Device and system design to modulate evolutionary stability

From OpenWetWare

Revision as of 12:33, 26 August 2005 by Jason R. Kelly (Talk | contribs)
(diff) ←Older revision | Current revision (diff) | Newer revision→ (diff)
Jump to: navigation, search

back to Evolutionary Stability

Performance stability can be modulated by the design of the device or system itself. Additionally, better understanding of the system design principles for modulating evolutionary stability may shed some light on the mechanisms natural systems utilize for stabilizing genetic elements. Some ideas for the system design strategies to modulate performance stability are provided below.

Contents

Redundancy

Design a system, such as a bi-stable switch, containing redundant copies of particular device components. The performance stability of the system will presumably increase because all the copies would need to break before the system loses function. However, it is possible that the second copy of the gene will result in increased recombination in the device, requiring codon optimization of the different copies to avoid homologous DNA sequences.

Simple selection circuits

Incorporating selective pressure for device components could enable long-term performance stability. For instance, the performance of a ring oscillator system could be maintained by adding a second device that selects for functionality of the repressor proteins in the system. This device will contain repressor-regulated promoters upstream of genes encoding proteins whose presence is deleterious to the cell, providing a selective pressure to maintain repressor function. The design challenge would be to design a selection device such that it does not interfere with the main system performance. In this case, the promoters regulating deleterious gene expression would need to be repressed by low levels of each repressor, and the ring oscillator would have to function reliably with some low level of each repressor protein present independent of state. Otherwise, the selection would be applied each time the oscillator changed state. This approach is not as general as codon optimization; however the inclusion of selection may enable system performance to be extended indefinitely, so long as the selective pressure remains in place.

Co-evolve an engineered device and a host cell

Attempt to evolve a host cell to gain a competitive advantage from maintaining the function of a simple engineered device by propagating the device over a large number of generations with selective pressure for appropriate device performance. Previous work with a tetracycline resistance plasmid by Lenski et al suggests this approach may be feasible (Lenski et al., 1994). After five hundred generations in media containing tetracycline the host cell evolved a competitive advantage from the tetracycline resistance plasmid independent of the antibiotic being present. This was shown by competition experiments between the plasmid-containing cell and an isogenic plasmid-free counterpart in media without tetracycline. The success of this approach likely varies dramatically depending on the particular device.

As an example, we can attempt to evolve a host cell to stabilize an inverter that regulates expression of a selectable marker that also has a counter-selection. The inverter will be induced to switch between ON and OFF states and subjected to selection for expression or non-expression of the marker. This selection for function will be repeated periodically while the cells are grown over a large number of generations. Finally, the growth of cells containing the device will be compared to isogenic device-free counterparts in non-selective conditions to evaluate if the host has evolved to gain a competitive advantage from maintaining the device. In support of providing more sophisticated selections to devices in continuous culture, we've designed a microfluidic chemostat integrated with a cell sorter (i.e. a SortoStat).

Mechanisms suggested by natural systems

Previous work suggests some natural mechanisms for increasing the evolutionary stability of biological systems. For example, the demand theory of gene expression developed by Savagneau explores why a single gene might be regulated by a transcriptional activator or repressor (Savageau, 1983). Demand theory outlines the system architecture that best leverages selective pressure on protein expression towards maintenance of the control system for protein expression. Although demand theory has been supported by examining native regulatory systems, we can provide a more direct proof by demonstrating demand theory via experimental evolution of a simple engineered system.

Additionally, a cue can be taken from the many viral genomes that utilize multiple reading frames for overlapping genetic elements in the same sequence. This mechanism is thought to serve as a means of information compression due to selection for small genome size in viral replication. It may also play a role in stabilizing non-essential genetic elements via ‘interlacing’ with essential elements. For instance, a non-essential gene that shared sequence with an essential gene may gain resistance to frameshift mutations since such a mutation would typically result in a loss of function of the essential gene, selecting against the mutant. This approach will be used to stabilize engineered systems by interlacing system components with essential genes or selectable markers.

Contact

Jason Kelly

Personal tools