available online at koffam/papers/2012_xu_bhan_koffas.pdfxu ,2, namita bhan 1and mattheos ag koffas...

COBIOT-1086; NO. OF PAGES 9

Engineering plant metabolism into microbes: from systemsbiology to synthetic biologyPeng Xu1,2, Namita Bhan1,2 and Mattheos AG Koffas1,2

Available online at www.sciencedirect.com

Plant metabolism represents an enormous repository of

compounds that are of pharmaceutical and biotechnological

importance. Engineering plant metabolism into microbes will

provide sustainable solutions to produce pharmaceutical and

fuel molecules that could one day replace substantial portions

of the current fossil-fuel based economy. Metabolic

engineering entails targeted manipulation of biosynthetic

pathways to maximize yields of desired products. Recent

advances in Systems Biology and the emergence of Synthetic

Biology have accelerated our ability to design, construct and

optimize cell factories for metabolic engineering applications.

Progress in predicting and modeling genome-scale metabolic

networks, versatile gene assembly platforms and delicate

synthetic pathway optimization strategies has provided us

exciting opportunities to exploit the full potential of cell

metabolism. In this review, we will discuss how systems and

synthetic biology tools can be integrated to create tailor-made

cell factories for efficient production of natural products and

fuel molecules in microorganisms.

Addresses1 Department of Chemical and Biological Engineering, Rensselaer

Polytechnic Institute, Troy, NY 12180, United States2 Center for Biotechnology and Interdisciplinary Studies, Rensselaer

Polytechnic Institute, Troy, NY 12180, United States

Corresponding author: Koffas, Mattheos AG ([email protected],

[email protected])

Current Opinion in Biotechnology 2012, 24:xx–yy

This review comes from a themed issue on Plant biotechnology

Edited by Natalia Dudareva and Dean DellaPenna

0958-1669/$ – see front matter, Published by Elsevier Ltd.

http://dx.doi.org/10.1016/j.copbio.2012.08.010

IntroductionPlant metabolism represents an enormous repository ofbioactive compounds known as natural products (NPs) thatare of pharmaceutical and biotechnological importance [1].Specialized in skeletal structures and functional groups,plant NPs continue to play a leading role in drug discovery[2]. For example, it has been estimated that more than 75%of FDA-approved antibiotics and anticancer drugs in thelast 25 years are NPs or inspired by NPs [3]. As a result,tremendous effort has been dedicated to the developmentof cost-efficient processes for the production of importantNPs. Currently, most of commercialized NPs are still

Please cite this article in press as: Xu P, et al.: Engineering plant metabolism into microbes: fromj.copbio.2012.08.010

www.sciencedirect.com

manufactured by extraction from their native plant sourcesor semi-synthesized from extracted NP intermediates.The major bottleneck in the plant extraction process isthe low yield and the complicated downstream purificationprocesses. While organic chemists have been able to ele-gantly synthesize/modify an array of complex NPs, suchchemical synthesis processes are usually overshadowed byinherent disadvantages such as low yield, use of toxiccatalysts and extreme reaction conditions that make themnot amenable to large scale production [4].

Alternatively, microbial metabolic engineering hasemerged as a promising approach to overcome theselimitations. With the introduction of a plant heterologouspathway, universal precursors/intermediates (i.e. acetyl-CoA, malonyl-CoA, isopentenyl pyrophosphate) gener-ated from the microbial central carbon metabolism andessential for the cell’s own survival need to be redirectedin order to efficiently produce value-added NPs frominexpensive feedstocks. For instance, genetically tract-able microorganisms (e.g. Escherichia coli and Saccharo-myces cerevisiae) engineered with plant secondarymetabolic pathways have demonstrated great success toproduce a range of NPs including fatty acids [5], terpe-noids [6,7,8��], flavonoids [9,10�], polyketides [11] andalkaloids [12]. Conventional metabolic engineering strat-egies have been primarily focused on redirecting carbonflux toward a target pathway by overexpression of rate-limiting steps and deletion of precursor competing-path-ways [13]. With the advent of the post-genomics era, astaggering volume of genomic, transcriptomic, proteomicand metabolomic information has been deposited; thus itis now possible to manipulate a specific metabolic net-work by tweaking the entire cell metabolism [14]. Con-currently, as the characterized genetic parts and deviceshave been accumulated and the cost of DNA synthesiscontinues to decline, synthetic biology and metabolicengineering has enabled the precise control of cell metab-olism to customize the production of plant NPs at a scaleand speed never seen before. In this review, we willdiscuss how systems and synthetic biology tools can beintegrated to design and create tailor-made cell factoriesfor efficient production of NPs and fuel molecules inmicroorganisms.

Synthetic metabolic pathway designOwing to the highly evolved transcriptional regulatorynetwork and metabolic feedback control, microorganismstend to maintain a constant level of essential precursormetabolite flux and will not commit to overproduce the

systems biology to synthetic biology, Curr Opin Biotechnol (2012), http://dx.doi.org/10.1016/

Current Opinion in Biotechnology 2012, 24:1–9

http://dx.doi.org/10.1016/j.copbio.2012.08.010http://dx.doi.org/10.1016/j.copbio.2012.08.010mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.copbio.2012.08.010http://www.sciencedirect.com/science/journal/09581669

2 Plant biotechnology


Figure 1

OptimalsolutionFlux

v 3

Fluxv 2

Flux v1

679.0 679.5 680.0 680.5m/z

681.0

Current Opinion in Biotechnology

Systems biology tools bring together layers of information that will advance our understanding to design and model intertwined metabolic network.

Genomics, transcriptomics, proteomics and constraints-based flux balance analysis (FBA) provide a global landscape of cell metabolism that will

benefit the design and control of biological systems for biotechnological applications. Left panel shows the original metabolic network and the right

panel shows optimized metabolic network with maximal carbon flux toward a desired product. On the right panel, dark green blocks show

overexpression or upregulation targets; red blocks show knockout or downregulation targets. Double arrows show interplay between the neighboring

components.

desired recombinant molecules [15]. It is often the casethat manipulation of local metabolic networks can onlylead to limited performance improvement. In the pastdecade, the field of systems biology has seen tremendousadvances in understanding and predicting the dynamicbehavior of complex biological systems [16]. Integratedexperimental and computational tools have allowed us toattain a global landscape of molecular interactions amongthe myriad cellular components. Toward the goal ofbetter engineering cell metabolism for NPs production,a number of high throughput omics tools and constraints-based flux balance models have been employed and ourability to quantitatively dissect cellular phenotypes hasbeen greatly enhanced (Figure 1). These achievementshave led to the emergence of the new concept of ‘systemsmetabolic engineering’ [14] that aims to move strainengineering on a global scale.

The large volume of plant genomic information gener-ated from high throughput DNA sequencing techniques



has provided immense possibility for understanding thebiosynthetic potential of plant metabolism [17]. Suchdata can be exploited to identify cryptic pathways orunknown gene clusters that are responsible for valuedNP biosynthesis. Sequence comparison betweenunknown DNAs and canonical genes allows the predic-tion of conserved catalytic motifs that potentially couldlead to new chemical entities [18]. For example, gen-ome mining has been successfully applied to understandthe biochemical programming and molecular basis oftwo types of NP biosynthetic systems: the modularpolyketide synthases and nonribosomal polyketidesynthetases [19]. Recently, bioinformatics analysisof genomic information and mRNA transcripts has ledto the prediction of enzymes and pathways involved interpenoid biosynthesis in both Catharanthus roseus [20]and Ganoderma lucidum [21]. De novo sequencing, assem-bly of transcriptomic data has also allowed the predic-tion of biosynthetic pathways for a range of NPsincluding prostratins [22], picrosides [23] and hypericins



http://dx.doi.org/10.1016/j.copbio.2012.08.010http://dx.doi.org/10.1016/j.copbio.2012.08.010

Engineering plant metabolism into microbes Xu, Bhan and Koffas 3


[24]. Targeted proteomics that directly correlate proteintranslation efficiency with pathway limitations based onthe accurate determination of enzyme abundance havebeen used to fine-tune/optimize heterologous mevalo-nate [25] and phenylalanine pathways [26] in E. coli,leading to improved production of amorpha-4,11-dieneand tyrosine, respectively. These high throughput sys-tems biology tools have largely improved our ability todesign and optimize metabolic pathways for efficientproduction of NPs in microorganisms.

At the same time, in silico computational biology toolsbased on flux balance analysis (FBA) or kinetics formu-lation have been developed to identify genetic manip-ulation targets that would redistribute carbon flux towarda compound of interest. The most popular in silico gen-ome-scale metabolic computational method is con-straints-based flux balance analysis [14]. It allows thedetermination of optimal reaction flux (i.e. maximal cellgrowth, maximal product formation or minimal precursordissipation) by searching for a linear space constrained bystoichiometry-based mass balance equations and pre-scribed physio-biochemical boundaries. More impor-tantly, possible flux distributions solved by linear orquadratic programming can be ranked and chosen toprioritize the implementation of genetic interventions.For example, strain optimization procedures such asOptORF [27], RobustKnock [28] and CiED [29,30] haveenabled efficient use of FBA as a tool to predict geneoverexpression and/or gene deletion targets. The ever-increasing size of biochemical datasets has motivated theuse of algorithms that incorporate layers of information tofurther constrain flux space and achieve more accuratepredictions. For example, transcriptional regulation net-work [31], chemical reaction thermodynamics [32], RNA-seq data [33], flux measurement data [34] and kineticconstraints [35] have all been incorporated into FBAframework to identify and optimize metabolic states invarious organisms. Recently, genetic interventions pre-dicted by OptForce have been successfully applied toimprove intracellular malonyl-CoA and the resultingstrain demonstrated around a 5-fold increase in flavonoidproduction in E. coli [10�]. Such achievements demon-strate that constraints-based flux balance models arepowerful tools for strain design and improvement.

Synthetic pathway construction: enablingtechnology to implement the vision ofsynthetic biologySynthetic biology is characterized by a constructiveapproach to understand and manipulate biological sys-tems [36]. Even though the cost of commercial synthesisof genes is declining, our ability to physically constructcomplex biological devices/pathways from basic DNAparts is still a critical hurdle in implementing the visionof synthetic biology [37�]. These limitations can becomeprohibitive when constructing multi-gene metabolic



pathways and complex regulatory circuits. For example,strain development through metabolic engineering lead-ing to valuable NPs typically involves the manipulationof a dozen precursors or rate-limiting pathways [7,8��].Conventional pathway construction approaches, whichlargely rely on smart design to assemble multi-genefragments in operon form, are limited in terms of auto-mation and context-independent pathway output.Therefore, synthetic metabolic engineering requiresefficient tools that allow precise and concerted assemblyof multiple gene fragments, leading to programmableor predictable pathway output customized for strainoptimization.

Last decade’s advancement in synthetic biology has ledto the emergence of several robust gene assembly plat-forms that are suited for metabolic pathway constructionand optimization. These gene assembly tools allow us toefficiently construct directional multi-gene pathwaysfrom synthesized or PCR-based gene fragments with30+ bp overlap regions (Figure 2). For example,Sequence and Ligation-Independent cloning (SLIC)[38,39] and Gibson isothermal assembly [40] rely on invitro homologous recombination and single-strandannealing. SLIC has been recently modified to allowone-step assembly of multiple medium-size DNA frag-ments [41] and large genomic DNA fragments (up to28 kb) [42] using in vitro single-strand overlappingannealing. In vivo assembly tools based on yeast hom-ologous recombination have also been developed forassembling large DNA fragments encoding an entirebiochemical pathway [43] or iteratively integratingmulti-gene pathways into yeast chromosomes [44].Coupled with yeast in vivo recombination, Gibson iso-thermal assembly has been used to assemble a self-replicating synthetic M. genitalium genome (1.08 Mb)[45�] and a high G + C genomic DNA fragment(454 kb) containing Synechococcus elongatus PCC 7942photosynthetic gene clusters [46]. Circular polymeraseextension cloning (CPEC) has been developed for one-step assembly of multi-gene pathways and combinatorialDNA libraries using complementary fragment annealingand PCR overlap extension [47].

Compared with conventional DNA cloning protocols,these advanced DNA assembly tools offer an efficientapproach to construct multi-gene pathways in a one-step,scar-less and sequence-independent manner. The down-side of these assembly tools is the terminal flankinghomology sequence of each DNA fragment must bespecifically designed for each assembly junction, whichtends to be laborious and error-prone [48]. Assembly ofgene fragments with identical end-terminus sequences(such as repeated promoters, repeated ribosome bindingsite and repeated terminators) can be problematic, as thiscan lead to constructs with missing, extra or rearrangedgene fragments in the wrong order. As such, these






Figure 2

SLICDNA Assembler

Gibson Isothermal

(a)

(b)

(c)

Denaturation PCR extension

AvrIIXbal

CDS-2

SpelNhel

Donorvector

Sall

Annealing

AvrII

CDS-1

AvrII

CDS-1

CDS-2

SpelNhel

SallNhel

Spel

CDS-2

SallNhel

Spel

CDS-2

CDS-1

AvrII

CDS-1

AvrII

Sall

T7 promoter/lacO Ribosome binding site T7 terminator

Destinationvector

Operonconfiguration

Pseudo-operonconfiguration

Mono-cistronicconfiguration

SpeINheI

SalI

Spe

l + S

all

Xbal + S

allSpel + Sall Avr

ll + Sa

llNhel + Sall

Avrll + Sall

blabla

ori

lacI

ori

lacI

ori

bla

lacI

ori

lacI

ori

blabla

lacI


DNA assembly tools for synthetic pathway construction. (a) Parallel gene assembly tools provide an efficient approach to construct directional multi-gene

pathways from synthesized or PCR-based gene fragments with 30+ bp overlap region. SLIC (sequence and ligation-independent cloning) and Gibson

isothermal assembly are based on in vitro homologous recombination and single-strand annealing; specifically, SLIC used a bifunctional DNA polymerase

(T4 DNA polymerase) that exhibits 30 exonuclease activity in the absence of dNTPs, whereas Gibson assembly used enzyme mixtures consisting of 50

exonuclease, Phusion DNA polymerase and Taq DNA ligase. DNA assembler is based on yeast in vivo homologous recombination. (b) Circular polymerase

extension cloning (CPEC) technique for gene assembly. Complementary overhangs attached to insertion gene fragment anneal to linearized vector and

can be circularized into an intact vector by PCR extension. (c) ePathBrick vectors provide a versatile platform to combinatorially assemble multi-gene

pathways with different configurations. The ePathBricks take advantage of four compatible restriction enzymes (AvrII, XbaI, SpeI and NheI), which

generate compatible cohesive ends and upon ligation result in a scar sequence that cannot be cleaved by either of the original restriction enzymes. Basic

molecular components flanked with these isocaudamer pairs can be assembled together by restriction enzyme digestion and ligation.

advanced cloning tools do not offer the capability torefine the regulatory control elements of each of theexpression cassettes if identical regulatory sequences(such as promoters, operators, ribosome binding sites



and terminators) are used. In addition, due to the pre-determined fragment context specified by the overlapsequence, these gene assembly tools are not scalable fordiversifying pathway configurations and shuffling gene






order [49��]. For these reasons, there is a need for newgene assembly tools adept at hierarchical pathway con-struction and generation of pathway diversities in acombinatorial manner.

Assembly of DNA fragments based on the BioBrickparadigm has become a standard practice in metabolicpathway construction [50,51] and genetic circuit design[52]. A versatile gene assembly platform, ePathBrick,compatible with BioBrick standards has been recentlydescribed for metabolic pathway construction and optim-ization [49��]. The engineered ePathBrick vectors featurefour compatible restriction enzyme sites (AvrII, XbaI,SpeI and NheI) allocated on strategic positions that supportthe modular assembly of a number of molecular com-ponents including regulatory control elements (promo-ters, operators, ribosome binding sites and terminators)and multi-gene pathways (Figure 2c). In addition, ePath-Bricks provide a platform for combinatorial generation ofpathway diversities with three distinct configurations.Specifically, for a multi-gene pathway, each of the path-way components can be organized either in operon,pseudo-operon, or monocistronic form (Figure 2c) whenusing different isocaudamer pairs. It was demonstratedthat a three-gene flavonoid pathway can be easily diver-sified to 54 pathway equivalents differing in pathwayconfiguration and gene order; coupled with high through-put screening techniques, we envision that this combi-natorial strategy would greatly improve our ability toexploit the full potential of microbial cell factories forrecombinant metabolite and NP production. A total path-way size of 36 kb could be assembled using four compa-tible ePathBrick vectors, which is sufficient for expressinga typical plant secondary metabolic pathway [49��]. TheePathBrick assembly, though limited by one gene at a time,provides a versatile platform to combinatorially constructand optimize multi-gene pathways — a feature that will beproven extremely useful for pathway engineering.

Synthetic pathway optimization throughstatic, dynamic and spatial controlOne of the most important goals of metabolic engineeringand synthetic biology has been the optimization of meta-bolic pathways for the production of biofuels [53,54�] andpharmaceuticals [7] in genetically tractable organisms. Amajor challenge in these efforts has been the accuratetuning and optimization of the expression level of each ofthe enzymes of the selected pathways. Toward this goal, anumber of synthetic biology approaches have beenapplied including modification of plasmid copy number[26], promoter strength [55] and gene codon usage [56]. Inaddition, delicately designed molecular control elementsor regulatory circuits have been integrated into the cellchassis to enable the host strain to precisely respond toenvironmental stimuli or cellular intermediates and drivecarbon flux toward the target pathway. For example,engineering promoter architecture has resulted in tunable



gene expression in both E. coli [57] and yeast [58];engineered metabolite-responsive riboswitches [59] andsynthetic ribosome binding sites [60] can be used toprecisely control protein translation; engineering genetictimers based on mutually repressible toggle switches havebeen used to control the timing of yeast sedimentation[61].

However, engineering microbial overproduction pheno-types remains a daunting task as it usually involves themanipulation of a handful of precursor or rate-limitingpathways that are subjected to tight cellular regulation.For example, precursor flux improvement by overexpres-sion of heterologous pathways may not be accommodatedby downstream pathways; as a result, accumulated ordepleted intermediates may compromise cell viabilityand pathway productivity [6]. In an attempt to addressthese issues, combinatorial transcriptional engineeringhas been proposed to optimize the expression level ofeach of the genes involved in a multi-gene pathway.Central to this endeavor is an approach called ‘multi-variate modular metabolic engineering’ (Figure 3a),which aims at removing pathway bottlenecks and opti-mizing a pathway’s potential through systematic investi-gation of pathway input variables [62]. For example,strategies including combinatorially tuning promoterstrength, plasmid copy numbers and translationalinitiation rates, have been deployed for optimizing thebiosynthesis of important metabolites, such as taxadiene[8��] and free fatty acids (P Xu et al., Modular optimiz-ation of multi-gene pathways for fatty acids production inE. coli, unpublished data). It is important to note here thatprecisely designed statistical experiments such as responsesurface methodology [63] can be used to screen maineffects and optimize pathway output instead of performinga full factorial search. Combined with synthetic promoterlibraries [57] or synthetic ribosome binding sites [60], themultivariate modular metabolic engineering strategies pro-vide a generalized approach to optimize cell factories forvaluable metabolite production.

In nature, functionally related enzymes are often orga-nized in close proximity by noncovalent interactions ormembrane-tethered anchor proteins. This spatial co-localization strategy enables products formed from oneenzyme to be efficiently translocated to another enzyme,thus preventing the accumulation of toxic intermediatesand minimizing the loss of reactive intermediates due todiffusion (Figure 3b and c). Recently, a number of scaf-fold-based approaches that mimic this substrate-channel-ing effect have been developed to optimize the metabolicflux through a target pathway. For example, heterologousenzymes involved in the mevalonate biosynthetic pathwayhave been tethered on specifically designed syntheticprotein scaffolds at stoichiometric ratios [64��]. This spatialcontrol led to a 77-fold improvement in product titers at lowenzyme expression levels. The same strategy has been






Figure 3

Promoter strength(a)

(b) (c)

(d)

RBS strength

gene I gene II

Enzyme I Enzyme II

Enzyme III

PU

PD

+

Upstreampathway

Downstreampathway

Sensor

Substrate

Intermediate

Product

1.0

1.0

0.8

0.6

0.4

0.2

0.0

0.8

0.6

0.4

0.2

PD

PU

Downstreampathway

Upstreampathway

0.0

10-3 10-2 10-1 100 10 1 10 2 10 3 10 4

10-3 10-2 10-1 100 10 1 10 2[Intermediate]

[Intermediate]

103 10 4 10 5 10 6 10 7 10 8

105 10 6 10 7 10 8

Pro

mot

er a

ctiv

ityP

rom

oter

act

ivity

Synthetic protein scaffold

Pro

duct

ion

pote

ntia

l

Strength of gene I Strength

of gene

II


Synthetic pathway optimization through static, spatial and dynamic regulatory control. (a) Schemes of multivariate modular pathway engineering to

balance cell metabolism and improve production titer. Customized promoters and ribosome binding sites were combinatorially used to remove

pathway bottlenecks by tuning gene expression at both transcriptional and translational level. Right panel shows precisely designed statistical

experiments (Response Surface Methodology) can be used to search for the optimal expression pattern for a two gene pathway. (b) Co-localization of

pathway enzymes on synthetic protein scaffold to channel carbon flux from substrate to product. (c) Spatial display of functionally related metabolic

enzymes on the yeast cell surface. (d) Dynamic sensor–regulator systems use a transcriptional factor that senses a key intermediate and dynamically

modulates the expression of genes involved in the target pathway. Right panel shows how accumulated intermediates affect the promoter activity of

the upstream pathway or downstream pathway. Specifically, PD (downstream pathway promoter) is positively regulated by the accumulated

intermediate; PU (upstream pathway promoter) is negatively regulated by the accumulated intermediate.

applied to improve the production of glucaric acid in E. coli[65]. Similarly, yeast cell surface displayed with engin-eered minihemicellulosome that assembles cellulose-degrading enzymes has enabled ethanol productiondirectly from plant biomass xylan [66�]. Potentially, these



substrate-channeling strategies could be applied to othersystems for the modular control over metabolic flux.

As heterologous pathways become larger and more com-plicated, it becomes increasingly difficult to optimize






them with static regulatory control [67]. Generally, optim-ization through static control is only applicable for aparticular environment and any perturbations that deviatefrom the prescribed condition would likely result inphenotype instability or suboptimal productivity. In con-trast to static control, native biological systems typicallyutilize dynamic regulatory networks to control metabolicflux in response to changing environments (Figure 3d).For example, one of the coherent strategies that occur inmost biological systems is gene expression regulationthrough negative/positive feedback loops. Mediated bya transcriptional regulator, a metabolic intermediatewould act as a signaling molecule to induce or repressthe expression of enzymes responsible for its synthesis orconsumption. Nature has evolved this strategy to allowmetabolic pathways to be dynamically modulated so thatcellular resources can be more efficiently utilized regard-less of the changing environment. In practice, thisstrategy could be applied to pathway optimization inthe case where accumulated toxic intermediates are detri-mental to cell growth. For example, a negative feedbackloop can be used to control pathways that are responsiblefor intermediate synthesis (Figure 3d); and a positivefeed-forward loop can be used to control pathways thatare responsible for intermediate consumption. Successfulapplication of this optimization strategy would requireidentification of promoters that are both positively andnegatively regulated by the same intermediate molecule.Recently, a fatty acyl-CoA responsive promoter has beenused to dynamically control the expression of genescoding for enzymes involved in the biodiesel pathway;the resulting dynamic sensor-regulator system has led to a3-fold increase in FAEEs (fatty acids ethyl esters) pro-duction compared with using constitutive promoters[68��] in E. coli.

PerspectiveThe advances in systems and synthetic biology havegreatly speeded up our ability to design, construct andoptimize cell factories for metabolic engineering appli-cations. Concerns about climate change, rising petroleumprices and emerging health problems will continue tomotivate us to explore novel solutions for the productionof fuel and pharmaceutical molecules. It is envisionedthat synthetic cell factories will partially replacepetroleum-based chemical synthesis in the next 5–10years. Growing interest in modeling cell metabolism willresult in the development of new computational tools thatincorporate dynamic gene expression patterns and offer amechanistic view of regulatory networks. Furthermore,gene assembly tools that are adept at high throughputcombinatorial pathway construction will physicallyenable us to exploit the full potential of cell metabolism.Knowledge about hierarchical regulatory architecture [69]and network motif [70] will transform our understandingof how to build modular and orthogonal genetic circuits tocontrol gene expression. Large scale gene-editing tools



such as MAGE [71�], CAGE [72] and TALENs [73] willallow the targeted regulation and modification of genomicDNAs. By combining these tools, metabolic engineerswill be well positioned to create tailor-made cell factoriesfor efficient, high-yield and economical production ofNPs and fuel molecules in the near future.

AcknowledgementsThis work was supported by NIH 5 R01 GM085323-04 award to MattheosKoffas.

References and recommended readingPapers of particular interest, published within the period of review,have been highlighted as:

� of special interest�� of outstanding interest

1. Facchini PJ, Bohlmann J, Covello PS, De Luca V, Mahadevan R,Page JE, Ro DK, Sensen CW, Storms R, Martin VJ: Syntheticbiosystems for the production of high-value plant metabolites.Trends Biotechnol 2012, 30:127-131.

2. Li JW, Vederas JC: Drug discovery and natural products: endof an era or an endless frontier? Science 2009, 325:161-165.

3. Newman DJ, Cragg GM: Natural products as sources of newdrugs over the last 25 years. J Nat Prod 2007, 70:461-477.

4. Chemler J, Koffas M: Metabolic engineering for plant naturalproduct biosynthesis in microbes. Curr Opin Biotechnol2008:597-605.

5. Zhang X, Li M, Agrawal A, San KY: Efficient free fatty acidproduction in Escherichia coli using plant acyl-ACPthioesterases. Metab Eng 2011, 13:713-722.

6. Leonard E, Ajikumar P, Thayer K, Xiao W, Mo J, Tidor B,Stephanopoulos G, Prather K: Combining metabolic and proteinengineering of a terpenoid biosynthetic pathway foroverproduction and selectivity control. Proc Natl Acad Sci U S A2010:13654-13659.

7. Westfall PJ, Pitera DJ, Lenihan JR, Eng D, Woolard FX, Regentin R,Horning T, Tsuruta H, Melis DJ, Owens A et al.: Production ofamorphadiene in yeast, and its conversion todihydroartemisinic acid, precursor to the antimalarial agentartemisinin. Proc Natl Acad Sci U S A 2012, 109:E111-E118.

8.��

Ajikumar PK, Xiao W-H, Tyo KEJ, Wang Y, Simeon F, Leonard E,Mucha O, Phon TH, Pfeifer B, Stephanopoulos G: Isoprenoidpathway optimization for taxol precursor overproduction inEscherichia coli. Science 2010, 330:70-74.

The authors demonstrated multivariate modular pathway engineeringapproch to optimize isoprenoid pathway for taxadiene production in E.coli. By combinatorially changing promoter strength and plasmid copynumber, the authors successfully identified optimal pathway expres-sion levels that balance the formation and consumption of isopentenylpyrophosphate and minimized the secrection of toxic metaboliteindole.

9. Lim C, Fowler Z, Hueller T, Schaffer S, Koffas M: High-yieldresveratrol production in engineered Escherichia coli. ApplEnviron Microbiol 2011:3451-3460.

10.�

Xu P, Ranganathan S, Fowler Z, Maranas C, Koffas M: Genome-scale metabolic network modeling results in minimalinterventions that cooperatively force carbon flux towardsmalonyl-CoA. Metab Eng 2011, 13:578-587.

By using a recently developed OptForce framework, the authors suc-cessfully identified minimal set of genetic interventions that cooperativelyforce carbon flux toward malonyl-CoA. The authros also experimentallyvalidated these predictions and the optimized metabolic network led to a5.6-fold increase in flavonoid production.

11. Boghigian BA, Zhang H, Pfeifer BA: Multi-factorial engineeringof heterologous polyketide production in Escherichia colireveals complex pathway interactions. Biotechnol Bioeng 2011,108:1360-1371.






12. Nakagawa A, Minami H, Kim JS, Koyanagi T, Katayama T, Sato F,Kumagai H: A bacterial platform for fermentative production ofplant alkaloids. Nat Commun 2011, 2:326.

13. Xu P, Koffas MAG: Metabolic engineering of Escherichia coli forbiofuel production. Biofuels 2010, 1:493-504.

14. Lee JW, Na D, Park JM, Lee J, Choi S, Lee SY: Systemsmetabolic engineering of microorganisms for natural and non-natural chemicals. Nat Chem Biol 2012, 8:536-546.

15. Gerosa L, Sauer U: Regulation and control of metabolic fluxesin microbes. Curr Opin Biotechnol 2011, 22:566-575.

16. Smolke CD, Silver PA: Informing biological design byintegration of systems and synthetic biology. Cell 2011,144:855-859.

17. Mitchell W: Natural products from synthetic biology. Curr OpinChem Biol 2011, 15:505-515.

18. Challis GL: Genome mining for novel natural productdiscovery. J Med Chem 2008, 51:2618-2628.

19. Bergmann S, Schümann J, Scherlach K, Lange C, Brakhage AA,Hertweck C: Genomics-driven discovery of PKS-NRPS hybridmetabolites from Aspergillus nidulans. Nat Chem Biol 2007,3:213-217.

20. Yu F, Thamm AM, Reed D, Villa-Ruano N, Quesada AL, Gloria EL,Covello P, De Luca V: Functional characterization of amyrinsynthase involved in ursolic acid biosynthesis in Catharanthusroseus leaf epidermis. Phytochemistry 2012 http://dx.doi.org/10.1016/j.phytochem.2012.05.002.

21. Chen S, Xu J, Liu C, Zhu Y, Nelson DR, Zhou S, Li C, Wang L,Guo X, Sun Y et al.: Genome sequence of the model medicinalmushroom Ganoderma lucidum. Nat Commun 2012, 3:913.

22. Barrero R, Chapman B, Yang Y, Moolhuijzen P, Keeble-Gagnere G, Zhang N, Tang Q, Bellgard M, Qiu D: De novoassembly of Euphorbia fischeriana root transcriptomeidentifies prostratin pathway related genes. BMC Genomics2011, 12:600.

23. Gahlan P, Singh H, Shankar R, Sharma N, Kumari A, Chawla V,Ahuja P, Kumar S: De novo sequencing and characterization ofPicrorhiza kurrooa transcriptome at two temperaturesshowed major transcriptome adjustments. BMC Genomics2012, 13:126.

24. He M, Wang Y, Hua W, Zhang Y, Wang Z: De Novo sequencing ofHypericum perforatum transcriptome to identify potentialgenes involved in the biosynthesis of active metabolites. PLoSONE 2012, 7:e42081.

25. Redding-Johanson AM, Batth TS, Chan R, Krupa R, Szmidt HL,Adams PD, Keasling JD, Lee TS, Mukhopadhyay A, Petzold CJ:Targeted proteomics for metabolic pathway optimization:application to terpene production. Metab Eng 2011,13:194-203.

26. Juminaga D, Baidoo EE, Redding-Johanson AM, Batth TS, Burd H,Mukhopadhyay A, Petzold CJ, Keasling JD: Modular engineeringof L-tyrosine production in Escherichia coli. Appl EnvironMicrobiol 2012, 78:89-98.

27. Kim J, Reed JL: OptORF: optimal metabolic and regulatoryperturbations for metabolic engineering of microbial strains.BMC Syst Biol 2010, 4:53.

28. Tepper N, Shlomi T: Predicting metabolic engineeringknockout strategies for chemical production: accounting forcompeting pathways. Bioinformatics 2009, 26:536-543.

29. Fowler ZL, Gikandi WW, Koffas MA: Increasing malonyl-CoAbiosynthesis by tuning the Escherichia coli metabolic networkand its application to flavanone production. Appl EnvironMicrobiol 2009, 75:5831-5839.

30. Chemler J, Fowler Z, McHugh K, Koffas M: Improving NADPHavailability for natural product biosynthesis in Escherichia coliby metabolic engineering. Metab Eng 2010:96-104.

31. Jensen PA, Lutz KA, Papin JA: TIGER: toolbox for integratinggenome-scale metabolic models, expression data, andtranscriptional regulatory networks. BMC Syst Biol 2011, 5:147.



32. Fleming RM, Thiele I: von Bertalanffy 1.0: a COBRA toolboxextension to thermodynamically constrain metabolic models.Bioinformatics 2011, 27:142-143.

33. Gowen CM, Fong SS: Genome-scale metabolic modelintegrated with RNAseq data to identify metabolic states ofClostridium thermocellum. Biotechnol J 2010, 5:759-767.

34. Ranganathan S, Suthers PF, Maranas CD: OptForce: anoptimization procedure for identifying all geneticmanipulations leading to targeted overproductions. PLoSComput Biol 2010, 6:e1000744.

35. Jamshidi N, Palsson BO: Formulating genome-scale kineticmodels in the post-genome era. Mol Syst Biol 2008:4.

36. Cheng AA, Lu TK: Synthetic biology: an emerging engineeringdiscipline. Annu Rev Biomed Eng 2012, 14:155-178.

37.�

Ellis T, Adie T, Baldwin GS: DNA assembly for synthetic biology:from parts to pathways and beyond. Integr Biol (Camb) 2011,3:109-118.

The authors gave a comprehensive review about the advantages andlimitations of currently used gene assembly platforms.

38. Li MZ, Elledge SJ: Harnessing homologous recombination invitro to generate recombinant DNA via SLIC. Nat Methods 2007,4:251-256.

39. Jeong J-Y, Yim H-S, Ryu J-Y, Lee HS, Lee J-H, Seen D-S,Kang SG: One-step sequence- and ligation-independentcloning as a rapid and versatile cloning method for functionalgenomics studies. Appl Environ Microbiol 2012, 78:5440-5443.

40. Gibson DG, Young L, Chuang RY, Venter JC, Hutchison CA,Smith HO: Enzymatic assembly of DNA molecules up to severalhundred kilobases. Nat Methods 2009, 6:343-345.

41. Schmid-Burgk JL, Xie Z, Frank S, Virreira Winter S, Mitschka S,Kolanus W, Murray A, Benenson Y: Rapid hierarchical assemblyof medium-size DNA cassettes. Nucleic Acids Res 2012, 40:e92.

42. Wang R-Y, Shi Z-Y, Chen J-C, Chen G-Q: Cloning large geneclusters from E. coli using in vitro single-strand overlappingannealing. ACS Synth Biol 2012, 1:291-295.

43. Shao Z, Zhao H: DNA assembler, an in vivo genetic method forrapid construction of biochemical pathways. Nucleic Acids Res2009, 37:e16.

44. Wingler LM, Cornish VW: Reiterative recombination for the invivo assembly of libraries of multigene pathways. Proc NatlAcad Sci U S A 2011, 108:15135-15140.

45.�

Gibson DG, Glass JI, Lartigue C, Noskov VN, Chuang R-Y,Algire MA, Benders GA, Montague MG, Ma L, Moodie MM et al.:Creation of a bacterial cell controlled by a chemicallysynthesized genome. Science 2010, 329:52-56.

The authors demonstrated for the first time creating a completed self-replicating synthetic genome from chemically synthesized nucleotides.

46. Noskov VN, Karas BJ, Young L, Chuang R-Y, Gibson DG, Lin Y-C,Stam J, Yonemoto IT, Suzuki Y, Andrews-Pfannkoch C et al.:Assembly of large, high G + C bacterial DNA fragments inyeast. ACS Synth Biol 2012, 1:267-273.

47. Quan J, Tian J: Circular polymerase extension cloning for high-throughput cloning of complex and combinatorial DNAlibraries. Nat Protocols 2011, 6:242-251.

48. Hillson NJ, Rosengarten RD, Keasling JD: j5 DNA assemblydesign automation software. ACS Synth Biol 2011, 1:14-21.

49.��

Xu P, Vansiri A, Bhan N, Koffas MAG: ePathBrick: a syntheticbiology platform for engineering metabolic pathways in E. coli.ACS Synth Biol 2012, 1:256-266.

By allocating four compatible restriciton enzyme sites (AvrII, XbaI, SpeIand NheI) on strategic positions, the authors modified the five Duetvectors. The engineered ePathBrick vectors support modular assemblyof multi-gene pathways, combinatorial generation of pathway diversitiesin three distinct configurations, thus providing a versatile platform forconstruction and optimization of metabolic pathways in E. coli.

50. Vick JE, Johnson ET, Choudhary S, Bloch SE, Lopez-Gallego F,Srivastava P, Tikh IB, Wawrzyn GT, Schmidt-Dannert C: Optimizedcompatible set of BioBrickTM vectors for metabolic pathwayengineering. Appl Microbiol Biotechnol 2011, 92:1275-1286.



http://dx.doi.org/10.1016/j.phytochem.2012.05.002http://dx.doi.org/10.1016/j.phytochem.2012.05.002http://dx.doi.org/10.1016/j.copbio.2012.08.010http://dx.doi.org/10.1016/j.copbio.2012.08.010



51. Du L, Villarreal S, Forster AC: Multigene expression in vivo:supremacy of large versus small terminators for T7 RNApolymerase. Biotechnol Bioeng 2012, 109:1043-1050.

52. Norville JE, Derda R, Gupta S, Drinkwater KA, Belcher AM,Leschziner AE, Knight TF: Introduction of customized inserts fors-treamlined assembly and optimization of BioBrick syntheticgenetic circuits. J Biol Eng 2010, 4:17.

53. Peralta-Yahya PP, Ouellet M, Chan R, Mukhopadhyay A,Keasling JD, Lee TS: Identification and microbial production ofa terpene-based advanced biofuel. Nat Commun 2011, 2:483.

54.�

Wargacki AJ, Leonard E, Win MN, Regitsky DD, Santos CN,Kim PB, Cooper SR, Raisner RM, Herman A, Sivitz AB et al.: Anengineered microbial platform for direct biofuel productionfrom brown macroalgae. Science 2012, 335:308-313.

By transplanting a alginate-degrading gene cluster using fosmid, theauthors successfully constructed an E. coli strian that efficiently producedethanol directly from brown macroalgae.

55. Anthony J, Anthony L, Nowroozi F, Kwon G, Newman J, Keasling J:Optimization of the mevalonate-based isoprenoidbiosynthetic pathway in Escherichia coli for production of theanti-malarial drug precursor amorpha-4,11-diene. Metab Eng2009, 11:13-19.

56. Bokinsky G, Peralta-Yahya PP, George A, Holmes BM, Steen EJ,Dietrich J, Soon Lee T, Tullman-Ercek D, Voigt CA, Simmons BAet al.: Synthesis of three advanced biofuels from ionic liquid-pretreated switchgrass using engineered Escherichia coli.Proc Natl Acad Sci U S A 2011, 108:19949-19954.

57. Cox RS, Surette MG, Elowitz MB: Programming geneexpression with combinatorial promoters. Mol Syst Biol 2007:3.

58. Blazeck J, Liu L, Redden H, Alper H: Tuning gene expression inYarrowia lipolytica by a hybrid promoter approach. ApplEnviron Microbiol 2011, 77:7905-7914.

59. Michener JK, Thodey K, Liang JC, Smolke CD: Applications ofgenetically-encoded biosensors for the construction andcontrol of biosynthetic pathways. Metab Eng 2012, 14:212-222.

60. Salis H, Mirsky E, Voigt C: Automated design of syntheticribosome binding sites to control protein expression. NatBiotechnol 2009, 27 946–U112.

61. Ellis T, Wang X, Collins JJ: Diversity-based, model-guidedconstruction of synthetic gene networks with predictedfunctions. Nat Biotechnol 2009, 27:465-471.

62. Yadav VG, De Mey M, Giaw Lim C, Ajikumar PK,Stephanopoulos G: The future of metabolic engineering andsynthetic biology: towards a systematic practice. Metab Eng2012, 14:233-241.

63. Xu P, Ding Z, Qian Z, Zhao C, Zhang K: Improved production ofmycelial biomass and ganoderic acid by submerged culture ofGanoderma lucidum SB97 using complex media. EnzymeMicrob Technol 2008, 42:325-331.



64.��

Dueber JE, Wu GC, Malmirchegini GR, Moon TS, Petzold CJ,Ullal AV, Prather KLJ, Keasling JD: Synthetic protein scaffoldsprovide modular control over metabolic flux. Nat Biotechnol2009, 27:753-759.

By using an artifical protein scaffold, the authors spatially recruitedmevalonate pathway enzymes and achieved modular control over meta-bolic flux. This substrate channeling strategy has led to a 77-fold increasein mevalonate production in E. coli.

65. Moon TS, Dueber JE, Shiue E, Prather KLJ: Use of modular,synthetic scaffolds for improved production of glucaric acid inengineered E. coli. Metab Eng 2010, 12:298-305.

66.�

Sun J, Wen F, Si T, Xu J-H, Zhao H: Direct conversion of xylan toethanol by recombinant Saccharomyces cerevisiae strainsdisplaying an engineered minihemicellulosome. Appl EnvironMicrobiol 2012, 78:3837-3845.

The authors reported the construction of a trifunctional minihemicellulo-some on the yeast cell surface. Direct ethanol production from birchwoodxylan was achieved in this study.

67. Holtz WJ, Keasling JD: Engineering static and dynamic controlof synthetic pathways. Cell 2010, 140:19-23.

68.��

Zhang F, Carothers JM, Keasling JD: Design of a dynamicsensor-regulator system for production of chemicals andfuels derived from fatty acids. Nat Biotechnol 2012, 30:354-359.

Using a fatty acyl-CoA-responsive promoter, the authors dynamicallyregulated the genes involved in biodiesel synthesis and achieved 3-foldincrease in FAEEs production. This strategy provides a generalizableapproach for controlling gene expression in alomost any pathways ifcognate senosrs for the pathway intermediate exists.

69. Chubukov V, Zuleta IA, Li H: Regulatory architecture determinesoptimal regulation of gene expression in metabolic pathways.Proc Natl Acad Sci U S A 2012, 109:5127-5132.

70. Ma W, Trusina A, El-Samad H, Lim WA, Tang C: Defining networktopologies that can achieve biochemical adaptation. Cell 2009,138:760-773.

71.�

Wang HH, Isaacs FJ, Carr PA, Sun ZZ, Xu G, Forest CR,Church GM: Programming cells by multiplex genomeengineering and accelerated evolution. Nature 2009,460:894-898.

The authors described multiplex automated genome engineering (MAGE)for large-scale programming and evolution of cells. Using this strategy,the authors has successfully constructed an E. coli strain producing 5-fold more lycopene compared with parental strain.

72. Isaacs FJ, Carr PA, Wang HH, Lajoie MJ, Sterling B, Kraal L,Tolonen AC, Gianoulis TA, Goodman DB, Reppas NB et al.:Precise manipulation of chromosomes in vivo enablesgenome-wide codon replacement. Science 2011, 333:348-353.

73. Miller JC, Tan S, Qiao G, Barlow KA, Wang J, Xia DF, Meng X,Paschon DE, Leung E, Hinkley SJ et al.: A TALE nucleasearchitecture for efficient genome editing. Nat Biotechnol 2011,29:143-148.




Engineering plant metabolism into microbes: from systems biology to synthetic biologyIntroductionSynthetic metabolic pathway designSynthetic pathway construction: enabling technology to implement the vision of synthetic biologySynthetic pathway optimization through static, dynamic and spatial controlPerspectiveAcknowledgementsReferences and recommended reading

available online at koffam/papers/2012_xu_bhan_koffas.pdfxu ,2, namita bhan 1and mattheos ag koffas...

Documents