8/11/2019 Challenges to Synbio

1/3

Can engineering approaches tame the complexity of living systems? Roberta Kwokexplores five

challenges for the field and how they might be resolved.

T

o read some accounts of syntheticbiology, the ability to manipulate lifeseems restricted only by the imagi-

nation. Researchers might soonprogram cells to produce vast quantities ofbiofuel from renewable sources, or to sensethe presence of toxins, or to release precisequantities of insulin as a body needs it all

visions inspired by the idea that biologists canextend genetic engineering to be more likethe engineering of any hardware. The for-mula: characterize the genetic sequences thatperform needed functions, the parts, com-bine the parts into devicesto achieve more complexfunctions, then insert thedevices into cells. As all life

is based on roughly the samegenetic code, synthetic biol-ogy could provide a toolboxof reusable genetic compo-nents biological versionsof transistors and switches to be pluggedinto circuits at will.

Such analogies dont capture the dauntingknowledge gap when it comes to how lifeworks, however. There are very few molecu-lar operations that you understand in the waythat you understand a wrench or a screwdriveror a transistor, says Rob Carlson, a principalat the engineering, consulting and design

company Biodesic in Seattle, Washington.And the difficulties multiply as the networksget larger, limiting the ability to design morecomplex systems. A 2009 review1showed thatalthough the number of published syntheticbiological circuits has risen over the past fewyears, the complexity of those circuits orthe number of regulatory parts they use hasbegun to flatten out.

Challenges loom at every step in the process,from the characterization of parts to the designand construction of systems. Theres a lot ofbiology that gets in the way of the engineer-ing, says Christina Agapakis, a graduate

student doing synthetic-biology research atHarvard Medical School in Boston, Massa-

chusetts. But difficult biology is not enough todeter the fields practitioners, who are alreadyaddressing the five key challenges.

Many of the parts are undefinedA biological part can be anything froma DNA sequence that encodes a specific

protein to a promoter, a sequence that facili-tates the expression of a gene. The problem isthat many parts have not been characterizedwell. They havent always been tested to showwhat they do, and even when they have, theirperformance can change with different cell

types or under different labo-ratory conditions.

The Registry of StandardBiological Parts, which is

housed at the MassachusettsInstitute of Technology inCambridge, for example, hasmore than 5,000 parts availableto order, but does not guaran-

tee their quality, says director Randy Rettberg.Most have been sent in by undergraduatesparticipating in the International GeneticallyEngineered Machine (iGEM) competition, anannual event that started in 2004. In it, studentsuse parts from a kit or develop new ones todesign a synthetic biological system. But manycompetitors do not have the time to character-ize the parts thoroughly.

While trying to optimize lactose fermentationin microbes, an iGEM team from the Univer-sity of Pavia in Italy tested several pro-moters from the registry by placing themin Escherichia coli, a standard laboratory

bacterium. Most of the promoters tested by theteam worked, but some had little documenta-tion, and one showed no activity. About 1,500

registry parts have been confirmed as workingby someone other than the person who depos-ited them and 50 have reportedly failed, saysRettberg. Issues have been reported for roughlyanother 200 parts, and it is unclear how many ofthe remaining parts have been tested.

The registry has been stepping up efforts toimprove the quality by curating the collection,encouraging contributors to include docu-mentation on part function and performance,and sequencing the DNA of samples of partsto make sure they match their descriptions,says Rettberg. Meanwhile, synthetic biologistsAdam Arkin and Jay Keasling at the Univer-

sity of California, Berkeley, and Drew Endyat Stanford University in Stanford, Californiaare launching a new effort, tentatively calledBIOFAB, to professionally develop and char-acterize new and existing parts. Late last year,the team was awarded US$1.4 million by theNational Science Foundation and is hiringstaff, says Arkin. Endy, moreover, has proposedmethods to reduce some of the variability inmeasurements from different labs. By meas-uring promoter activity relative to a referencepromoter, rather than looking at absolute activ-ity, Endys team found that it could eliminatehalf the variation arising from experimental

conditions and instruments2

.

We are still like the

Wright Brothers,

putting pieces of woodand paper together.

Luis Serrano

M.K

NOWLES

J.SWART

Images such as these run in magazines

TheNew Yorker(left) and Wiredportray

synthetic biology as simple design and

construction. The truth is that many

of the parts are not well characterized,

or work unpredictably in different

configurations and conditions.

THE HYPEThe parts work like Lego

288

Vol 463|21 January 2010

288

NATURE|Vol 463|21 January 2010NEWS FEATURE

20 Macmillan Publishers Limited. All rights reserved10

8/11/2019 Challenges to Synbio

2/3

Measurements are tricky to standardize,however. In mammalian cells, for example,genes introduced into a cell integrate unpre-

dictably into the cells genome, and neighbour-ing regions often affect expression, says MartinFussenegger, a synthetic biologist at the SwissFederal Institute of Technology (ETH) Zurich.This is the type of complexity that is very dif-ficult to capture by standardized characteriza-tion, he says.

The circuitry is unpredictableEven if the function of each part isknown, the parts may not work as

expected when put together, says Keasling.Synthetic biologists are often caught in alaborious process of trial-and-error, unlike the

more predictable design procedures found inother modern engineering disciplines.

We are still like the Wright Brothers, puttingpieces of wood and paper together, saysLuis Serrano, a systems biologist at theCentre for Genomic Regulation in Bar-celona, Spain. You fly one thing andit crashes. You try another thing andmaybe it flies a bit better.

Bioengineer Jim Collins and his col-leagues at Boston University in Massa-chusetts crashed a lot when implementinga system called a toggle switch in yeast.His lab built one roughly ten years ago

in E. coli3

: the team wanted to make cellsexpress one gene call it gene A andthen prompt them with a chemical signalto turn off A and express another gene, B. Butthe cells refused to express B continuously;they always shifted back to expressing A. Theproblem, says Collins, was that the promoterscontrolling the two genes were not balanced, soA overpowered B. It took about three years oftweaking the system to make it work, he says.

Computer modelling could help reduce thisguesswork. In a 2009 study4, Collins and hiscolleagues created several slightly different ver-sions of two promoters. They used one version

of each to create a genetic timer, a system thatwould cause cells to switch from expressingone gene to another after a certain lag time.They then tested the timer, fed the results backinto a computational model and predicted howtimers built from other versions would behave.Using such modelling techniques, researcherscould optimize computationally rather thantest every version of a network, says Collins.

But designs might not have to work per-fectly: imperfect ones can be refined using aprocess called directed evolution, says FrancesArnold, a chemical engineer at the CaliforniaInstitute of Technology in Pasadena. Directed

evolution involves mutating DNA sequences,screening their performance, selecting the best

candidates and repeating the process until thesystem is optimized. Arnolds lab, for instance,is using the technique to evolve enzymesinvolved in biofuel production.

The complexity is unwieldyAs circuits get larger, the process ofconstructing and testing them becomes

more daunting. A system developed by Keas-lings team5, which uses about a dozen genes toproduce a precursor of the antimalarial com-

pound artemisinin in microbes, is perhaps thefields most cited success story. Keasling esti-mates that it has taken roughly 150 person-yearsof work including uncovering genes involved inthe pathway and developing or refining partsto control their expression. For example, theresearchers had to test many part variants beforethey found a configuration that sufficientlyincreased production of an enzyme needed toconsume a toxic intermediate molecule.

People dont even think about tacklingthose projects because it takes too much timeand money, says Reshma Shetty, co-founder ofthe start-up firm Ginkgo BioWorks in Boston,

Massachusetts. To relieve similar bottlenecks,Ginkgo is developing an automated process

to combine genetic parts. The parts havepre-defined flanking sequences, dictated by aset of rules called the BioBrick standard, and

can be assembled by robots.At Berkeley, synthetic biologist J. Christopher

Anderson and his colleagues are developing asystem that lets bacteria do the work. Engi-neered E. colicells, called assembler cells, arebeing equipped with enzymes that can cut andstitch together DNA parts. Other E. colicells,engineered to act as selection cells, will sortout the completed products from the leftoverparts. The team plans to use virus-like particlescalled phagemids to ferry the DNA from theassembler to the selection cells. Anderson saysthat the system could shorten the time neededfor one BioBrick assembly stage from two days

to three hours.

Many parts are incompatibleOnce constructed and placed intocells, synthetic genetic circuits can

have unintended effects on their host. ChrisVoigt, a synthetic biologist at the Universityof California, San Francisco, ran into thisproblem while he was a postdoc at Berkeleyin 2003. Voigt had assembled genetic parts,mainly from the bacterium Bacillus subtilis,into a switch system that was supposed to turnon expression of certain genes in response to achemical stimulus. He wanted to study the sys-

tem independently of B. subtilis other geneticnetworks, so he put the circuit into E. colibut it didnt work.

You looked under the microscope and thecells were sick, says Voigt. One day it woulddo one thing, and another day it would doanother thing. He eventually saw in the litera-ture that one of the circuits parts dramaticallydisrupted E. colis natural gene expression.There was nothing wrong with the design ofthe circuit, he says. It was just that one partwas not compatible.

Synthetic biologist Lingchong You at DukeUniversity in Durham, North Carolina, and

his colleagues found that even a simple circuit,comprising a foreign gene that promoted itsown expression, could trigger complex behav-iour in host cells6. When activated in E. coli, thecircuit slowed down the cells growth, whichin turn slowed dilution of the genes proteinproduct. This led to a phenomenon called bist-ability: some cells expressed the gene, whereasothers did not.

To lessen unexpected interactions, research-ers are developing orthogonal systems thatoperate independently of the cells naturalmachinery. Synthetic biologist Jason Chin ofthe Medical Research Council Laboratory of

Molecular Biology in Cambridge, UK, and hiscolleagues have created a protein-production

SLIMFILMS

H.C

AMPBELL

The magazines Scientific American (top) and

IEEE Spectrumportrayed synthetic biology as

being similar to microchip design or electrical

wiring. Although computational modelling may

help scientists to predict cell behaviour, the

cell is a complex, variable, evolving operating

system, very different from electronics.

THE HYPECells can simply be rewired

289

Vol 463|21 January 2010

289

NATURE|Vol 463|21 January 2010 NEWS FEATURE

20 Macmillan Publishers Limited. All rights reserved10

8/11/2019 Challenges to Synbio

3/3

system in E. colithat is separate from the cellsbuilt-in system7. To transcribe DNA into RNA,the team uses a polymerase enzyme that rec-ognizes genes only if they have a specific pro-moter sequence that is not present in the cellsnatural genes. Similarly, the systems orthogo-nal O-ribosomes, which translate RNA intoprotein, can read only O-mRNA that containsa specific sequence, and O-mRNA is unread-able by natural ribosomes.

A parallel system gives biologists the

freedom to tweak components without dis-rupting the machinery needed for the cell tosurvive, says Chin. For example, his team hasstripped down the DNA sequence encodingpart of the O-ribosome to speed up produc-tion. This allows the cell to boot up proteinmanufacture more quickly, he says.

Another solution is to physically isolate thesynthetic network from the rest of the cell.Wendell Lim, a synthetic biologist at the Uni-versity of California, San Francisco, is experi-menting with the creation of membrane-boundcompartments that would insulate the geneticcircuits. Lims team is working in yeast, but

similar principles could be applied to bacte-rial cells, he says.

Variability crashes the systemSynthetic biologists must also ensurethat circuits function reliably. Molecu-

lar activities inside cells are prone to randomfluctuations, or noise. Variation in growth con-ditions can also affect behaviour. And over thelong term, randomly arising genetic mutationscan kill a circuits function altogether.

Michael Elowitz, a synthetic biologist at theCalifornia Institute of Technology in Pasadena,observed the cells capacity for randomness

about ten years ago when his team built agenetic oscillator8. The system contained three

genes whose interactions caused the produc-tion of a fluorescent protein to go up and down,making cells blink on and off. However, notall cells responded the same way. Some werebrighter, and some were dimmer; some blinkedfaster, others slower; and some cells skipped acycle altogether.

Elowitz says that the differences might havearisen for multiple reasons. A cell can expressgenes in bursts rather than a steady stream.Cells also may contain varying amounts of

mRNA and protein-production machinery,such as polymerase enzymes and ribosomes.Furthermore, the number ofcopies of the genetic circuit ina cell can fluctuate over time.

Jeff Hasty, a syntheticbiologist at the University ofCalifornia, San Diego, andhis colleagues described anoscillator with more consistent behaviour9in 2008. Using a different circuit design andmicrofluidic devices that allowed fine controlof growth conditions, the team made nearlyevery monitored cell blink at the same rate

though not in sync. And in this issue ofNature

(see page 326)10, Hastys team reports the abil-ity to synchronize the blinking by relying oncellcell communication. But Hasty says thatrather than trying to eliminate noise, research-ers could use it to their advantage. He notes thatin physics, noise can sometimes make a signaleasier to detect. I dont think you can beat it,so I think you ought to try to use it, says Hasty.For example, noise could allow some cells torespond differently to the environment fromothers, enabling the population to hedge itsbets, says Elowitz.

Meanwhile, geneticist George Church

at Harvard Medical School in Boston,Massachusetts, is exploring ways to make a

bacterial strain more stable. Church saysthat this might be achieved by introducingmore accurate DNA-replication machin-

ery, changing genome sites to make themless prone to mutation and putting extracopies of the genome into cells. Althoughstability may not be a serious issue forsimple systems, it will become important asmore components are assembled, he says.

Time to deliver?Despite the challenges, synthetic biologistshave made progress. Researchers haverecently developed devices that allowE. coli tocount events such as the number oftimes they have divided and to detect lightand dark edges. And some systems have

advanced from bacteria to more complexcells. The field is also gaining legitimacy,with a new synthetic-biology centre at Impe-rial College London and a programme at

Harvard Universitys recently launched WyssInstitute for Biologically Inspired Engineer-ing in Boston. The time has come for syn-thetic biologists to develop more real-worldapplications, says Fussenegger. The field hashad its hype phase, he says. Now it needs todeliver.

Keaslings artemisinin precursor system isapproaching commercial reality, with Paris-based pharmaceutical company Sanofi-Aventis

aiming to have the product available at anindustrial scale by 2012. And several com-

panies are pursuing biofuelproduction via engineeredmicrobes. But most applica-tions will take time.

As the cost of DNA syn-thesis continues to drop andmore people begin to tinker

with biological parts, the field could progressfaster, says Carlson. Its a question of whetherthe complexity of biology yields to that kind ofan effort. Roberta Kwok is a freelance writer in the San

Francisco Bay Area.

1. Purnick, P. E. M. & Weiss, R. Nature Rev. Mol. Cell Biol. 10,410422 (2009).

2. Kelly, J. R. et al. J. Biol. Engineer.3,4(2009).3. Gardner, T. S., Cantor, C. R. & Collins, J. J.Nature 403,

339342 (2000).4. Ellis, T., Wang, X. & Collins, J. J. Nature Biotechnol. 27,

465471 (2009).5. Ro, D.-K. et al. Nature 440, 940943 (2006).6. Tan, C., Marguet, P. & You, L. Nature Chem. Biol. 5, 842848

(2009).7. An, W. & Chin, J. W.Proc. Natl Acad. Sci. USA 106,

84778482 (2009).8. Elowitz, M. B. & Leibler, S. Nature 403,335338 (2000).9. Stricker, J. et al. Nature 456, 516519 (2008).10. Danino, T., Mondragn-Palomino, O., Tsimring, L. &

Hasty, J. Nature 463, 326330 (2010).

See Editorial, page 269.

The field has had its

hype phase. Now it

needs to deliver.

Martin Fussenegger

ISSUE1OFTHEADVENTURESINSYNTHETICBIOLOGY.S

TORY:DREWE

NDY&ISADORADEESE.A

RT:CHUCKWADEY

REVPAGE/ETCGROUPWWW.ETCGROUP.ORG

THE HYPEA promise of unprecedented powerNaturehas portrayed synthetic biologists as wieldingthe power to hack life (right), and in its Guide to

Synthetic Biology,the civil society organization ETC

Group even likened their activity to playing God. But in

reality, the field has yet to deliver much of practical use.

290

Vol 463|21 January 2010

290

NATURE|Vol 463|21 January 2010NEWS FEATURE

20 Macmillan Publishers Limited. All rights reserved10