Attachment_3_1000_Molecules_DARPA-BAA-13-37_ATCG_Tools.pdf

Approved for Public Release, Distribution Unlimited

Tool and Concepts under development in Living Foundries: ATCG (DARPA-BAA-12-60)

August 9, 2013

Approved for Public Release, Distribution Unlimited

The following slides were developed by current performers within the Living Foundries: ATCG program (DARPA-BAA-12-60).

Information regarding the overall goal of the project, quantitative impact, and current stage of development are included for reference by potential proposers to the Living Foundries: 1000 Molecules program.

Dissemination of this information does not necessarily reflect the position or the policy of the Government, nor a recommendation for collaboration or teaming.

SAGE – Software Actuated Genome Engineering GOAL: Developing and disseminating a software controlled droplet based engineering platform to expedite the design, editing and evolution of biological systems.

The impact will be to qualitatively change the scope of biological manipulations that established and new researchers can contemplate.

We have implemented MAGE (Multiplex Automated Genome Engineering) on Advanced Liquid Logic cartridges modified to perform cell growth, heat shock, chilling, oligonucleotide introduction, electroporation and cell recovery under fully automated software control.

In LF-ATGC, we are miniaturizing the platform to operate on 10K droplets with nL volumes, under complete software control with sophisticated sensing capabilities.

We envision that researchers will “code-up” biological experiments on a Phase II software programmable platform that reduces the barrier to entry for biological manufacturing and design.

A joint project with Advanced Liquid Logic (Michael Pollack), Duke University (Prof. Richard Fair), Harvard University (Prof.

George Church) and Stanford University (Prof. Horowitz)

Teaming Contact: Prof. Mark Horowitz (horowitz@stanford.edu)

Problem Addressed

Act Capabilities Proof it works

What enzymatic pathway would biosynthesize chemical X?

Given target X, Act outputs all possible (ranked) paths, even for unnatural targets, and with predicted enzymes

Engineered Tylenol‐producing E. coli strain in the wet lab

What are all chemicals I could biosynthesize?

Act can create lists of all possible biosynthesizable chemicals

Act predicts 3500+ chemicals as biosynthesizable, e.g., 2‐ carene, L‐DOPS

Future: Can we iteratively improve predictions?

Future: Act will learn from analytics over previously constructed strains

Overall Goal Predict biosynthetic pathways to arbitrary (un)natural targets

Quantitative Impact

Reduce months of literature/DB search to seconds and be integrate‐able into automated build systems

Current Development

Pathway predictor software tool available http://act.berkeley.edu/

Desired chemical e.g., Tylenol

Path search in metabolic space

PathwayRanked by viability

Engineered organism

Contact:

saurabhs@berkeley.edu http://act.berkeley.edu/

(Anderson Lab)L‐>R: native to non‐native chemicals, each line a heterologous enzyme

Challenge: Optical monitoring of fluorescent reporters currently limited to 3 reporters by cross talk

New Technique: Selective excitation of spectrally overlapping fluorescent reporters with shaped laser pulses

Objective: Simultaneous monitoring of at least 8 different fluorescent proteins. Extend to selective excitation of light activated proteins.

EBFP

TagBFP

ECFP

Test system: Excitation and emission spectra of three fluorescent proteins overlap in blue region

Contact:

Professor Herschel Rabitz Princeton University hrabitz@princeton.edu

Optical Monitoring of Engineered Genetic Circuits Princeton University and Lockheed Martin the “lagoon” selection phage:

encodes library memberinfection

SPAPMP

if SP encodes non-functional library member no pIII is produced

SPAPMP

x if SP encodes functional library member pIII is produced

SPAPMP

constant inflow mutagenesis (continuous) infectious progeny

SPAPMP

host cells with accessory plasmid & mutagenesis plasmid

AP

gene III

MP

non-infectious progeny

SPAPMP

x constant outflow

Engineering Biology Through Phage‐Assisted Continuous Directed Evolution (PACE)

• Goals: PACE increasse the speed of laboratory protein evolution by ~100‐fold over current methods, enabling hundreds or thousands of rounds of evolution to take place on the time scale of days to weeks.

Our Living Foundries efforts expand the capabilities of PACE (Aim 1) and apply PACE to the evolution of novel polymer biosyntheses (Aim 2) and expanded genetic codes (Aim 3)

• Progress:

• Significantly streamlined and simplified PACE

• Developed real‐time fitness monitor during PACE

• Linked biopolymer synthesis to gene expression toward PACE of biosynthetic pathways

• Linked tRNA aminoacylation to phage propagation for PACE of expanded genetic codes

• Developed and applied stepwise PACE to dramatically reprogram RNA polymerases that recognize promoters with 20/23 (87%) mutated nucleotides

Esvelt, Carlson, Liu Nature 472, 499 (2011); Dickinson, Leconte, Liu Proc. Natl. Acad. Sci. USA in press (2013); Leconte, Dickinson, Liu Biochemistry 52, 1490 (2013) We welcome collaboration; contact: drliu@fas.harvard.edu; for more information see http://evolve.harvard.edu

Quantitative Impact (in 1st year):

– Prototypes delivered in diverse application areas:

1. Pharmaceuticals: Non-ribosomal peptide synthase (NRPS) gene clusters refactored and 200 optimized.

2. Materials: Partnership with DSM established and 100 six-gene clusters delivered in 12 weeks for a chemical precursor for a super absorbent material.

3. Agriculture: Optimized the nitrogen fixation gene cluster in a form ready for transfer to a new host; ~1000 16-gene clusters built in this process.

4. Genetic programming: 200 new transcription factors (TALE-TFs) built that target unique human genes

– Quantitatively accelerated the design-build-test cycle

1. Faster design: system refactoring accelerated from 7 years to 10 months

2. Faster build: 0.15 clusters/week-person to 50 clusters per week-person

3. Rapid informed construct verification achieved through linkage with deep sequencing

– Core MIT-Broad Foundry established March 2013

1. Co-located with sequencing platform

2. Hiring of director (D. Ben Gordon)

3. Early establishment of assembly platform at Broad (nitrogenase and TALE-TF libraries complete)

4. Integration with transcriptomics / proteomics starting

5. Formal establishment of Industrial consortium nearly complete

Current stage of development (and technology needs):

– Integration of computer aided design with LIMS process integration

– Changing design paradigm with rule-based design and learning (EUGENE, READ, and RAVEN)

– Movement of molecular biology processes into a manufacturing environment

– Staged integration of new technologies (encapsulation, microfluidics, DIAL PCR, “registry in a tube”)

– Need methods and standards for high-throughput data integration across sequencing, transcriptomics, and proteomics platforms

– Need methods for synthesis and assembly of large libraries with high complexity.

– Interested in new high-throughput screening methods.

Goal: Accelerate the prototyping of genetic designs, via

– New technology development (Design-Build-Learn-Test)

– Technology implementation (Manufacturing Pipeline)

– Dissemination and co-development (Industrial Consortium) info@SynBioFoundry.org

Platform technology for production of novel genetically encoded materials that enhance DoD capabilities

Genomically Recoded Organisms (GROs) for virus resistance, genetic isolation & site-specific incorporation of nonstandard amino acids (NSAAs) for new polymers and therapeutics iROTS program aims to engineer the translation system for new classes of biopolymers & orthogonal cellular programs

We will:

• Expand the number of possible co-opted and dedicated codons

• Improve the efficiency of NSAA incorporation 10+ fold

• Develop orthogonal ribosomes that allow for evolution of the large subunit

• Construct at least 3 improved OTS— NSAA pairs

• Increase the efficiency of Multiplex Automated Genome Engineering (MAGE) by ~5x

Contact: Farren Isaacs – farren.isaacs@yale.edu

‐ Time to engineering‐ready microbial genome sequence:

1 month

‐ Quality Control of Assemblies:

100% of assemblies verified, 1X sequence coverage

‐ Assemblies/month: 1000 ‐ Parts per assembly: 2‐12 ‐ Integrated DNA per cycle: 20 kb ‐ Length of Assemblies: 2‐20 kb

(90% efficiency) ‐ Assembly cycle time: 3 days

‐ Integration loci per transformation: 1 (with drug marker)

‐ Host: S. cerevisiae

‐ Time to engineering‐ready microbial genome sequence:

6 hours

‐ Quality Control of Assemblies:

100% of assemblies verified, 15X sequence coverage

‐ Assemblies/month: 4000 ‐ Parts per assembly: 2‐20 ‐ Integrated DNA per cycle: 60 kb ‐ Length of Assemblies: 2‐20 kb

(90% efficiency) ‐ Assembly cycle time: 3 hours

‐ Integration loci per transformation: 3 (NO drug marker)

‐ Host: Many (S. cerevisiae, K.

lactis, K. marxianus, H.

polymorpha, etc)

Sunil Chandran, Dir. ASE; Amyris Inc chandran@amyris.com

Automation of Strain Engineering

SOA 2012

Automation of Strain Engineering

SOA 2014

Living Foundries ATCG enabling technology

Genotype annotation tools: raw genome sequence converted to annotated host genome (Task A)

Quality Control at Scale:

Next Generation Sequencing applied to verify assemblies (Task D)

DNA assembly breakthrough: Ligase Cycling Reaction delivers higher‐ throughput assembly to maintain high volume of DNA editing without sacrificing cycle time, or number of parts assembled.

(Task C)

Multiple Integration technology: more extensive engineering/cycle, shorter time to new pathways & materials (Task E)

Extending engineering to exotic hosts (Task E)

Living Foundries ATCG Projects

Pamela Silver (PI) Jim Collins, Peng Yin (Co‐PIs)

Wyss Institute for Biologically Inspired Engineering, Harvard University

Overall Goal: Compartmentalization and spatial organization to optimize biosynthesis of novel molecules

Nucleic acid nanotechnology for co‐localization and optimization of biosynthesis

• Formed novel DNA‐based structures

• Created RNA‐based scaffolds and structures in bacteria

• Organized multiple enzymes on RNA scaffolds

• Demonstrated ability to increase biosynthesis of alkanes and other molecules

• with scaffold organized reactions Novel circuit design technology

• Designed and characterized over 100 new riboregulators

• Built novel genetic circuits for detecting scaffold architectures and biosynthesis Membrane‐bound vesicles for encapsulation of biosynthetic reactions

• Developed new technology for building and programming vesicles in bacteria

Contact information: pamela_silver@hms.harvard.edu

Synthetic Biology Technologies Available from the Foundation for Applied Molecular Evolution

Contract information Limited to those developed under the following DARPA contract* Title: Reengineering DNA to Enable Ultra Large Scale DNA Assembly Organization: Foundation for Applied Molecular Evolution Principal Investigator: Steven A. Benner Contact: sbenner@ffame.org Contract number: HR0011-12-C-0064.

Overview FfAME has developed a tool for assembling ultra-large DNA constructs (UL-DNA, up to 50,000 base pairs) via the autonomous self-assembly of single stranded fragments. This technology exploits artificially expanded genetic information systems (AEGIS), which add extra nucleotides to the four (GACTZP) already found in natural DNA

Success stories The technology has delivered:

1. A self-assembled gene encoding kanamycin resistance, shown to be functional in E. coli by its conferring kanamycin resistance upon this strain (Merritt et al., manuscript in preparation).

2. Fragments for genes encoding S13 and T4 RNA phages (totaling ~50,000 bps), delivered to collabora-tors at the University of Texas for “booting” of active phages. These will be benchmarked in Phase 2.

3. Strains of E. coli that convert plasmids containing two different six-letter alphabets (GACTZP and

GACTSB) into entirely natural DNA constructs. These will be available to collaborators.

4. A software package that accepts the desired sequence as input, and outputs fragments to be synthesized for autonomous self-assembly using (a) GACTZP alphabets and (b) GACTSB alphabets, in both (c) anneal-extend-ligate formats and (d) anneal-ligate formats. This will be available to collaborators.

5. Remediation steps to allow the CustomArray DNA synthesizer to generate extendable and ligatable DNA fragments. This provides collaborators with a realistic view of the utility of array-based DNA synthesis.

6. Separate 4-, 8-, 12-, 16, 24- and 32-fragment construction experiments with only standard nucleotides, ideally designed to give standard oligonucleotides their best chance of working, to “push to failure” conventional autonomous assembly, demonstrating its limits and the point at which the new technology is needed (Matsuura et al., manuscript in preparation).

7. A pipeline for alternative fragment synthesis with GACTSB and GACTZP alphabets not involving the CustomArray instrument.

8. Quantitation of errors throughout, in both standard and AEGIS nucleotides, assigning tentatively error rates to (a) synthesis, (b) assembly, (c) polymerase extension, (d) conversion, (e) PCR, and (f) sequencing itself. This provides collaborators with realistic statements about the fidelity limits of modern DNA-targeted synthetic biology.

9. Protocols for constructing the assemblies, as well as commercial pipelines that deliver the needed components to collaborators/customers to apply for themselves the new technology.

10. A new implementation of the S:B pair with improved tautomerization properties.

* This is a small fraction of the technologies that FfAME and its affiliate, Firebird Biomolecular Sciences, make available at this time to the synthetic biology community.