Orthogonal replication, or replication totally separate from the host’s replication, is quite difficult to achieve. In yeast, this can be accomplished with a cytoplasmic plasmid (1) or in the mitochondria. This is achievable because of a membrane inbetween the host’s replication compartment and the orthgonal one. However, in E coli, all cellular components other than the periplasm are in a single compartment. Orthogonality, therefore, is much more difficult to achieve.
Orthogonal replication. Achieves minimal crossover in replication. (9)
Labs have attempted to do this, but none have succeeded in raising in vivo orthogonal mutation rate without raising whole-cell mutation rate (2). Groups have even attempted making an orthogonal replication system using the T7 DNAp (3), but haven’t succeeded and therefore are not sure about actual orthogonality. PACE has circumevented (4) this by phage infection followed by continuous passaging. However, this can only be used when there is a connection between selection and pIII expression, which can oftentimes be difficult to optimize and completely unsuitable in many cases.
It would be desirable to have a system that is completely orthogonal from the genome in E coli for continuous directed evolution. Hypothetically, there are 2 possibilities for this system: either have a different replication system, such as a linear protein-primed replicative system or have a different nucleotide, such as RNA. The former can be accomplished with a linear replication, much like the one in yeast, and the latter can be accomplished using in vivo Qbeta replication.
In order to get an in vivo linear protein-primed replication system in E coli, phage PRD1’s replication system must be hijacked. PRD1 is a linear protein-primed phage that replicates in E coli and a relative to the better studied phage phi29 (B subtilis), as well as the linear plasmid pBClin15 (B cereus). However, PRD1 and phi29 are both localized to the nucleoid, (5), and since phi29 polymerase can replicate ssDNA in vitro, (6) it can be inferred that there is a high probability of it doing the same in the cell with the lagging strand fragments. This would make the system not fully orthogonal from the genome. The phage polymerase could be evolved to not accept ssDNA fragments, but there is also a possibility that it cannot be. The DNA polymerase would also need to be mutated to increase its mutation rate, a non-trivial task.
In order to get an in vivo RNA replication system in E coli, phage qbeta’s replication system must be hijacked. Qbeta is an ssRNA+ phage that replicates in E coli and has been extensively studied in vitro (7) and in vivo. In 1965, Spiegelman published a paper on this RNA phage replicating in vitro, mutating to become more efficient in Spiegelman’s specific environments, earning it the name of “Spiegelman’s Monster”. If that process of replication can be harness in vivo, it could easily be a completely and inherently orthogonal system from the genome. On orthogonality, Spiegelman noted-
“Consider an RNA virus approaching a cell some 10^6 times its size and into which the virus is going to inject its only strand of genetic information. Even if the protein-coated ribosomal RNA molecules are ignored, the cell cytoplasm still contains of the order of 10,000 free RNA molecules of various sorts. The viral RNA contains the information required for the synthesis of the new kind of polymerase designed to make RNA copies from RNA. If this replicase were indifferent and accepted any RNA it happened to meet, what chance would the single original strand injected have of ever being replicated?” (8)
There are still issues, however. Many copies of the qbeta derived RNA means many point mutation copies will only be weakly expressed. Furthermore, there is strong selective pressure to not only evolve for the desired trait, but also to reduce its RNA genome in size and to avoid RNAse degredation. Still, qbeta has one of the highest mutation rates of any virus with the bonus of inherent orthogonality, making it a good candidate for this system.
Overall, it will be difficult to create an orthogonal replication system for directed evolution in E coli, but the finished system would allow for directing evolution on a massive scale.
(Yes, these are not a full references, just put important information)
- “An orthogonal DNA replication system in yeast”, Arjun Ravikumar, http://www.nature.com/nchembio/journal/v10/n3/full/nchembio.1439.html#ref3
- “Targeted gene evolution in Escherichia coli using a highly error-prone DNA polymerase I”, Manel Camps, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC187833/
- “Building an Orthogonal Replication Systems for Performing Directed Evolution in Escherichia coli: A Strategic Review and a Summary of the Initial Steps in Cloning Bacteriophage T7 gp4 Primase/Helicase”, Wendy Ma, http://jemi.microbiology.ubc.ca/sites/default/files/Ma%20et%20al..pdf
- “A system for the continuous directed evolution of biomolecules”, Kevin M. Esvelt, http://www.nature.com/nature/journal/v472/n7344/abs/nature09929.html
- “Nuclear and nucleoid localization are independently conserved functions in bacteriophage terminal proteins”, Modesto Redrejo-Rodríguez, http://onlinelibrary.wiley.com/doi/10.1111/mmi.12404/full
- “Rapid Amplification of Plasmid and Phage DNA Using Phi29 DNA Polymerase and Multiply-Primed Rolling Circle Amplification”, Frank B. Dean, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC311129/
- “The synthesis of a self-propagating and infectious nucleic acid with a purified enzyme.”, S Spiegelman, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC219765/
- “The Sol Spiegelman Papers RNA Replicase and Spiegelman’s “Little Monster,” 1961-1969″, https://profiles.nlm.nih.gov/ps/retrieve/Narrative/PX/p-nid/197