How emerging targeted mutation technologies could change the way we study human genetics

This is a guest post from Mari Niemi at the Wellcome Trust Sanger Institute. Mari is a graduate researcher whose research combines the results of human genetic studies with zebrafish models to study human disease.

The turn of the year 2012/13 saw the emergence of a new and exciting – and some may even say revolutionary – technique for targeted genome engineering, namely the clustered regularly interspaced short palindromic repeat (CRISPR)-system. Harboured with the cells of many bacteria and archaea, in the wild CRISPRs act as an adaptive immune defence system chopping up foreign DNA. However, they are now being harnessed for genetic engineering in several species, most notably in human cell lines and the model animals mouse (Mus musculus) and zebrafish (Danio rerio). This rapid genome editing is letting us to study the function of genes and mutations and may even help improve the treatment of genetic diseases. But what makes this technology better than what came before, what are its downsides, and how revolutionary will it really be?

Genetic engineering – then and now

Taking a step backward, the ability to edit specific parts of an organism’s genetic material is certainly not novel practice. In the last decade or two, zinc finger nucleases (ZFNs) and more recently employed transcription activator-like endonucleases (TALENs) saw the deletion and introduction of genetic material, from larger segments of DNA to single base-pair point mutations, at desired sites become reality. ZFNs and TALENs are now fairly established methods, yet constructing these components and applying them in the laboratory can be extremely tedious and time-consuming due to the complex ways in which they binding with DNA. Clearly, there is much room for improvement and a desire for faster, cheaper and more efficient techniques in the prospect of applying genome engineering in treatment of human disease.

The ‘revolutionary’ benefits of CRISPRs are certainly easy to recognise as soon as one begins working with them. The system is simply extremely efficient, effective and scalable to a level never seen before. The CRISPR-system can be designed to bind directly to genomic DNA using a custom-made stretch of RNA, which includes a 20 nucleotide-long sequence that is complementary to the genomic target site and thus binds to it. This RNA is stuck together with a special DNA-digesting protein (or endonuclease) termed Cas9, which creates a break in the DNA double helix upon binding. Crucially, this break then activates one the cell’s own DNA repair mechanisims (called non-homologous end-joining, or NHEJ), and this error-prone mechanism will often accidentally insert or delete small bits of DNA (called “indels”) at the target site. When these indels are targeted to areas that code for proteins they can cause the genetic code to become misaligned (often called “reading out of frame” or “frameshifts”), causing the protein to be misread and terminated early. If you do not want to simply break a gene, but instead want to change its sequence, you can use a different DNA repair mechanism called homology-directed repair (HDR), which replaces broken DNA with similar-looking (or “homologous”) DNA available in the cell. By giving the cell a ‘donor’ DNA-strand, and inducing a targeted break with CRISPR, we can swap the cell’s normal DNA for our own donor sequence during repair. Something similar can be done with ZFNs and TALENs, but HDR induced by CRISPR is much more efficient.

Improved methods for targeted mutagenesis

The increased genetic editing efficiency of CRISPRs offers possibilities for great improvements to many previously described research strategies, including loss-of-gene-function (LoF) analysis. So called knockout mouse and zebrafish lines, which have a specific gene deleted or broken, have been extensively used for decades to model human disease and to examine the functional implications of ‘knocking out’ normal function of disease-associated proteins. These knockout lines have historically been difficult and time-consuming to produce using the classical breeding and molecular genetics approach. To add to these complications, as many human diseases develop through complex interactions between several genes and environmental factors, it is often difficult to dissect which genetic changes truly alter disease susceptibility and manifestation. The emergence of the CRISPR technique has certainly stirred functional geneticists worldwide, who have been anxious to get their hands on this new technique – and for good reason: This emerging technique now allows for easy and fast generation of desired knockouts, often also in those genes which have previously been difficult or impossible to target.

Emerging applications for functional and clinical genetics research

Another extraordinary prospect for CRISPR-usage in LoF modelling is the possibility for efficient multiplexing of targets. This allows for creation of animals with two, three or even more target genes knocked out, making it possible to carefully examine the disease-contribution of each gene Due to its efficiency and scalability, there is now much interest for modelling gene function of complete sets of disease candidate lists implicated in genome wide association studies. Indeed, some of our work on CRISPRs at the Sanger Institute has also revolved around the idea of refining true associations between genes, expression networks and disease. The scalability and ease of the CRISPR-system will also ultimately allow us to add new knockouts to our Zebrafish Mutation Project, which aims to functionally characterise each of the 26,000 protein-coding zebrafish genes and their expression networks, many of which are also implicated in disease development in humans.

In addition to disease and gene modelling, the CRISPR-system brings forward a vast range of possibilities for new approaches to clinical research and disease treatments. As an example, CRISPRs may provide interesting opportunities for testing clinically associated mutations of individual patients in vivo by efficiently ‘flipping in’ the mutant human version in a model animal. Such studies would bear important information on which mutations are actually causing disease and possibly even provide diagnoses to families with children suffering from previously uncharacterised developmental disorders. Furthermore, published studies and our own experiments are allowing us to delete up to 6000 base pairs of targeted DNA, and new studies are pushing this figure upwards: this will enable us to delete or replace sizeable genomic regions containing non-coding regulatory elements and whole exons.

Looking much further into the future, we could even envisage using genome editing as a direct therapy. Imagine if we could remove stem cells from a patient with a genetic disease, use the CRISPR system to fix the genetic mutation in these cells, and then transplant these fixed cells back into the patient. But there are of course numerous hurdles to get over before this can become a reality.

Every method has a downside

On these accounts, one can hardly deny the CRISPR-system’s potential for originating new and pivotal methods for human genetics research. Nevertheless, great benefits often come with equal complications. For instance, the resolution by which the system can currently target the genome is on average one in every eight bases due to specific target sequence requirements. Yet, the most crucial hurdle that will need to be overcome before CRISPRs can be applied in treatments for human disease is the substantial risk of unwanted mutations elsewhere in the genome (“off-target mutations”), triggered by the technique’s high mutagenic efficiency. Studies have shown differing degrees of off-target mutagenesis in regions that look similar to the original target, but so far, accurate characterisation off-target effects on a genome-wide scale have not yet been carried out. Methods for reducing these potentially harmful mutations are under development, and significant improvements have been made in the recent months. Nonetheless, whole-genome sequencing of numerous tissue samples would be required to better understand the extent of genetic changes occurring through CRISPR-treatment before further steps towards utilising the system in e.g. stem cell transplantation therapies could be considered.

To conclude, the CRISPR-system offers an improved efficiency for previously established directed genome engineering methods and also brings forth opportunities for a variety of new genetic modification techniques. These approaches can be applied in research both in model organisms as well as human cell lines, and new methods are constantly under development. However, the novelty of the system still leaves many questions unanswered, and the extent of CRISPRs’ harmful effects remains unexplored ground. Despite these issues, the prospects for CRISPR application in research and therapeutics in the future seem fairly promising in these exciting times.