Research Article: Genetic interaction mapping with microfluidic-based single cell sequencing

Date Published: February 7, 2017

Publisher: Public Library of Science

Author(s): John R. Haliburton, Wenjun Shao, Adam Deutschbauer, Adam Arkin, Adam R. Abate, Quan Zou.

http://doi.org/10.1371/journal.pone.0171302

Abstract

Genetic interaction mapping is useful for understanding the molecular basis of cellular decision making, but elucidating interactions genome-wide is challenging due to the massive number of gene combinations that must be tested. Here, we demonstrate a simple approach to thoroughly map genetic interactions in bacteria using microfluidic-based single cell sequencing. Using single cell PCR in droplets, we link distinct genetic information into single DNA sequences that can be decoded by next generation sequencing. Our approach is scalable and theoretically enables the pooling of entire interaction libraries to interrogate multiple pairwise genetic interactions in a single culture. The speed, ease, and low-cost of our approach makes genetic interaction mapping viable for routine characterization, allowing the interaction network to be used as a universal read out for a variety of biology experiments, and for the elucidation of interaction networks in non-model organisms.

Partial Text

Cells rely on interactions between biomolecules to achieve complex and dynamic capabilities[1]. For example, cells use genetically encoded signaling proteins to interrogate environmental conditions necessary for adaptation and survival, such as by detecting competitors and responding by secreting an antibiotic. The complete set of biomolecular interactions that a cell uses is often depicted as a connected network known as a genetic interaction diagram[2–4]. With complete knowledge of the interaction network of a cell it is possible, in theory, to predict how the cell will respond to any given stimulus. While achieving such predictive power in practice is not currently possible, even partial understanding of the interaction network is valuable and is a core concept in systems biology[5, 6]. For example, in the study of human health genetic networks are useful for understanding how pathways are dysregulated in disease or drug metabolism. Additionally there is interest in using genetic interactions to better understand novel and synthetic properties of microorganisms, such as the ability to digest environmental contaminants or produce biofuels from cellulosic biomass. Consequently, there is immense interest in novel methods to systematically map genetic interaction networks [7–11].

Mapping genetic interactions requires comparing the phenotypes of single gene perturbations to the phenotypes of double gene perturbations. Making libraries of single genetic knockouts is straightforward, but producing libraries of double mutants is supremely challenging. A common way to produce this library is to cross libraries of single knockouts to generate strains containing defined double-knockout combinations. Alternatively, the single-knockout library can be complemented with a library of additional genes of complementary function (Fig 1a). Genetic interactions within the libraries are scored by measuring the fitness (or growth) of each double mutant strain. Moreover, the culture conditions can be varied, such as by depriving the cells of an important nutrient or adding a drug, to study how genetic interactions change under these conditions. This can be used, for instance, to elucidate the targets of a drug or to deduce key proteins important for signal processing.

We have demonstrated a method to rapidly screen genetic interactions in a single culture. We produced genetic interaction libraries comprising two genetic perturbations and used single cell linkage PCR and NGS to reliably quantify the levels of every member in the library. This should make our approach useful for non-model bacterial systems, wherein genomic modification (by transposons, CRISPR-Cas9, or targeted modification) is the only requirement. In addition, the massive scalability afforded by droplet microfluidics should enable higher order interactions, such as 3-gene interactions, to be tested.

 

Source:

http://doi.org/10.1371/journal.pone.0171302

 

0 0 vote
Article Rating
Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments