Scientists develop unique CRISPR-based method for more precise, systematic gene tagging

The genome editing tool known as CRISPR/Cas9 has opened many doors for biomedical research by allowing scientists to precisely edit DNA sequences and alter function. The technology offers hope for correcting genetic defects and preventing disease.

Scientists deploy glowing fluorescent tags to help them visualize the precise location of a protein of interest using gene editing. Researchers at the Allen Institute for Cell Science in Seattle, WA, have developed a CRISPR-based method to more precisely tag the genes encoding important proteins better than previously possible. The team is building a library of stem cell lines that tags 10 important genes linked to key cell structures and will make these cultures available to the research community.

Ruwanthi Gunawardane, Brock Roberts, and Amanda Haupt, and colleagues at the Allen Institute for Cell Science, are the authors of the paper "Systematic gene tagging using CRISPR/Cas9 in human stem cells to illuminate cell organization," published October 15, 2017, in Molecular Biology of the Cell, a journal of the American Society for Cell Biology.

This new tagging and screening strategy has very little margin for error and avoids the "scar" on the genetic code created by other tagging methods, which can produce unwanted effects on gene regulation, researchers say. The library of tagged stem cells being constructed represents the beginning of what researchers hope will be a much larger collection of cell lines with tags on specific proteins studied extensively by cell biologists. The intent is that these stem cell lines will serve as genetic and pharmacological screens, and models for disease as the cells differentiate.

For a closer look at how this stem cell library was created, read the following Q&A with responses provided by Gunawardane (RG) and Roberts (BR):

What are the main benefits and drawbacks of endogenous gene tagging?

RG: Endogenous gene tagging, unlike other methods, introduces the tag proximal to the gene of interest at its genomic loci. This results in the expression of the tagged fusion at physiological levels since it is regulated by its native promoter and allows one to study the localization and dynamics of the protein and structure under "normal" conditions. Traditional overexpression methods such as transient transfections with plasmids or viral transduction can introduce many unwanted effects including mislocation and/or cytotoxicity. Unlike overexpression systems, endogenous tagging also results in reduced fluorescence background and mislocalization artifacts and produces exceptional images. The drawback of endogenous tagging is that it is inefficient and can often be imprecise, but both these issues can be addressed as discussed below.

How is HDR-mediated tagging different/better?

BR: HDR (homology-directed repair) is the cell's repair mechanism that results in the incorporation of a foreign sequence into the genomic DNA (such as done here with GFP). Therefore, HDR allows one to incorporate a tag precisely into a target locus/gene to produce an endogenous fusion protein. Other methods of introducing a tag sequence leave a "scar" that may lead to unwanted effects in gene regulation, for example.

What makes your CRISPR/Cas9-mediated tagging protocol special, unique, novel?

RG: We use a modified ribonuclear protein method for our gene editing. To minimize continuous Cas9 activity that can introduce many unwanted on and off target editing (traditional plasmid-based systems), we use Cas9 protein pre-complexed to the crRNA and donor template. We also utilize high quality commercially synthesized crRNAs that have been designed to minimize potential off targets and donor templates with features for optimal editing. The successful application of this method on a large number of targets using a stem cell model suggests this particular method may be broadly applicable across many loci of the human genome. The extent to which we have performed quality control on this collection of edited cell lines is an additional unique feature of our work.

Why did you choose to label these 10 particular intracellular structures?

BR: Our goal is to produce lines that label most of the major structures commonly studied by cell biologists. This group of 10 structures is just the first set in this larger collection. The specific proteins (genes) were chosen because they have been used extensively by the cell biology community, they localize and outline the structure of interest, and they have been previously tagged (although not always endogenously) successfully with GFP.

You are making these cell lines available for use. How do you hope the scientific community/society will benefit from them?

RG: We hope the scientific community will use these cell lines to study various aspects of basic, developmental, and translational biology in a cell model that is diploid and nontransformed. We hope that basic cell biologists more accustomed to transformed cell lines will take the plunge into stem cells because they need not produce their own cell lines but instead can start with this collection and that these scientists will benefit from the relatively "normal" biology and pluripotency of these cells. We hope these lines will also be used for genetic and pharmacological screens and disease modeling by differentiation into various cell types.

In labeling these 10 cellular proteins, what is an example of a challenge you faced and how did you overcome it?

BR: While the challenges of systematic endogenous tagging of human genes in stem cells were relatively unknown at the beginning of the project, the efficiency of HDR, achieving precise incorporation of the tag, and scaling these methods for multiple genes in parallel were some of the challenges we faced. We achieved this by devising a FACS (fluorescence-activated cell sorting) enrichment strategy to allow identification of edits that are as low as 0.05%, novel PCR-based assays for identifying precisely edited clones, and semi-automation and multiplexing of screening assays for parallel processing of approximately 100 clones in our gene editing and quality control workflow.

You did a lot of quality control. Do you expect the clonal lines to remain stable?

RG: We do find that the tagged protein levels and pluripotency remain stable over time, although this testing has been limited to 10-20 passages post editing. However, we do observe some genome instability in some of the clonal lines upon continual passaging. This is a feature of stem cell lines that has to be considered and monitored, if applicable to the study, as part of routine culturing of stem cells.

What will you do next?

BR: We plan to expand our collection and continue to report the data resulting from those efforts. We have also initiated tagging genes that are not expressed in stem cells, but turn on after differentiation (such as cardio-specific genes). Additionally, we plan to explore in much greater detail the dynamic behavior of the structures both in stem cells and cardiomyocytes differentiated from this collection of gene-edited iPSCs. We also plan to soon report on the performance of machine learning models that infer the relationships between cellular organelles, which are in turn based on a high throughput microscopy pipeline that is churning out data from the cell lines in this report.

Anything else people should know?

RG: In general, the success we have had in editing the 25 genes we have targeted to date strongly suggest that the human genome, especially in stem cells, is amenable to editing, which potentially opens up many more exciting and important editing possibilities for other genes of interest for researchers. We share our editing and screening strategies/methods, the crRNA sequences (allencell.org), and donor plasmids (Addgene) for those interested in tagging the same genes/proteins in their own cell lines.