Building a Genome from Scratch: Lessons Learned from Sc2.0

If anyone understands the challenges associated with powering a eukaryotic cell with a designer genome built from scratch, it’s Leslie Mitchell, a postdoc in Jef Boeke’s lab at NYU. Mitchell has been leading experimental design and technological development for the Synthetic Yeast Genome Project (Sc2.0) since 2012. In this role, she collaborates with Sc2.0 International Consortium team members spanning 4 continents to provide remote mentorship and solve challenges associated with synthetic chromosome design features and assembly.

The goal of Sc2.0 is to design, build, assemble, and test the function of an entirely synthetic designer yeast genome. Earlier this month, the Sc2.0 International Consortium announced the completed design of the Sc2.0 genome, construction of five new synthetic chromosomes, and described the 3D organization of synthetic chromosomes in the nucleus, all of which earned them the March 10 cover story for Science. To date, 30% of the Sc2.0 genome has been constructed in cells, and the team hopes to have the entire genome – all 16 chromosomes – completed before the end of 2017.

According to Mitchell, the genome engineering effort of Sc2.0 can offer some powerful lessons for GP-write, both from an organizational and technical perspective.

Organizational Insights

A well defined plan should be in place at the launch of the project, which all participants must agree upon, including funding, space, personnel, QA/standards, material transfers, publication policy, intellectual property, software, ownership of the project, training and education, and compliance with local laws. Additionally, because GP-write will involve global participation similar to Sc2.0, it will be important to maintain a collaborative and inclusive culture across international borders.

One organizational feature that has driven success of Sc2.0 is the distributive nature of the project. While Sc2.0 chromosome design is centralized, involving a close collaboration between yeast geneticists of Jef Boeke’s lab and computational biologists from Joel Bader’s lab at Johns Hopkins University, synthesis and assembly are parallelized between teams around the world. Not only does this partition the workload associated with chromosome assembly, it also distributes the financial burden as each team is responsible for obtaining their own funding.

Technical Insights

Following a common design standard has been a key strategy for Sc2.0. All design features that are written into Sc2.0 chromosomes adhere to a set of overarching design principles, the goals of which are to direct growth to wild-type levels while simultaneously increasing genome flexibility and stability.

Another lesson learned from Sc2.0 is to take on a piece-by-piece assembly strategy. Such a strategy is practical in that it’s much easier to manipulate smaller segments of DNA in vitro, for example 30-60 kb segmented into 10 kb ‘chunks’. More importantly, this strategy allows the team to evaluate synthetic DNA function in cells along the way and quickly back track to any design features that negatively impact cell fitness. For the Sc2.0 project it is now clear that despite densely spaced clusters of edits (mean distance of ~400bp), the overall design is robust as few ‘bugs’, or designer changes that affect cell fitness, have been uncovered.

Sc2.0 has also highlighted limitations of gene synthesis and how these limitations may impact our ability to design and build synthetic genomes. While cost is a much-discussed issue associated with gene synthesis, a less publicized problem is that not all sequences can be easily synthesized de novo. It all comes down to sequence composition. For instance ‘low complexity DNA’ such as homopolymer runs and regions with extreme GC or AT richness can be particularly difficult to synthesize. Of course with enough time, money and effort most DNA sequences can eventually be built; de novo designed mammalian genome-scale synthesis however will require a drop in cost by orders of magnitude together with technology improvements to enable DNA synthesis without restrictions on sequence composition.

DNA delivery is another area ripe for technology development for GP-write. Yeast cells are easily transformed and incorporate DNA into their genome readily using their natural capacity for homologous recombination. Together these features have enabled an efficient Sc2.0 DNA delivery strategy encompassing segments of 30-60kb of designer DNA. The Sc2.0 strategy can applied on some level for GP-write, but there will be delivery challenges that will need to be addressed as we move into mammalian systems, in particular with respect to the delivery of increasingly large segments of designer DNA for targeted delivery.

An additional challenge that GP-write will face with respect to building artificial mammalian chromosomes is centromere engineering. Unlike budding yeast point centromeres, which are ~125 bp and easy to synthesize, mammalian centromeres are composed of megabases of highly repetitive sequences. Mitchell and her colleagues are currently trying to overcome this engineering challenge with the goal of building mammalian artificial chromosomes that will be stable over many generations of cells. Centromere engineering strategies starting from the ‘top-down’, by minimizing native chromosomes, as well as from the ‘bottom-up’, via de novo establishment of centromere function, will both be important to pursue.

With encoded features that enable genetic flexibility and increased genomic stability, Sc2.0 will be a designer genome with new capabilities. As a result, we will soon be able to ask new biological questions – about evolution, minimal genome sequences that support viability under different conditions, and the requirement of specific genomic features such as repeats and introns. Completion of Sc2.0 will represent a major stepping-stone for GP-write, which will build on the knowledge and technological advances of this project.

We invite you to join Mitchell and her colleagues at the next GP-write meeting on May 9-10th, 2017, which will be held at the New York Genome Center. Mitchell will be speaking on the topic of Genome Engineering Foundries at 11:30 am on May 10th. View the agenda.

Registration is limited to 250 in-person attendees on a first-come, first-served basis so don’t forget to register today! We look forward to seeing you there!