Synthetic Biology E-Book

• Herausgeber: Wiley-VCH Verlag GmbH & Co. KGaA
• Kategorie: Fachliteratur
• Serie: Advanced Biotechnology
• Sprache: Englisch

Seitenzahl: 878

Veröffentlichungsjahr: 2018

Beschreibung

A review of the interdisciplinary field of synthetic biology, from genome design to spatial engineering.

Written by an international panel of experts, Synthetic Biology draws from various areas of research in biology and engineering and explores the current applications to provide an authoritative overview of this burgeoning field. The text reviews the synthesis of DNA and genome engineering and offers a discussion of the parts and devices that control protein expression and activity. The authors include information on the devices that support spatial engineering, RNA switches and explore the early applications of synthetic biology in protein synthesis, generation of pathway libraries, and immunotherapy.

Filled with the most recent research, compelling discussions, and unique perspectives, Synthetic Biology offers an important resource for understanding how this new branch of science can improve on applications for industry or biological research.

Leseprobe

Table of Contents

Cover

Title page

Copyright

About the Series Editors

Part I: DNA Synthesis and Genome Engineering

Chapter 1: Competition and the Future of Reading and Writing DNA

1.1 Productivity Improvements in Biological Technologies

1.2 The Origin of Moore’s Law and Its Implications for Biological Technologies

1.3 Lessons from Other Technologies

1.4 Pricing Improvements in Biological Technologies

1.5 Prospects for New Assembly Technologies

1.6 Beyond Programming Genetic Instruction Sets

1.7 Future Prospects

References

Chapter 2: Trackable Multiplex Recombineering (TRMR) and Next‐Generation Genome Design Technologies: Modifying Gene Expression in

E. coli

by Inserting Synthetic DNA Cassettes and Molecular Barcodes

2.1 Introduction

2.2 Current Recombineering Techniques

2.3 Trackable Multiplex Recombineering

2.4 Current Challenges

2.5 Complementing Technologies

2.6 Conclusions

References

Chapter 3: Site‐Directed Genome Modification with Engineered Zinc Finger Proteins

3.1 Introduction to Zinc Finger DNA‐Binding Domains and Cellular Repair Mechanisms

3.2 Approaches for Engineering or Acquiring Zinc Finger Proteins

3.3 Genome Modification with Zinc Finger Nucleases

3.4 Validating Zinc Finger Nuclease‐Induced Genome Alteration and Specificity

3.5 Methods for Delivering Engineered Zinc Finger Nucleases into Cells

3.6 Zinc Finger Fusions to Transposases and Recombinases

3.7 Conclusions

References

Chapter 4: Rational Efforts to Streamline the

Escherichia coli

Genome

4.1 Introduction

4.2 The Concept of a Streamlined Chassis

4.3 The

E. coli

Genome

4.4 Random versus Targeted Streamlining

4.5 Selecting Deletion Targets

4.6 Targeted Deletion Techniques

4.7 Genome‐Reducing Efforts and the Impact of Streamlining

4.8 Selected Research Applications of Streamlined‐Genome

E. coli

4.9 Concluding Remarks, Challenges, and Future Directions

References

Chapter 5: Functional Requirements in the Program and the Cell Chassis for Next‐Generation Synthetic Biology

5.1 A Prerequisite to Synthetic Biology: An Engineering Definition of What Life Is

5.2 Functional Analysis: Master Function and Helper Functions

5.3 A Life‐Specific Master Function: Building Up a Progeny

5.4 Helper Functions

5.5 Conclusion

Acknowledgments

References

Part II: Parts and Devices Supporting Control of Protein Expression and Activity

Chapter 6: Constitutive and Regulated Promoters in Yeast: How to Design and Make Use of Promoters in

S. cerevisiae

6.1 Introduction

6.2 Yeast Promoters

6.3 Natural Yeast Promoters

6.4 Synthetic Yeast Promoters

6.5 Conclusions

References

Chapter 7: Splicing and Alternative Splicing Impact on Gene Design

7.1 The Discovery of “Split Genes”

7.2 Nuclear Pre‐mRNA Splicing in Mammals

7.3 Splicing in Yeast

7.4 Splicing without the Spliceosome

7.5 Alternative Splicing in Mammals

7.6 Controlled Splicing in

S. cerevisiae

7.7 Splicing Regulation by Riboswitches

7.8 Splicing and Synthetic Biology

Acknowledgments

Chapter 8: Design of Ligand‐Controlled Genetic Switches Based on RNA Interference

8.1 Utility of the RNAi Pathway for Application in Mammalian Cells

8.2 Development of RNAi Switches that Respond to Trigger Molecules

8.3 Rational Design of Functional RNAi Switches

8.4 Application of the RNAi Switches

8.5 Future Perspectives

References

Chapter 9: Small Molecule‐Responsive RNA Switches (Bacteria): Important Element of Programming Gene Expression in Response to Environmental Signals in Bacteria

9.1 Introduction

9.2 Design Strategies

9.3 Mechanisms

9.4 Complex Riboswitches

9.5 Conclusions

References

Chapter 10: Programming Gene Expression by Engineering Transcript Stability Control and Processing in Bacteria

10.1 An Introduction to Transcript Control

10.2 Synthetic Control of Transcript Stability

10.3 Managing Transcript Stability

10.4 Potential Mechanisms for Transcript Control

10.5 Conclusions and Discussion

Acknowledgments

References

Chapter 11: Small Functional Peptides and Their Application in Superfunctionalizing Proteins

11.1 Introduction

11.2 Permissive Sites and Their Identification in a Protein

11.3 Functional Peptides

11.4 Conclusions

Acknowledgments

References

Part III: Parts and Devices Supporting Spatial Engineering

Chapter 12: Metabolic Channeling Using DNA as a Scaffold

12.1 Introduction

12.2 Biosynthetic Applications of DNA Scaffold

12.3 Design of DNA‐Binding Proteins and Target Sites

12.4 DNA Program

12.5 Applications of DNA‐Guided Programming

References

Chapter 13: Synthetic RNA Scaffolds for Spatial Engineering in Cells

13.1 Introduction

13.2 Structural Roles of Natural RNA

13.3 Design Principles for RNA Are Well Understood

13.4 Applications of Designed RNA Scaffolds

13.5 Conclusion

References

Chapter 14: Sequestered: Design and Construction of Synthetic Organelles

14.1 Introduction

14.2 On Organelles

14.3 Protein‐Based Organelles

14.4 Lipid‐Based Organelles

14.5

De novo

Organelle Construction and Future Directions

Acknowledgments

References

Part IV: Early Applications of Synthetic Biology: Pathways, Therapies, and Cell-Free Synthesis

Chapter 15: Cell‐Free Protein Synthesis: An Emerging Technology for Understanding, Harnessing, and Expanding the Capabilities of Biological Systems

15.1 Introduction

15.2 Background/Current Status

15.3 Products

15.4 High‐Throughput Applications

15.5 Future of the Field

Acknowledgments

References

Chapter 16: Applying Advanced DNA Assembly Methods to Generate Pathway Libraries

16.1 Introduction

16.2 Advanced DNA Assembly Methods

16.3 Generation of Pathway Libraries

16.4 Conclusions and Prospects

References

Chapter 17: Synthetic Biology in Immunotherapy

17.1 The Need for a New Therapeutic Paradigm

17.2 Rationale for Cellular Therapies

17.3 Synthetic Biology Approaches to Cellular Immunotherapy Engineering

17.4 Challenges and Future Outlook

Acknowledgment

References

Part V: Societal Ramifications of Synthetic Biology

Chapter 18: Synthetic Biology: From Genetic Engineering 2.0 to Responsible Research and Innovation

18.1 Introduction

18.2 Public Perception of the Nascent Field of Synthetic Biology

18.3 Frames and Comparators

18.4 Toward Responsible Research and Innovation (RRI) in Synthetic Biology

18.5 Conclusion

Acknowledgments

References

Index

End User License Agreement

List of Tables

Chapter 4: Rational Efforts to Streamline the

Escherichia coli

Genome

Table 4.1 Published biotechnology‐related applications of streamlined‐genome

E. coli

.

Chapter 8: Design of Ligand‐Controlled Genetic Switches Based on RNA Interference

Table 8.1 Developed RNAi switches.

Chapter 11: Small Functional Peptides and Their Application in Superfunctionalizing Proteins

Table 11.1 Available functional peptide tags.

Chapter 12: Metabolic Channeling Using DNA as a Scaffold

Table 12.1 Advantages and disadvantages between DNA, protein, and RNA scaffolds.

Table 12.2 DNA scaffolds used to order enzymes of biosynthetic pathways.

Chapter 13: Synthetic RNA Scaffolds for Spatial Engineering in Cells

Table 13.1 Comparison of features between RNA structure prediction software packages.

Chapter 16: Applying Advanced DNA Assembly Methods to Generate Pathway Libraries

Table 16.1 Summary of different advanced DNA methods that could be used for combinatorial library generation.

Chapter 17: Synthetic Biology in Immunotherapy

Table 17.1 Major categories of cell‐based immunotherapies and application areas currently under investigation.

List of Illustrations

Chapter 1: Competition and the Future of Reading and Writing DNA

Figure 1.1 Estimates of the maximum productivity of DNA synthesis and sequencing enabled by commercially available instruments. Productivity of DNA synthesis is shown only for column‐based synthesis instruments, as data for sDNA fabricated on commercially available DNA arrays is unavailable; exceptions are discussed in the text. Shown for comparison is Moore’s law, the number of transistors per chip.

Figure 1.2 Commercial prices per base for DNA sequencing, column‐synthesized oligonucleotides, and gene‐length sDNA. Reported prices for array‐synthesized oligos vary widely, and no time series is available. Market pricing for genes can vary by up to an order of magnitude, depending on sequencing composition and complexity.

Chapter 2: Trackable Multiplex Recombineering (TRMR) and Next‐Generation Genome Design Technologies: Modifying Gene Expression in

E. coli

by Inserting Synthetic DNA Cassettes and Molecular Barcodes

Figure 2.1 The λ‐Red system and the replication fork annealing model of recombination. Either a double‐ or single‐stranded recombineering substrate, consisting of the DNA sequence to be inserted flanked by homology arms, is transformed into cells. The λ‐Red proteins facilitate recombination by digesting one strand of DNA in the case of dsDNA (Exo), by inhibiting RecBCD nuclease activity (Gam), and by protecting and conveying the ssDNA to the replication fork (Beta). Then, the ssDNA acts as a mismatched Okazaki fragment and binds to the lagging strand via its homology arms. This process results, upon completion of cell duplication, with one wild‐type daughter cell and one recombineered, heterozygous‐like daughter cell.

Figure 2.2 Overview of trackable multiplex recombineering (a) and tunable trackable multiplex recombineering (b). TRMR and T

2

RMR cassettes are designed and synthesized in multiplex followed by transformation into

Escherichia coli

. The

E. coli

population is then placed under selective pressure. Alleles that are enriched during selection are identified by microarray (TRMR) or next‐generation sequencing (T

2

RMR), and their relative fitness is determined. Cassette design for each technique is shown at the top. Black regions are shared DNA and gray regions are from the targeting oligos. HA1 and HA2, homology regions; P, barcode priming site; G, barcode identifying the gene; B, barcode identifying the BCD; BlastR, blasticidin resistance gene; KanR, kanamycin resistance gene; stop, three frame stop codons; Ts, terminator spacer; Tp, terminator pause; Pi, promoter insulator; LacIO, LacI‐regulated synthetic inducible promoter (apFAB906); BCD, bicistronic design (dual RBS).

Figure 2.3 Construction and incorporation of TRMR (a) and T

2

RMR (b) cassettes. In both TRMR and T

2

RMR, targeting oligos are ligated with a shared DNA cassette encoding a specific genetic function. The ligated synDNA cassettes are amplified by rolling circle amplification and then cleaved to create a linear dsDNA substrate. (c) The linear synDNA cassettes are recombineered into cells, targeting all genes at one time.

Figure 2.4 Validation of TRMR (a) and T

2

RMR (b) cassettes using the

lacZ

gene. In TRMR, the “up” cassette causes

lacZ

to be expressed, while the “down” cassette results in a loss of expression. In T

2

RMR, varying the four libraries (“off,” “low,” “intermediate,” and “high”), and the amount of inducer (IPTG) allows

lacZ

to be expressed over a ~10

4

‐fold range.

Figure 2.5 T

2

RMR has significantly increased ability to discriminate between MOPS minimal medium and LB‐rich medium. The Pearson dissimilarity (0 indicates perfectly linearly correlated, and 2 indicates negatively correlated) between MOPS and LB samples for each library type is shown. * indicates p < 0.05 Benjamini−Hochberg corrected significance.

Chapter 3: Site‐Directed Genome Modification with Engineered Zinc Finger Proteins

Figure 3.1 Targeted genomic modification using zinc finger nucleases (ZFNs). A pair of ZFPs fused to the

Fok

I nuclease domain is designed to target opposite strands of DNA. When dimerization of the

Fok

I domain occurs following ZF binding, a double‐strand break (DSB) is created in the DNA (shown by lightning bolt). The cell chooses to use either the nonhomologous end joining (NHEJ) or homologous recombination (HR) pathway to repair the DSB. NHEJ can be used for gene disruption via targeted mutagenesis using one pair of ZFNs or deletions/inversions with two pairs of ZFNs. If gene correction via HR is desired, a homologous template sequence is provided that will be used by the cellular HR machinery replace the endogenous sequence near the DSB. Alternatively, targeted gene addition at or near the site of the targeted DNA cleavage can be achieved by flanking the sequence to be inserted with homologous arms.

Figure 3.2 A nuclease assay for detecting gene targeting. ZFNs are used to create a targeted DNA DSB. PCR is used to amplify the targeted sites and the DNA is heated and cooled to reanneal the strands, creating mismatched heteroduplexes of DNA. The Surveyor nuclease cleaves the heteroduplexes only at the sites of mismatched DNA, leaving the homodimers unmodified. Gel electrophoresis can then be used to observe and quantitate the efficiency of gene targeting using ZFNs at the genomic level.

Chapter 4: Rational Efforts to Streamline the

Escherichia coli

Genome

Figure 4.1 Schematic map of selected features of the

E. coli

K‐12 MG1655 genome, numbered on the perimeter in base pair. Outward from the center, rings depict (1) strain‐specific K‐12 genomic islands longer than 4 kbp [29], (2) essential genes (www.shigen.nig.ac.jp/ecoli/pec/index.jsp), (3) ribosomal RNA operons, (4) IS elements, (5) prophages [18], and (6) macrodomains [30]. Ori and ter indicate the origin and terminus of replication, respectively.

Figure 4.2 Comparison of the pan‐genome and core‐genome sizes, defined by homologous gene clusters (HGCs). Data and classification criteria are from [33] and are based on the analysis of 186 sequenced

E. coli

genomes. HGCs are generated by sequence similarity (95% of HGCs have <0.242 substitutions per site). The soft‐core genome is defined as all HGCs that have members in at least 95% of the 186 genomes. The strict core genome is defined as all HGCs that have members in all genomes. The pan‐genome is defined as all HGCs.

Figure 4.3 General scheme of standard deletion procedures. (a) Overview of the circular DNA‐based method. Boxes A and B represent >500‐bp DNA segments flanking the genomic region to be deleted. Ab

R

stands for an antibiotic resistance marker gene; ori indicates a replication origin functioning only under permissive conditions. (b) Overview of the λ‐Red‐mediated, linear DNA‐based deletion method. Two alternative routes for generating deletions are shown. A, B, and C represent arbitrarily chosen 40–60‐bp DNA segments (homology boxes). Arrowheads represent I‐SceI cleavage sites. Ab

R

and csm stand for an antibiotic resistance marker and a counterselectable gene, respectively.

Figure 4.4 Deletion map of reduced‐genome

E. coli

strains. Rings depict features mapped to the genome of

E. coli

K‐12 MG1655, numbered on the perimeter in kilobase pair. Outward from the center, (1) strain‐specific K‐12 genomic islands longer than 4 kbp [96], (2) essential genes (www.shigen.nig.ac.jp/ecoli/pec/index.jsp), and (3)–(8) set of deletions constructed by Goryshin

et al

. [43], Yu

et al

. [73], Hashimoto

et al

. [97], Pósfai

et al

. (MDS42: black boxes, MDS69: black and gray boxes) [29, 85], Mizoguchi

et al

. (MGF‐01) [47], and Hirokawa

et al

. (DGF‐298) [98], respectively. Ori and ter indicate the origin and terminus of replication, respectively.

Figure 4.5 Hypothetical relationship between the fitness of the cell and the extent of genome streamlining.

Chapter 5: Functional Requirements in the Program and the Cell Chassis for Next‐Generation Synthetic Biology

Figure 5.1 A schematic view of functional analysis [5]. Master and helper functions are as defined in the text.

Figure 5.2 Replication of the program, reproduction of the chassis. The sequence of the

Mycoplasma mycoides

DNA transplanted into

Mycoplasma capricolum

is identical to that at the end of the experiment. Transplantation has triggered degradation of the

M. capricolum

DNA, while the DNA of

M. mycoides

replicates and dilutes out the component of the initial host. At the end of the experiment, the components of the cells are identical to those of

M. mycoides

, not to those of the recipient

M. capricolum

.

Figure 5.3 Excerpt of the metabolism of pyrimidines and DNA synthesis. The building blocks for DNA stem from NDPs, not nucleoside triphosphates (NTPs). This creates an imbalance in the case of cytosine, because CDP is not produced during the

de novo

synthesis. This explains why, in general, C is the limiting nucleotide, driving A+T enrichment of the genome in most situations. Nucleoside diphosphokinase is reversible; however ATP is in excess over adenosine diphosphate (ADP), so that production of CDP is limiting via this route. CDP comes mainly from mRNA turnover via phosphorolysis (polynucleotide phosphorylase) or RNase activity, with further phosphorylation using cytidylate kinase.

Chapter 6: Constitutive and Regulated Promoters in Yeast: How to Design and Make Use of Promoters in

S. cerevisiae

Figure 6.1 Modified natural yeast promoters. Top, typical bipartite structure of yeast promoters. Bottom left, promoter libraries obtained by point mutation. Random point mutations, illustrated as stars, are introduced by error‐prone PCR along the sequence of the starting promoter. Bottom right, promoter libraries obtained by substituting non‐consensus sequences with random oligonucleotides. By concentrating the mutations in the non‐consensus regions, it is possible to fine‐tune the strength of the starting promoter. N: nucleotide; ORF: open reading frame; TATA: TATA element; TFBS: transcription factor binding site; TSS: transcription initiation start site.

Figure 6.2 Synthetic hybrid promoters are obtained by combining a core promoter and one or more transcription factor binding sites (TFBSs). Each TFBS is specifically recognized by a transcription factor. By selecting the TFBSs, it is possible to choose which regulation the synthetic hybrid promoter should display. The combination of two or more different TFBSs results in a combinatorial regulation. The multiplication of the TFBS copy number results in the adjustment of the promoter strength.

Chapter 8: Design of Ligand‐Controlled Genetic Switches Based on RNA Interference

Figure 8.1 Schematic of the RNAi pathway in mammalian cells, including endogenous miRNA processing and ectopic shRNA or siRNA expression.

Figure 8.2 RNAi switch design strategies with a variety of trigger molecules. The RNA motifs that bind to specific trigger molecules are introduced into the appropriate regions in pri‐miRNA, pre‐miRNA, shRNA, or siRNA. The motifs embedded in the RNA are then optimized to generate functional RNAi switches.

Figure 8.3 3D design scheme of the protein‐responsive shRNA switch. (a) Protein‐responsive shRNA switches are three‐dimensionally designed by connecting an RNP motif to the corresponding dsRNA

in silico

. After specifying the position of the devices with reference to the Dicer cleavage sites and comparing the locations of the bound proteins of the X and Y switches, the bound trigger protein on switch X rotates ~30° in a counterclockwise direction around the axis of the dsRNA and is located ~2.6 Å more distant from the Dicer cleavage sites than switch Y. (b) Dicer can access and process switch Y via trigger protein binding; the switch then induces the knockdown of its target gene via RNAi (left). Dicer is inaccessible to switch X in the presence of the trigger protein because the RNP interaction faces Dicer and inhibits its access (right). The prevention of Dicer function causes the derepression of gene knockdown.

Figure 8.4 Applications of RNAi switches. RNAi ON/OFF switches can be applied to RNAi reporter and cell fate conversion systems that respond to specific trigger molecules.

Chapter 9: Small Molecule‐Responsive RNA Switches (Bacteria): Important Element of Programming Gene Expression in Response to Environmental Signals in Bacteria

Figure 9.1 (a,b) Examples of synthetic riboswitch libraries.

Chapter 10: Programming Gene Expression by Engineering Transcript Stability Control and Processing in Bacteria

Figure 10.1 Primary routes for RNase E‐mediated mRNA degradation. “5′ entry” is initiated when an mRNA undergoes 5′‐PP removal, catalyzed by the pyrophosphohydrolase enzyme RppH, creating a 5′‐P that can be recognized and bound by RNase E (shown by “+” symbol). “Direct entry” (at right) is 5′ independent entry by RNase E that occurs without recognition and binding to a 5′‐P moiety. Following RNase E binding, an initial cleavage event generates 3′‐OH‐ and 5′‐P‐terminated RNAs that are efficient substrates for 3′ → 5′ degradation to monomers and further rounds of RNase E binding and cleavage.

Figure 10.2 Naturally occurring transcript stability control mechanisms. (a) mRNAs that are highly occupied by translating ribosomes RNA will have occluded sites for RNase E 5′ entry and direct entry, leading to relatively long transcript half‐life and high levels of gene expression. Exposed transcripts (i.e., with lower ribosome density), such as mRNAs transcribed with bacteriophage polymerases with fast elongation rates, are more susceptible to RNase E attack due to a lack of occluding ribosomes. (b) sRNA and asRNA operate through Hfq‐mediated binding to the RBS (green box) and/or start codon region of a target mRNA, which prevents ribosome docking and likely recruits RNase E to the transcript. (c) The addition of poly(A) tails to a transcript, usually by poly(A) polymerase (PAP I), creates a foothold for binding by polynucleotide phosphorylase (PNPase), a 3′ → 5′ exoribonuclease.

Figure 10.3 Examples of engineered transcript stability control. (a) Synthetic secondary structure hairpins within the 5′ UTR can increase half‐life by preventing RNase E 5′ entry; direct entry by RNase E is still possible. (b) 5′ ribozyme‐mediated transcript cleavage creates a 5′‐OH not recognized by RppH and therefore cannot become a 5′‐P for RNase E to bind; degradation of these processed transcripts occurs through RNase E direct entry. (c) Riboregulators and riboswitches typically work by

cis

‐RNA sequestration of the RBS, which can be relieved by either

trans

‐RNA binding to the

cis

‐RN or ligand binding to the

cis

‐RNA. These binding events free the RBS from the

cis

‐RNA and therefore allow translation.

Figure 10.4 Model‐driven design workflow for engineering gene expression with TSC. A basic systems‐level model is used to identify goals for genetic device outputs. These goals inform the creation of a mechanistic model based on biochemical understanding, which is used to identify component specifications needed for device function. Components are then engineered and/or evolved with

in vitro

selection to meet the design specifications. Transcript design methods employing biophysical models of RNA folding are employed to enable the assembly of individual RNA components into functional devices. The mechanistic model is then refined to account for engineered component characteristics used to predict device outputs. Systems‐level functions are obtained through the assembly of multiple static and dynamic RNA devices.

Figure 10.5 Examples of unutilized transcript stability control mechanisms. (a) In naturally occurring biological systems, decoy RNAs attenuate sRNA‐mediated mRNA degradation by shunting sRNAs away from cognate transcripts, increasing message stability and the level of genetic expression. (b) The

lysC

transcript of

E. coli

contains a riboswitch that controls access to the RBS and RNase E target sites, showing the functional integration of multiple TSC mechanisms.

Chapter 12: Metabolic Channeling Using DNA as a Scaffold

Figure 12.1 Spatial and temporal organization of biosynthetic enzymes based on different types of scaffolds. (a) Biosynthetic enzymes are typically randomly distributed inside the cell. The conversion of the substrate may therefore be limited by the diffusion rate and the concentration of the substrate and localization of the enzymes. (b) Immobilizing biosynthetic enzymes using synthetic protein scaffolds can bring the enzymes into close proximity and therefore enhance the metabolic flux. In the absence of a large superscaffold, the precise arrangement of enzymes is unpredictable and is limited by the tertiary structure of the protein scaffold. (c) Biosynthetic enzymes with predictable RNA binding sites have been assembled using synthetic RNA aptamers. The enzymes are in close proximity and in the predefined order, which enables faster conversion of the substrate into the end product. (d) An assembly line based on the DNA scaffold promotes positioning of biosynthetic enzymes in close proximity and the predefined order. The substrate conversion is faster with less unwanted side products. The enzymes are linked to DNA‐binding domains, which recognize specific nucleotide sequences. The order of the enzymes can easily be changed by changing the order of the specific nucleotide sequence on the DNA program, which can lead to different end products.

Figure 12.2 The biosynthesis of

L

‐threonine in

E. coli

is enhanced by the DNA scaffold. (a) The three‐step conversion of aspartate semialdehyde to

L

‐threonine. (b) Arrangements of DNA‐target sites on the DNA program with indicated production rates for

L

‐threonine are depicted. The DNA scaffold includes the chimeric proteins, homoserine dehydrogenase (HDH; E1), homoserine kinase (HK; E2), and threonine synthase (TS; E3), fused to DNA‐binding domains (ADB). Consecutive arrangements of DNA‐target sites for threonine synthase (E3), the third enzyme in the biosynthesis of

L

‐threonine, improved the production rate for

L

‐threonine. The DNA‐target sites specific for the individual chimeric proteins are separated with 8‐, 18‐, or 28‐bp spacers between each DNA‐target site. The fastest rate of

L

‐threonine production in

E. coli

was obtained with the DNA program [1 : 1 : 2], with DNA‐binding sites separated by 8 bp (see also [10]).

Figure 12.3 A DNA scaffold enhances the biosynthesis of

trans

‐resveratrol in

E. coli

. (a) In the biosynthetic pathway of resveratrol, the 4‐cumaric acid is converted to resveratrol in a two‐step reaction with the biosynthetic enzymes 4‐coumarate–CoA ligase (4CL) and stilbene synthase (STS). (b) Close proximity of the 4CL and STS enzymes can be achieved by fusing the enzymes with linker polypeptides or by introducing DNA scaffolds where the enzymes (4CL or STS) are fused to the DNA‐binding domains (Zif268 or PBSII). The chimeric protein of the enzyme and DNA‐binding domain binds to a specific nucleotide sequence present on the DNA program. The DNA‐target sites specific for the individual chimeric proteins are separated with a 2‐bp spacer between each of four tandem repeats [11].

Figure 12.4 Biosynthesis of (a) 1,2‐propanediol and (b) mevalonate in

E. coli

. (c) Schemes of consecutive, bidirectional, and mixed consecutive and bidirectional arrangements of DNA scaffolds with different stoichiometry and positions of DNA‐target sites that were tested for the improved biosynthesis of 1,2‐propanediol or mevalonate. DNA scaffolds can be used to overcome the limitations in biosynthetic pathways that occur because of individual enzymes with lower activity, compared with other enzymes in the same biosynthetic pathway. By changing the order or number of DNA‐target sites, we can increase reaction yields, fine‐tune biosynthesis production, and minimize side products. If the first enzyme in the biosynthetic pathway is most active, others can be distributed on both sides around the first, resulting in 1 : 2 molar ratios in favor of enzymes with low activity. Such groups of enzyme binding sites can then be multiplied on the DNA scaffold to achieve better molar ratios between the DNA scaffold and enzymes. (d) Impact of different scaffold architectures on 1,2‐propanediol and mevalonate production [11].

Figure 12.5 Targeting DNA

in vitro

and

in vivo

with zinc finger domains. (a) The binding affinity of zinc finger domains (e.g., Zif268) to their specific nucleotide target sequence was determined using surface plasmon resonance (SPR). With increasing concentrations of purified zinc finger protein, the response signal increases, indicating protein binding. (b) The zinc finger domain (PBSII) was fused to the N‐terminal half (PBSII‐nYFP), and Zif268 was fused to the C‐terminal half (cYFP‐Zif268) of the yellow fluorescent protein (YFP). Purified PBSII‐nYFP and cYFP‐Zif268 protein chimeras were mixed, either with DNA scaffolds, containing PBSII, or Zif268 target sites separated by 2‐bp spacer, or DNA scaffolds with random nucleotide sequences. Fluorescence was then measured [11]. (c) The binding of the DNA‐binding domain (e.g., Zif268)

in vivo

was tested with the inhibition of β‐galactosidase expression. The expression of the tested zinc finger was under the control of an arabinose‐inducible promoter. The

lacZ

gene was controlled by the P

SYN

promoter, which contained either the zinc finger target site or random DNA target site (CTCTATCAATGATAGAG). β‐Galactosidase activity is measured in the presence of 1% or absence (0%) of arabinose and normalized to the galactosidase levels of the unrepressed state. The β‐galactosidase activity is detected when the DNA‐binding protein (e.g., zinc finger A) is not expressed (no arabinose). Arabinose induces the expression of zinc finger A, which binds to the DNA‐target site “a” upstream of the β‐galactosidase gene, and represses the expression of β‐galactosidase. If the DNA‐target site “b” is not recognized by the DNA‐binding protein, the expression of β‐galactosidase is not affected.

Figure 12.6 Spatial position of biosynthetic enzymes is defined by DNA program. (a) Scheme of two types of DNA scaffolds that differ in spacer lengths separating the DNA‐target sites. A first scaffold plasmid (left) carries four copies of Zif268 and four copies of PBSII binding sites, separated by an insertion of 850 bp along part of the plasmid backbone. A second scaffold plasmid carries four copies of the Zif268 and PBSII binding sites separated by 2‐bp‐long spacers. (b) Enzyme clustering improves the production of

trans

‐resveratrol, which was measured in

Escherichia coli

‐expressing fusion enzymes (Zif268‐4CL and PBSII‐STS) with different DNA program plasmids [1]

4

‐850 bp‐[1]

4

and [1 : 1]

4

2‐bp spacers (for details see Figure 12.4a) [11]. (c) The spatial orientation of the enzymes is governed with a spacer between the DNA‐target sites. The 2‐ and 8‐bp spacers orientate chimeric enzymes on the same side of the DNA program (up). The 4‐bp spacer between the target sites orientates the enzymes on opposite sides of the DNA program (below). The best production of

trans

‐resveratrol in

E. coli

was achieved when the binding sites for Zif268 and PBSII were separated with 8 bp [11].

Figure 12.7 Applications of DNA‐guided programming. (a) For many biosynthetic pathways, the first enzymes in the cascade are the same, and the end product is determined by the enzymes that are lower in the cascade. By immobilization of specific enzymes on a DNA scaffold, when others are left out, we can determine which end product will be preferentially synthesized. This is a powerful tool for influencing the biosynthetic flux to produce less unwanted products and a cleaner end product. (b) Similar to protein scaffolds, DNA scaffolds can be used for defining the order in which protein kinases phosphorylate each other or a chain of other posttranslational protein modifications. Signaling pathways can, therefore, be modulated using different scaffolds. The DNA scaffold could also be used for information processing such as rewiring intracellular signaling pathways and designing new protein networks for constructing new biological devices with selected features.

Chapter 13: Synthetic RNA Scaffolds for Spatial Engineering in Cells

Figure 13.1 Prevalence and diversity of secondary structure in natural RNA. (a) The alanine‐carrying transfer RNA shown here has the typical clover leaf structure common among tRNA. (b) The theophylline‐binding riboswitch (from PDB: 1O15_A) is a canonical riboswitch. (c) The PP7 aptamer [16] binds to the PP7 coat protein with low nanomolar affinity. (d) The

Homo sapiens

TERC lncRNA (NR_001566.1) is an example of a natural lncRNA that serves as a scaffold.

Figure 13.2 Design principles for RNA structure and function. (a) RNA secondary structure can be predicted from the primary sequence using a variety of software packages. (b) RNA can self‐assemble into 2D or 3D structures

in vitro

. (c) Researchers have developed a variety of synthetic parts, such as synthetic riboregulators, synthetic ribozymes, ligand‐regulated riboregulators, and ligand‐regulated ribozymes [51–54]. (d)

In vitro

selection can be used to enhance the function of RNAs through iterative rounds of amplification and selection.

Figure 13.3 Applications of RNA scaffolds

in vivo

. (a) mRNA are modified to include either several repeats of an aptamer or two different aptamers in close proximity. The former approach results in concentrated foci of fluorescent protein fusions to RNA‐binding domains (RBDs) [78] and in the latter, two halves of the protein with RBD fusions [79], only complement to be fluorescent on the mRNA scaffold. (b) Enzymes fused to RBDs localize to self‐assembled RNA scaffolds with aptamers presented. Channeling of intermediate metabolites can lead to enhanced pathway flux toward biofuels or other high value products [1]. (c) Pentamer of bacteriophage Φ29 pRNA [23] from PDB file 1FOQ. Tagging the monomers with functional units like siRNA can make them useful drug delivery vehicles [6, 80]. (d) The clover leaf tRNA sequence can be tagged with recombinant RNA and epitopes as shown to allow for its synthesis and purification [81].

Chapter 14: Sequestered: Design and Construction of Synthetic Organelles

Figure 14.1 Possible strategies for engineering a synthetic organelle. Complexity of intra‐ and intercellular spatial organization spans from enzymes with inherent substrate channeling to symbiotic cocultures. This review highlights work in the middle ground, from nanocompartments to repurposed organelles.

Figure 14.2 Four core design principles for a synthetic organelle.

Figure 14.3 The structure and function of a bacterial microcompartment, the carboxysome. (a) Structural model showing some of the structural components and enzymatic cargo. (b) Schematic of the carbon‐concentrating mechanism of carboxysome function. (c) Model of shell permeability based on CcmK4 (PDB: 2A10).

Figure 14.4 Genetic organization of α‐ and β‐carboxysome regulons from two different bacteria. Color scheme indicates relatedness.

Figure 14.5 Model of an encapsulin. (a) Genomic organization in

T. maritima

highlight signal sequence of cargo protein. (b) Structural model based on encapsulin X‐ray structure (PDB: 3DKT) and ferritin‐like protein (PDB: 3HL1).

Figure 14.6 Schematic of potential synthetic organelles in the budding yeast.

Chapter 15: Cell‐Free Protein Synthesis: An Emerging Technology for Understanding, Harnessing, and Expanding the Capabilities of Biological Systems

Figure 15.1 Advantages for cell‐free biology. By bypassing cellular objectives and opening the reaction environment, cell‐free protein synthesis allows for increased freedom of design as a result of the benefits highlighted here.

Figure 15.2 A new paradigm for cell‐free biomanufacturing. Cell‐free protein synthesis is able to separate catalyst synthesis (cell growth) from catalyst utilization (protein synthesis). This allows resources to be funneled toward the product of interest in ways not possible

in vivo

.

Figure 15.3 Historical trends for different CFPS systems. Batch protein yields for the papers cited in this review are arranged by platform (a) and product type (b). In addition, cell‐free protein synthesis has seen successes at a variety of volumes (c). ECE,

E. coli

extract; WGE, wheat germ extract; SCE,

S. cerevisiae

extract; ICE, insect cell extract; CHO, Chinese hamster ovary cell extract; LTE,

L. tarentolae

extract; STR,

Streptomyces

extract; BSE,

B. subtilis

extract; PDMS, polydimethylsiloxane; GFP, green fluorescent protein; iPSCs, induced pluripotent stem cells; rhGM‐CSF, recombinant human granulocyte macrophage colony‐stimulating factor.

Figure 15.4 The production of proteins containing ncAAs is a frontier of CFPS. Amber suppression, shown here, is the most common method for ncAA incorporation in CFPS platforms but is hampered by competition between the amber‐suppressing tRNA and release factor 1 (RF1). Several methods have been developed to prevent this competition. Also, new strains lacking RF1 should address this issue.

Figure 15.5 CFPS is a useful approach for the production of membrane proteins. Several methods have been implemented to mimic the cell membrane in cell‐free protein synthesis: (a) lipid bilayer, (b) liposome, (c) micelle, (d) bicelle, (e) nanodisc, and (f) tethered bilayer lipid membrane.

Chapter 16: Applying Advanced DNA Assembly Methods to Generate Pathway Libraries

Figure 16.1 Overview of the combinatorial library approach for pathway improvement. When improving a multi‐gene pathway, variations of the pathway components including promoters, RBSs, coding DNA sequences (CDSs), or transgenic regions are generated by either mutagenesis, homolog cloning, or

in silico

design (promoters and CDSs are used as examples in the figure). The diversified components are then assembled by various DNA assembly techniques to form a library of combinations. Cells hosting this pathway library will then be screened for the optima of the desired phenotype. Labels “p1‐3” standard for promoters. Labels “t1‐3” standard for terminators.

Figure 16.2 Preparation of DNA fragments for large library generation. Each unique design represents a unique promoter or gene. Varied strength promoters, orthologous genes, or mutated pathway components generate diversity. If the DNA is assembled with homology regions, upstream and downstream of the DNA fragment of interest, the pathway can assemble properly into many different combinations. Each strategy has incorporated different lengths of homology, which can contribute to the efficiency of correct assembly. These DNA fragments are then subjected to the desired DNA assembly reaction with the linearized vector and transformed into the host.

Figure 16.3 Fermentation profiles of the evolutionary rounds for the pathway libraries. (a,b) Cellobiose consumption and ethanol production of the cellobiose utilization pathway from the promoter‐based directed evolution. The black square represents the parent pathway with no mutations in the

PDC1

and

ENO2

promoter. The circles are the first round of error‐prone PCR of both promoters. The triangles represent the second and final rounds of directed evolution mutagenesis. (c,d) Cellobiose consumption and ethanol production of the cellobiose utilization pathway from the protein‐based directed evolution. The black square represents the wild‐type pathway with no mutations in the β‐glucosidase and the cellodextrin transporter. The circle is the first round of error‐prone PCR of both proteins. The triangle represents the second round of directed evolution.

Chapter 17: Synthetic Biology in Immunotherapy

Figure 17.1 Schematic of adoptive T‐cell therapy. Endogenous tumor‐infiltrating lymphocytes (TILs) are T cells with natural tumor reactivity and can be isolated from tumor biopsies, expanded

ex vivo

, and reinfused into the cancer patient. Alternatively, non‐tumor‐reactive T cells can be isolated, genetically modified to express a tumor‐reactive T‐cell receptor (TCR) or chimeric antigen receptor (CAR), expanded

ex vivo

, and reinfused into the patient.

Figure 17.2 CARs redirect T‐cell specificity toward tumor targets. (a) Schematic of first‐, second‐, and third‐generation CARs. The single‐chain variable fragment (scFv) derived from a tumor‐antigen‐specific antibody serves as the extracellular sensing domain, and the cytoplasmic tail of the CD3ζ chain serves as the intracellular signaling domain of the CAR. In second‐ and third‐generation CARs, one or two costimulatory domains such as CD28 and 4‐1BB are directly fused to the CD3ζ chain to enhance T‐cell signaling. (b) Schematic of single‐chain, bispecific OR‐gate CARs. T cells expressing an OR‐gate signal processing system can kill any target cell that expresses either antigen A or antigen B. (c) Schematic of an AND‐NOT‐gate CAR pair. The first receptor is a conventional CAR that targets antigen A. The second is a chimeric inhibitory receptor (iCAR) that targets antigen B and contains the cytoplasmic domain of an inhibitory receptor (e.g., PD‐1 or CTLA‐4). Presence of antigen A triggers CAR signaling, while presence of antigen B triggers iCAR signaling. The inhibitory function of the iCAR overrides any activation signal that may result from the conventional CAR, thus executing A‐NOT‐B signal computation. (d) Schematic of an AND‐gate CAR pair. The first receptor is a conventional first‐generation CAR that targets antigen A and contains only the CD3ζ chain without costimulatory signals. The second is a chimeric costimulatory receptor that targets antigen B and contains both CD28 and 4‐1BB costimulatory signals but no CD3ζ chain. Both antigens must be present to trigger a sufficiently robust T‐cell response to execute therapeutic function. (e) Schematic of a “remote‐controlled” CAR system. Here, the CAR protein is split into two parts, with the first fragment being a conventional CAR that contains the FK506 binding protein (FKBP) instead of the CD3ζ chain at the C‐terminus. The second fragment consists of a membrane‐tethered CD3ζ chain fused to the FKBP‐rapamycin binding (FRB). Presence of a rapamycin analog (rapalog) molecule triggers dimerization between FKBP and FRB, thereby reconstituting a full CAR protein and enabling CAR signaling in response to antigen binding. (f) Schematic of a synthetic Notch (synNotch) receptor‐regulated CAR expression system. Upon binding to antigen A, the synNotch receptor releases a TF, which translocates to the nucleus and triggers CAR expression from a cognate promoter. This CAR molecule is subsequently able to trigger T‐cell activation upon binding to antigen B, resulting in AND‐gate signal computation in a sequential manner.

Figure 17.3 Synthetic biological constructs and circuits enable controlled enhancement of T‐cell function. (a) “Armored” T cells are engineered to overexpress costimulatory ligands, cytokine receptors, chemokine receptors, or immunostimulatory cytokines that can boost T‐cell proliferation, persistence, and effector functions in an autocrine manner. Once secreted, immunostimulatory cytokines (e.g., IL‐2, IL‐12, IL‐15, etc.) can also signal in paracrine fashion to trigger the recruitment, growth, and antitumor responses of native immune cells. (b) T cells can be genetically modified to resist inhibitory signals present on tumor cells (e.g., PD‐L1) or within the tumor microenvironment (e.g., TGF‐β) by expressing dominant‐negative receptors (DNRs) or knocking out inhibitory receptors. DNRs lack signal transduction domains and competitively sequester immunosuppressive ligands away from native inhibitory receptors. Genetic knockout of inhibitory receptor expression abrogates receptor‐mediated recognition of immunosuppressive factors, thus reducing T‐cell dysfunction and exhaustion. (c) Inverted cytokine receptors (ICRs) are fusions between the extracellular ligand binding of an inhibitory receptor and the intracellular signaling domain of an immunostimulatory receptor. Encounter with immunosuppressive cytokines (e.g., IL‐4) in the tumor microenvironment activates expression programs that enhance T‐cell proliferation, persistence, and effector functions.

Figure 17.4 A chemically inducible caspase 9 kill switch. Inactive pro‐caspase 9 monomers are linked to the human FK506 binding protein FKBP and constitutively expressed in the engineered cell. Upon addition of the chemical inducer of dimerization AP1903, the FKBP domains dimerize and lead to the cross‐linking and activation of caspase 9, which triggers downstream events in the apoptosis pathway and results in cell death.

Chapter 18: Synthetic Biology: From Genetic Engineering 2.0 to Responsible Research and Innovation

Figure 18.1 Focus group’s evaluation of SB, before and after they receive information. The x‐axis means: −100 totally opposed; 0 neutral; +100 totally endorsing.

Figure 18.2 The dominant comparator for SB could come from either of three preexisting technology debates.

Figure 18.3 The SYNMOD game app allows players to create and combine different peptide modules to design new antibiotics. The game is freely available for iOS X and Android devices. See http://www.biofaction.com/project/synmod‐mobile‐game/

Guide

Cover

Table of Contents

Begin Reading

Pages

xv

xvi

1

3

4

5

6

7

8

9

10

11

12

13

15

16

17

18

19

20

22

23

24

25

26

27

28

29

30

31

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

49

50

51

52

53

54

55

56

57

58

59

60

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

150

151

152

153

154

156

157

158

159

161

162

163

164

165

166

167

168

169

170

171

173

174

175

177

178

179

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

230

231

232

233

234

235

237

239

240

241

242

243

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

337

338

339

340

341

343

344

345

346

347

349

350

351

352

353

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

Related Titles

Luisi, P.P. (ed.)

Chemical Synthetic Biology

2011

Print ISBN: 978-0-470-71397-6

Further Volumes of the “Advanved Biotechnology“ Series:

Published:

Villadsen, J. (ed.)

Fundamental Bioengineering

2016

Print ISBN: 978-3-527-33674-6

Love, J. Ch. (ed.)

Micro- and Nanosystems for Biotechnology

2016

Print ISBN: 978-3-527-33281-6

Wittmann, Ch., Liao, J.C. (eds.)

Industrial Biotechnology

Microorganisms (2 Volumes)

2017

Print ISBN: 978-3-527-34179-5

Wittmann, Ch., Liao, J.C. (eds.)

Industrial Biotechnology

Products and Processes

2017

Print ISBN: 978-3-527-34181-8

Yoshida, T. (ed.)

Applied Bioengineering

2017

Print ISBN: 978-3-527-34075-0

Nielsen, J., Hohmann, S. (eds)

Systems Biology

2017

Print ISBN: 978-3-527-33558-9

Coming soon:

Chang, H. N.

Emerging Areas in Bioengineering

2018

Print ISBN: 978-3-527-34088-0

Planned:

Lee, G. M., Faustrup Kildegaard, H.

Cell Culture Engineering

Recombinant Protein Production

2018

Print ISBN: 9783527343348

series

Synthetic Biology: Parts, Devices and Applications

Edited by Christina Smolke

Volume Editor

Christina Smolke

Stanford School of Medicine

Y2E2 Building, Room #B07

473 Via Ortega

Stanford, CA 94305

United States

Series Editors

Sang Yup Lee

KAIST

Department of Chemical ' Biomolecular

Engineering

291 Daehak-ro, Yuseong-gu

34141 Daejeon

Republic of Korea

Jens Nielsen

Chalmers University of Technology

Department of Biology and Biological Engineering

Kemivägen 10

41296 Göteborg

Sweden

Gregory Stephanopoulos

Massachusetts Institute of Technology

Department of Chemical Engineering

77 Massachusetts Avenue

Cambridge, MA 02139

USA

Cover

fotolia_yurakp and fotolia_EvgeniyBobrov

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de.

© 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Print ISBN: 978-3-527-33075-1

ePDF ISBN: 978-3-527-68808-1

ePub ISBN: 978-3-527-68809-8

Mobi ISBN: 978-3-527-68807-4

oBook ISBN: 978-3-527-68810-4

Cover Design Adam-Design, Weinheim, Germany

About the Series Editors

Sang Yup Lee is distinguished Professor at the Department of Chemical and Biomolecular Engineering at the Korea Advanced Institute of Science and Technology. At present, Prof. Lee is the Director of the Center for Systems and Synthetic Biotechnology, Director of the BioProcess Engineering Research Center, and Director of the Bioinformatics Research Center. He has published more than 500 journal papers, 64 books, and book chapters, and has more than 580 patents (either registered or applied) to his credit. He has received numerous awards, including the National Order of Merit, the Merck Metabolic Engineering Award, the ACS Marvin Johnson Award, Charles Thom Award, Amgen Biochemical Engineering Award, Elmer Gaden Award, POSCO TJ Park Prize, and HoAm Prize. He is Fellow of American Association for the Advancement of Science, the American Academy of Microbiology, American Institute of Chemical Engineers, Society for Industrial Microbiology and Biotechnology, American Institute of Medical and Biological Engineering, the World Academy of Science, the Korean Academy of Science and Technology, and the National Academy of Engineering of Korea. He is also Foreign Member of National Academy of Engineering, USA. In addition, he is honorary professor of the University of Queensland (Australia), honorary professor of the Chinese Academy of Sciences, honorary professor of Wuhan University (China), honorary professor of Hubei University of Technology (China), honorary professor of Beijing University of Chemical Technology (China), and advisory professor of the Shanghai Jiaotong University (China). Apart from his academic associations, Prof. Lee is the editor‐in‐chief of the Biotechnology Journal and is also contributing to numerous other journals as associate editor and board member. Prof. Lee is serving as a member of Presidential Advisory Committee on Science and Technology (South Korea).

Jens Nielsen is Professor and Director to Chalmers University of Technology (Sweden) since 2008. He obtained an MSc degree in chemical engineering and a PhD degree (1989) in biochemical engineering from the Technical University of Denmark (DTU) and after that established his independent research group and was appointed full professor there in 1998. He was Fulbright visiting professor at MIT in 1995–1996. At DTU, he founded and directed the Center for Microbial Biotechnology. Prof. Nielsen has published more than 350 research papers and coauthored more than 40 books, and he is inventor of more than 50 patents. He has founded several companies that have raised more than 20 million in venture capital. He has received numerous Danish and international awards and is member of the Academy of Technical Sciences (Denmark), the National Academy of Engineering (USA), the Royal Danish Academy of Science and Letters, the American Institute for Medical and Biological Engineering and the Royal Swedish Academy of Engineering Sciences.

Gregory Stephanopoulos is the W.H. Dow Professor of Chemical Engineering at the Massachusetts Institute of Technology (MIT, USA) and Director of the MIT Metabolic Engineering Laboratory. He is also Instructor of Bioengineering at Harvard Medical School (since 1997). He received his BS degree from the National Technical University of Athens and PhD from the University of Minnesota (USA). He has coauthored about 400 research papers and 50 patents, along with the first textbook on metabolic engineering. He has been recognized by numerous awards from the American Institute of Chemical Engineers (AIChE) (Wilhelm, Walker and Founders awards), American Chemical Society (ACS), Society of Industrial Microbiology (SIM), BIO (Washington Carver Award), the John Fritz Medal of the American Association of Engineering Societies, and others. In 2003, he was elected member of the National Academy of Engineering (USA) and in 2014 President of AIChE.

Part IDNA Synthesis and Genome Engineering

Chapter 1Competition and the Future of Reading and Writing DNA

Robert Carlson

Biodesic and Bioeconomy Capital, 3417 Evanston Ave N, Ste 329, Seattle, WA, 98103, USA

Constructing arbitrary genetic instruction sets is a core technology for biological engineering. Biologists and engineers are pursuing even better methods to assemble these arbitrary sequences from synthetic oligonucleotides (oligos) [1]. These new assembly methods in principle reduce costs, improve access, and result in long sequences of error‐free DNA that can be used to construct entire microbial genomes [2]. However, an increasing diversity of assembly methods is not matched by any obvious corresponding innovation in producing oligos. Commercial oligo production employs a very narrow technology base that is many decades old. Consequently, there is only minimal price and product differentiation among corporations that produce oligos. Prices have stagnated, which in turn limits the economic potential of new assembly methods that rely on oligos. Improvements may come via recently demonstrated assembly methods that are capable of using oligos of lower quality and lower cost as feedstocks. However, while these new methods may substantially lower the cost of gene‐length double‐stranded DNA (dsDNA), they also may be economically viable only when producing many orders of magnitude with more dsDNA than what is now used by the market. The commercial success of these methods, and the broader access to dsDNA they enable, may therefore depend on structural changes in the market that are yet to emerge.

1.1 Productivity Improvements in Biological Technologies

In considering the larger impact of technological monoculture in DNA synthesis, it is useful to contrast DNA synthesis and assembly with DNA sequencing. In particular, it is instructive to compare productivity estimates of commercially available sequencing and synthesis instruments (Figure 1.1). Reading DNA is as crucial as writing DNA to the future of biological engineering. Due to not just commercial competition but also competition between sequencing technologies, both prices and instrument capabilities are improving rapidly. The technological diversity responsible for these improvements poses challenges in making quantitative comparisons. As in previous discussions of these trends, in what follows I rely on the metrics of price [$/base] and productivity [bases/person/day].

Figure 1.1 Estimates of the maximum productivity of DNA synthesis and sequencing enabled by commercially available instruments. Productivity of DNA synthesis is shown only for column‐based synthesis instruments, as data for sDNA fabricated on commercially available DNA arrays is unavailable; exceptions are discussed in the text. Shown for comparison is Moore’s law, the number of transistors per chip.

(Intel; Carlson, 2010 [3]; Loman et al. 2012 [4]; Quail et al. 2012 [5]; Liu, 2012 [6].)

Figure 1.1 also directly compares the productivity enabled by commercially available sequencing and synthesis instruments to Moore’s law, which describes the exponential increase in transistor counts in CPUs over time. Readers new to this discussion are referred to References 3 and 4 for in‐depth descriptions of the development of these metrics and the utility of a comparison with Moore’s law [3, 7]. Very briefly, Moore’s law is a proxy for productivity; more transistors enable greater computational capability, which putatively equates to greater productivity.

Visual inspection of Figure 1.1 reveals several interesting features. First, general synthesis productivity has not improved for several years because no new instruments have been released publicly since about 2008. Productivity estimates for instruments developed and run by oligo and gene synthesis service providers are not publicly available.1

Second, it is clear that DNA sequencing platforms are improving very rapidly, now much faster than Moore’s law.

Moore’s law and its economic and social consequences are often used to benchmark our expectations of other technologies. Therefore, developing an understanding of this “law” provides a means to compare and contrast it with other technological trends.

1.2 The Origin of Moore’s Law and Its Implications for Biological Technologies

Moore’s law is often mistakenly described as a technological inevitability or is assumed to be some sort of physical phenomenon. It is neither; Moore’s law is a business plan, and as such it is based on economics and planning. Gordon Moore’s somewhat opaque original statement of what became the “law” was a prediction concerning economically viable transistor yields [8]. Over time, Moore’s economic observation became an operational model based on monopoly pricing, and it eventually enabled Intel to outcompete all other manufacturers of general CPUs. Two important features distinguish CPUs from other technologies and provide insight into the future of trends in biological technologies: the first is the cost of production, and the second is the monopoly pricing structure.

Early on Intel recognized the utility of exploiting Moore’s law as a business plan. A simple scaling argument reveals the details of the plan. While transistor counts increased exponentially, Intel correspondingly reduced the price per transistor at a similar rate. In order to maintain revenues, the company needed to ship proportionally more transistors every quarter; in fact, the company increased its shipping numbers faster than prices fell, enabling consistent revenue to grow for several decades. This explains why Intel former CEO Andy Grove reportedly constantly pushed for an even greater scale [9].

In this sense, Moore’s law was always about economics and planning in a multibillion‐dollar industry. In the year 2000, a new chip fab cost about $1 billion; in 2009, it cost about $3 billion. Now, according to The Economist, Intel estimates that a new chip fab costs about $10 billion [9]. This apparent exponential increase in the cost of semiconductor processing is known as Rock’s law. It is often argued that Moore’s law will eventually expire due to the physical constraints of fabricating transistors at small length scales, but it is more likely to become difficult to economically justify constructing fabrication facilities at the cost of tens to hundreds of billions of dollars. Even through the next several iterations, these construction costs will dictate careful planning that spans many years. No business spends $10 billion without a great deal of planning, and, more directly, no business finances a manufacturing plant that expensive without demonstrating a long‐term plan to repay the financiers. Moreover, Intel must coordinate the manufacturing and delivery of very expensive, very complex semiconductor processing instruments made by other companies. Thus Intel’s planning and finance cycles explicitly extend many years into the future. New technology has certainly been required to achieve each planning goal, but this is part of the ongoing research, development, and planning process for Intel.

Moore’s law served a second purpose for Intel and one that is less well recognized but arguably more important; it was a pace selected to enable Intel to win. Intel successfully organized an entire industry to move at a pace only it could survive. And only Intel did survive. While Intel still has competitors in products such as memory or GPUs, companies that produced high volume, general CPUs have all succumbed to the pace of Moore’s law. The final component of this argument is that, according to Gordon Moore, Intel could have increased transistor counts faster than the historical rate.2 In fact, Intel ran on a faster internal innovation clock than it admitted publicly, which means that Moore’s law was, as one Intel executive put it, a “marketing head fake” [10]. The inescapable conclusion of this argument is that the management of Intel made a very careful calculation; they evaluated product rollouts to consumers – the rate of new product adoption, the rate of semiconductor processing improvements, and the financial requirements for building the next chip fab line – and then set a pace that nobody else could match but that left Intel plenty of headroom for future products. In effect, if not intent, Intel executed a strategy that enabled it to set CPU prices and then to reduce those prices at a rate no other company could match.

This long‐term planning, pricing structure, and the resulting lack of competition contrasts quite strongly with the commercial landscape for biological technologies. Whereas the exponential pace of doubling of transistor counts was controlled by just one company, productivity in DNA sequencing has recently improved faster than Moore’s law due to competition not just among companies but also among technologies. Conversely, the lack of improvement in synthesis productivity suggests that the narrow technology base for writing DNA has reached technical and, therefore, economic limits. Nonetheless, while Figure 1.1