmid A. Bioorganometallic chemistry:
biocatalytic oxidation reactions with biomimetic NAD+/NADH co-factors and
[Cp*Rh(bpy)H]+ for selective organic synthesis. J Organomet Chem 2004;689(25):
478390.
Lynd LR, Wyman CE, Gerngross TU. Biocommodity engineering. Biotechnol Prog 1999;15:
77793.
Maier T, Leibundgut M, Ban N. The crystal structure of a mammalian fatty acid synthase.
Science 2008;321(5894):131522.
Mansson MO, Siegbahn N, Mosbach K. Site-to-site directed immobilization of enzymes
with bis-NAD analogues. Proc Natl Acad Sci U S A 1983;80(6):148791.
Martn del Campo JS, Rollin J, Myung S, Chun Y, Chandrayan S, Patio R, et al. High-yield
production of dihydrogen from xylose by using a synthetic enzyme cascade in a cellfree system. Angew Chem Int Ed 2013;52(17):458790.
Martin del Campo JS, You C, Kim J-E, Patio R, Zhang Y-HP. Discovery and characterization
of a novel ATP/polyphosphate xylulokinase from a hyperthermophilic bacterium
Thermotoga maritima. J Ind Microbiol Biotechnol 2013;40:6619.
Mateo C, Chmura A, Rustler S, van Rantwijk F, Stolz A, Sheldon RA. Synthesis of
enantiomerically pure (S)-mandelic acid using an oxynitrilasenitrilase bienzymatic
cascade: a nitrilase surprisingly shows nitrile hydratase activity. Tetrahedron Asymmetry 2006;17(3):3203.
Matthaei JH, Jones OW, Martin RG, Nirenberg MW. Characteristics and composition of
RNA coding units. Proc Natl Acad Sci U S A 1962;48(4):66677.
Mertens R, Liese A. Biotechnological applications of hydrogenases. Curr Opin Biotechnol
2004;15(4):3438.
Meynial-Salles I, Forchhammer N, Croux C, Girbal L, Soucaille P. Evolution of a Saccharomyces
cerevisiae metabolic pathway in Escherichia coli. Metab Eng 2007;9(2):1529.
Moehlenbrock M, Minteer S. Extended lifetime biofuel cells. Chem Soc Rev 2008;37:
118896.
Moradian A, Benner SA. A biomimetic biotechnological process for converting starch to
fructose: thermodynamic and evolutionary considerations in applied enzymology. J
Am Chem Soc 1992;114(18):69807.
Myung S, Zhang Y-HP. Non-complexed four cascade enzyme mixture: simple purication
and synergetic co-stabilization. PLoS One 2013;8:e61500.
Myung S, Wang YR, Zhang Y-HP. Fructose-1,6-bisphosphatase from a hyper-thermophilic
bacterium Thermotoga maritima: characterization, metabolite stability and its implications. Proc Biochem 2010;45:18827.
Myung S, Zhang X-Z, Zhang Y-HP. Ultra-stable phosphoglucose isomerase through immobilization of cellulose-binding module-tagged thermophilic enzyme on low-cost
high-capacity cellulosic adsorbent. Biotechnol Prog 2011;27:96975.
Myung S, You C, Zhang Y-HP. Recyclable cellulose-containing magnetic nanoparticles: immobilization of cellulose-binding module-tagged proteins and synthetic metabolon
featuring substrate channeling. J Mater Chem B 2013;1:441927.
Myung S, Rollin J, You C, Sun F, Chandrayan S, Adams MWW, et al. In vitro metabolic engineering of hydrogen production at theoretical yield from sucrose. Metab Eng 2014;
24(1):707.
Nakamura CE, Whited GM. Metabolic engineering for the microbial production of 1,3propanediol. Curr Opin Biotechnol 2003;14(5):4549.
Ninh PH, Honda K, Sakai T, Okano K, Ohtake H. Assembly and multiple gene expression of
thermophilic enzymes in Escherichia coli for in vitro metabolic engineering.
Biotechnol Bioeng 2014. http://dx.doi.org/10.1002/bit.25338.
Noireaux V. Principles of cell-free genetic circuit assembly. Proc Natl Acad Sci U S A 2003;
100:126727.
Opgenorth PH, Korman TP, Bowie JU. A synthetic biochemistry molecular purge valve
module that maintains redox balance. Nat Commun 2014;5:4113.
Orita I, Sakamoto N, Kato N, Yurimoto H, Sakai Y. Bifunctional enzyme fusion of 3hexulose-6-phosphate synthase and 6-phospho-3-hexuloisomerase. Appl Microbiol
Biotechnol 2007;76(2):43945.
Patnaik R, Louie S, Gavrilovic V, Perry K, Stemmer WPC, Ryan CM, et al. Genome shufing
of Lactobacillus for improved acid tolerance. Nat Biotechnol 2002;20(7):70712.
Paul CE, Gargiulo S, Opperman DJ, Lavandera I, Gotor-Fernndez V, Gotor V, et al. Mimicking nature: synthetic nicotinamide cofactors for C=C bioreduction using enoate reductases. Org Lett 2013;15:1803.
Paul CE, Arends IWCE, Hollmann F. Is simpler better? Synthetic nicotinamide cofactor analogues for redox chemistry. ACS Catal 2014;4:78897.
Peralta-Yahya PP, Zhang F, del Cardayre SB, Keasling JD. Microbial engineering for the production of advanced biofuels. Nature 2012;488(7411):3208.
Qi P, You C, Zhang YHP. One-pot enzymatic conversion of sucrose to synthetic amylose by
using enzyme cascades. ACS Catal 2014;4:13117.
Resnick SM, Zehnder AJB. In vitro ATP regeneration from polyphosphate and AMP by
polyphosphate:AMP phosphotransferase and adenylate kinase from Acinetobacter
johnsonii 210A. Appl Environ Microbiol 2000;66(5):204551.
Rieckenberg F, Ardao I, Rujananon R, Zeng A-P. Cell-free synthesis of 1,3-propanediol
from glycerol with a high yield. Eng Life Sci 2014;14:3806.
Riedel K, Bronnenmeier K. Intramolecular synergism in an engineered exo-endo-1,4-glucanase fusion protein. Mol Microbiol 1998;28:76775.
Rodriguez SB, Leyh TS. An enzymatic platform for the synthesis of isoprenoid precursors.
PLoS One 2014;9(8):e105594.
Rollin JA, Tam W, Zhang Y-HP. New biotechnology paradigm: cell-free biosystems for
biomanufacturing. Green Chem 2013;15:170819.
Rollin JA, Ye XH, Martin dCJS, Adams MWW, Zhang Y-HP. Novel hydrogen detection apparatus along with bioreactor systems. Adv Biochem Eng Biotechnol 2014. http://dx.doi.
org/10.1007/10_2014_274.
Rupp S. New Bioproduction Systems: From Molecular Circuits to Novel Reactor Concepts
in Cell-Free Biotechnology. Berlin Heidelberg: Springer; 2013. p. 121.
Ryan JD, Fish RH, Clark DS. Engineering cytochrome P450 enzymes for improved activity
towards biomimetic 1,4-NADH cofactors. Chembiochem 2008;9(16):257982.
Sakai H, Nakagawa T, Tokita Y, Hatazawa T, Ikeda T, Tsujimura S, et al. A high-power glucose/oxygen biofuel cell operating under quiescent conditions. Energy Environ Sci
2009;2:1338.
Santacoloma PA, Gr Sin, Gernaey KV, Woodley JM. Multienzyme-catalyzed processes:
next-generation biocatalysis. Org Process Res Dev 2010;15(1):20312.
Santos H, da Costa MS. Compatible solutes of organisms that live in hot saline environments. Environ Microbiol 2002;4(9):5019.
Satoh Y, Tajima K, Tannai H, Munekata M. Enzyme-catalyzed poly(3-hydroxybutyrate)
synthesis from acetate with CoA recycling and NADPH regeneration in Vitro. J Biosci
Bioeng 2003;95(4):33541.
Schoevaart R, van Rantwijk F, Sheldon RA. A four-step enzymatic cascade for the one-pot
synthesis of non-natural carbohydrates from glycerol. J Org Chem 2000;65(21):
69403.

Schoffelen S, van Hest JCM. Multi-enzyme systems: bringing enzymes together in vitro.
Soft Matter 2012;8(6):173646.
Schultheisz HL, Szymczyna BR, Scott LG, Williamson JR. Pathway engineered enzymatic
de novo purine nucleotide synthesis. ACS Chem Biol 2008;3(8):499511.
Schultheisz HL, Szymczyna BR, Williamson JR. Enzymatic synthesis and structural characterization of 13C, 15N-poly(ADP-ribose). J Am Chem Soc 2009;131(40):145718.
Scism RA, Bachmann BO. Five-component cascade synthesis of nucleotide analogues in an
engineered self-immobilized enzyme aggregate. Chembiochem 2010;11(1):6770.
Serdakowski A, Dordick J. Enzyme activation for organic solvents made easy. Trends
Shaw A, Bott R, Day AG. Protein engineering of -amylase for low pH performance. Curr
Opin Biotechnol 1999;10:34952.
Sheldon RA, van Pelt S. Enzyme immobilisation in biocatalysis: why, what and how. Chem
Soc Rev 2013;42:622335.
Sheppard AW, Gillespie I, Hirsch M, Begley C. Biosecurity and sustainability within the
growing global bioeconomy. Curr Opin Environ Sustain 2011;3(12):410.
Smith K, Cho K-M, Liao J. Engineering Corynebacterium glutamicum for isobutanol production. Appl Microbiol Biotechnol 2010;87(3):104555.
Sokic-Lazic D, Minteer SD. Citric acid cycle biomimic on a carbon electrode. Biosens
Bioelectron 2008;24(4):93944.
Somerville C, Youngs H, Taylor C, Davis SC, Long SP. Feedstocks for lignocellulosic biofuels.
Science 2010;329(5993):7902.
Steen EJ, Kang Y, Bokinsky G, Hu Z, Schirmer A, McClure A, et al. Microbial production of
fatty-acid-derived fuels and chemicals from plant biomass. Nature 2010;463(7280):
55962.
Stefer F, Sieber V. Refolding of a thermostable glyceraldehyde dehydrogenase for application in synthetic cascade biomanufacturing. PLoS One 2013;8(7):e70592.
Sun J, Hopkins RC, Jenney FE, McTernan PM, Adams MWW. Heterologous expression and
maturation of an NADP-dependent [NiFe]-hydrogenase: a key enzyme in biofuel production. PLoS One 2010;5(5):e10526.
Sun FF, Zhang XZ, Myung S, Zhang Y-HP. Thermophilic Thermotoga maritima ribose-5phosphate isomerase RpiB: optimized heat treatment purication and basic characterization. Protein Expr Purif 2012;82:3027.
Swartz JR. Cell-free bioprocessing. Chem Eng Prog 2013;2013(11):405.
Taylor LEI, Dai Z, Decker SR, Brunecky R, Adney WS, Ding S-Y, et al. Heterologous expression of glycosyl hydrolases in planta: a new departure for biofuels. Trends Biotechnol
2008;26:41324.
Thauer K, Jungermann K, Decker K. Energy conservation in chemotrophic anaerobic bacteria. Bacteriol Rev 1977;41:10080.
Thauer RK, Kaster AK, Seedorf H, Buckel W, Hedderich R. Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 2008;6:57991.
Thoden JB, Holden HM, Wesenberg G, Raushel FM, Rayment I. Structure of carbamoyl
phosphate synthetase: a journey of 96 A from substrate to product. Biochemistry
1997;36(21):630516.
Vasic-Racki D. History of industrial biotransformations dreams and realities. In: Liese A,
Seebald S, Wandrey C, editors. Industrial biotransformations. KGaA, Weinheim:
Wiley-VCH; 2006. p. 137.
Wang Y, Zhang Y-HP. Cell-free protein synthesis energized by slowly-metabolized maltodextrin. BMC Biotechnol 2009a;9:58.
Wang Y, Zhang Y-HP. Overexpression and simple purication of the Thermotoga maritima
6-phosphogluconate dehydrogenase in Escherichia coli and its application for NADPH
regeneration. Microb Cell Fact 2009b;8:30.
Wang Y, Huang W, Sathitsuksanoh N, Zhu Z, Zhang Y-HP. Biohydrogenation from biomass
sugar mediated by in vitro synthetic enzymatic pathways. Chem Biol 2011;18:
37280.
Wei B, Dai M, Yin P. Complex shapes self-assembled from single-stranded DNA tiles. Nature 2012;485(7400):6236.
Welch P, Scopes RK. Studies on cell-free metabolism: ethanol production by a yeast glycolytic system reconstituted from puried enzymes. J Biotechnol 1985;2(5):25773.
Wichmann R, Vasic-Racki D. Cofactor regeneration at the lab scale. Adv Biochem Eng
Wildeman SMAD, Sonke T, Schoemaker HE, May O. Biocatalytic reductions: from lab curiosity to rst choice. Acc Chem Res 2007;40:12606.
Wong CH, Whitesides GM. Enzyme-catalyzed organic-synthesis NAD(P)H cofactor
regeneration by using glucose-6-phosphate and the glucose-6-phosphatedehydrogenase from Leuconostoc-Mesenteroides. J Am Chem Soc 1981;103(16):
48909.
Wong CH, Daniels L, Ormejohnson WH, Whitesides GM. Enzyme-catalyzed organicsynthesis NAD(P)H regeneration using dihydrogen and the hydrogenase from
Methanobacterium thermoautotrophicum. J Am Chem Soc 1981;103(20):62278.
Woodward J, Orr M, Cordray K, Greenbaum E. Enzymatic production of biohydrogen. Nature 2000;405:10145.
Wyman CE. Biomass ethanol: technical progress, opportunities, and commercial challenges. Annu Rev Energy Environ 1999;24:189226.
Xu Z, Jing K, Liu Y, Cen P. High-level expression of recombinant glucose dehydrogenase
and its application in NADPH regeneration. J Ind Microbiol Biotechnol 2007;34(1):
8390.
Xu Y, Masuko S, Takieddin M, Xu H, Liu R, Jing J, et al. Chemoenzymatic synthesis of homogeneous ultralow molecular weight heparins. Science 2011;334(6055):498501.
Xu Y, Cai C, Chandarajoti K, Hsieh P-H, Li L, Pham TQ, et al. Homogeneous low-molecularweight heparins with reversible anticoagulant activity. Nat Chem Biol 2014;10:
24850.
Yaoi T, Miyazaki K, Oshima T, Komukai Y, Go M. Conversion of the coenzyme specicity of
isocitrate dehydrogenase by module replacement. J Biochem 1996;119(5):10148.
1483
Ye X, Wang Y, Hopkins RC, Adams MWW, Evans BR, Mielenz JR, et al. Spontaneous highyield production of hydrogen from cellulosic materials and water catalyzed by enzyme cocktails. ChemSusChem 2009;2(2):14952.
Ye X, Zhang C, Zhang Y-HP. Engineering a large protein by combined rational and random
approaches: stabilizing the Clostridium thermocellum cellobiose phosphorylase. Mol
Biosyst 2012a;8:181523.
Ye X, Honda K, Sakai T, Okano K, Omasa T, Hirota R, et al. Synthetic metabolic
engineeringa novel, simple technology for designing a chimeric metabolic pathway.
Microb Cell Fact 2012b;11(1):120.
Yim H, Haselbeck R, Niu W, Pujol-Baxley C, Burgard A, Boldt J, et al. Metabolic engineering
of Escherichia coli for direct production of 1,4-butanediol. Nat Chem Biol 2011;7(7):
44552.
Yordanov B, Kim J, Petersen RL, Shudy A, Kulkarni VV, Phillips A. Computational design of
nucleic acid feedback control circuits. ACS Synth Biol 2014;3(8):60016.
You C, Zhang Y-HP. Cell-free biosystems for biomanufacturing. Adv Biochem Eng
Biotechnol 2013;131:89119.
You C, Zhang YHP. Annexation of a high-activity enzyme in a synthetic three-enzyme
complex greatly decreases the degree of substrate channeling. ACS Synth Biol 2014;
3:3806.
You C, Myung S, Zhang Y-HP. Facilitated substrate channeling in a self-assembled
trifunctional enzyme complex. Angew Chem Int Ed 2012;51:878790.
You C, Chen H, Myung S, Sathitsuksanoh N, Ma H, Zhang X-Z, et al. Enzymatic transformation of nonfood biomass to starch. Proc Natl Acad Sci U S A 2013;110:71827.
Yu X, Liu T, Zhu F, Khosla C. In vitro reconstitution and steady-state analysis of the fatty
acid synthase from Escherichia coli. Proc Natl Acad Sci U S A 2011;108(46):186438.
Zebda A, Gondran C, Le Goff A, Holzinger M, Cinquin P, Cosnier S. Mediatorless highpower glucose biofuel cells based on compressed carbon nanotube-enzyme electrodes. Nat Commun 2011;2:370.
Zeng A-P, Sabra W. Microbial production of diols as platform chemicals: recent progresses. Curr Opin Biotechnol 2011;22(6):74957.
Zhang Y-HP. Reviving the carbohydrate economy via multi-product bioreneries. J Ind
Microbiol Biotechnol 2008;35(5):36775.
Zhang Y-HP. A sweet out-of-the-box solution to the hydrogen economy: is the sugarpowered car science ction? Energy Environ Sci 2009;2(2):27282.
Zhang Y-HP. Production of biocommodities and bioelectricity by cell-free synthetic enzymatic pathway biotransformations: challenges and opportunities. Biotechnol Bioeng
2010a;105:66377.
Zhang Y-HP. Renewable carbohydrates are a potential high density hydrogen carrier. Int J
Hydrog Energy 2010b;35:1033442.
Zhang Y-HP. Simpler is better: high-yield and potential low-cost biofuels production
through cell-free synthetic pathway biotransformation (SyPaB). ACS Catal 2011a;1:
9981009.
Zhang Y-HP. Substrate channeling and enzyme complexes for biotechnological applications. Biotechnol Adv 2011b;29:71525.
Zhang Y-HP. What is vital (and not vital) to advance economically-competitive biofuels
production. Proc Biochem 2011c;46:2091110.
Zhang Y-HP. Next generation bioreneries will solve the food, biofuels, and environmental trilemma in the energy-food-water nexus. Energy Sci Eng 2013;1:2741.
Zhang Y-HP, Xu J-H, Zhong JJ. A new high-energy density hydrogen carrier carbohydrate
might be better than methanol. Int J Energy Res 2013;37:76979.
Zhang Y-HP, Huang W-D. Constructing the electricitycarbohydratehydrogen cycle for a
sustainability revolution. Trends Biotechnol 2012;30(6):3016.
Zhang X-Z, Zhang Y-HP. One-step production of biocommodities from lignocellulosic biomass by recombinant cellulolytic Bacillus subtilis: opportunities and challenges. Eng
Life Sci 2010;10:398406.
Zhang J, Shao J, Kowal P, Wang PG. Enzymatic Synthesis of Oligosaccharides. Wiley-VCH
Verlag GmbH & Co. KGaA; 2005 [137167 pp.].
Zhang Y-HP, Himmel M, Mielenz JR. Outlook for cellulase improvement: screening and
selection strategies. Biotechnol Adv 2006;24(5):45281.
Zhang Y-HP, Evans BR, Mielenz JR, Hopkins RC, Adams MWW. High-yield hydrogen production from starch and water by a synthetic enzymatic pathway. PLoS One 2007;
2(5):e456.
Zhang Y-HP, Sun J-B, Zhong J-J. Biofuel production by in vitro synthetic pathway transformation. Curr Opin Biotechnol 2010;21:6639.
Zhang X-Z, Sathitsuksanoh N, Zhu Z, Zhang Y-HP. One-step production of lactate from cellulose as sole carbon source without any other organic nutrient by recombinant cellulolytic Bacillus subtilis. Metab Eng 2011a;13:36472.
Zhang Y-HP, Myung S, You C, Zhu Z, Rollin JA. Toward low-cost biomanufacturing through
in vitro synthetic biology: bottom-up design. J Mater Chem 2011b;21(47):1887786.
Zheng H, Ohno Y, Nakamori T, Suye S-I. Production of L-malic acid with xation of HCO
3
by malic enzyme-catalyzed reaction based on regeneration of coenzyme on electrode
modied by layer-by-layer self-assembly method. J Biosci Bioeng 2009;107(1):
1620.
Zhu ZG, Wang YR, Minteer S, Zhang Y-HP. Maltodextrin-powered enzymatic fuel cell
through a non-natural enzymatic pathway. J Power Sources 2011;196:75059.
Zhu ZG, Sun F, Zhang X, Zhang Y-HP. Deep oxidation of glucose in enzymatic fuel cells
through a synthetic enzymatic pathway containing a cascade of two thermostable
dehydrogenases. Biosens Bioelectron 2012;36:1105.
Zhu F, Zhong X, Hu M, Lu L, Deng Z, Liu T. In vitro reconstitution of mevalonate pathway
and targeted engineering of farnesene overproduction in Escherichia coli. Biotechnol
Bioeng 2014a;111:1396405.
Zhu Z-G, Kin Tam T, Sun F, You C, Zhang Y-HP. A high-energy-density sugar biobattery
based on a synthetic enzymatic pathway. Nat Commun 2014b;5:3026.

2015 - Productionofbiofuelsandbiochemicalsbyinvitrosynthe (Retrieved-2016!01!25)

Uploaded by

Copyright:

Available Formats

2015 - Productionofbiofuelsandbiochemicalsbyinvitrosynthe (Retrieved-2016!01!25)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2015 - Productionofbiofuelsandbiochemicalsbyinvitrosynthe (Retrieved-2016!01!25)

Uploaded by

Copyright:

Available Formats

Biotechnology Advances 33 (2015) 14671483

Contents lists available at ScienceDirect

Research review paper

Production of biofuels and biochemicals by in vitro synthetic biosystems:

Available online 27 October 2014

Corresponding author. Tel.: +1 540 231 7414.

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

Martn del Campo et al. (2013); Myung et al. (2014)

Reaction control and

High product yield due to co-production of ethanol, resulting in no sugar losses.

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

breakthroughs will revolutionize industrial bioprocess by producing a

low cooling energy) and low capital investment, especially important

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

extra reduced coenzymes, such reduced coenzymes should be removed

2.2. Examples of biofuels and bioelectricity

H2 O ribulose 5phosphate 2NADPH 2H CO2

2NADPH 2H 2H2 2NADP

6 Ribulose 5phosphate H2 O 5 G6P Pi

Furthermore, I proposed to generate glucose 1-phosphate (G-1-P)

As a result, the whole stoichiometric reaction can be written as

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

The overall reaction (Eq. (6)) is spontaneous and endothermic

C6 H12 O6 glucose 2NAD

2 pyruvate 2 CoA 2 acetylCoA 2 CO2

2 acetylCoA 2 NADH C4 H10 O nbutanol H2 O 2 CoA 2NAD

C6 H12 O6 glucose C4 H10 O nbutanol 2 CO2 H2 O

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

et al., 2008), Bacillus subtilis (Li et al., 2011), Corynebacterium

C6 H12 O6 glucose 2NAD

C4 H10 O isobutanol 2 CO2 H2 O 2NAD

C6 H12 O6 glucose C4 H10 O isobutanol 2 CO2 H2 O

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

ribulose5phosphate CO2 2NADH 2H

C6 H10 O5 7H2 O 24e 6CO2 24H anode compartment

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

2.3.1. Synthetic amylose

n C12 H22 O11 cellobiose n glucose C6 H10 O5 n amylose

glucose 2 C2 H6 O ethanol 2 CO2

C6 H10 O5 n cellulose C6 H10 O5 n=2 amylose n C2 H6 O ethanol n CO2

fermentation in second generation cellulosic bioreneries, where

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

catalyzed by lactate dehydrogenase (Eq. (23)). As a result, one glucose

2 pyruvate 2 NADH 2H 2 C3 H6 O3 lactate 2 NAD

C6 H12 O6 glucose 2 C3 H6 O3 lactate

By compromising reaction temperature and optimizing enzyme

glycerol NADPH H 1; 3PDO NADP H2 O

2 Pyruvate 2 CoA NAD NADP

2 acetylCoA NADPH NADH H 2CO2

P3HB 2 CoA NAPD

2 Pyruvate O2 P3HB 2 CO2 H2 O2

3. Economic analysis of the hydrogen case

The most appealing biomanufacturing feature of in vitro synthetic

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

biochemicals, will also be amenable to cost-competitive production via

Y.-H.P. Zhang / Biotechnology Advances 33 (2015) 14671483

solution to address numerous challenges of the hydrogen economy

a good starting point is to use thermoenzymes. Many examples exist