In order to improve CRISPR’s efficiency and accuracy, two research groups – one from Columbia University and another from MIT and Harvard – independently showed that it is possible to insert DNA sequences into the genome using the CRISPR enzyme coupled with transposon elements. This breakthrough enriches the CRISPR arsenal with an efficient way to insert sequences in the chromosome in a targeted manner.
Typical genome editing relies on DNA repair
Altering an organism’s genome is not an easy task. Cells have safety mechanisms, aiming to correct errors in the DNA sequence caused by mutagenic factors – such as radiation or chemical mutagens – or mistakenly-introduced during DNA replication. The DNA repair mechanisms are sophisticated enzymatic systems activated by environmental conditions or DNA damage, and protect DNA or try to revert it to its original sequence. Some common repair mechanisms are homologous recombination, non-homologous end joining, base excision repair, and many more. These mechanisms are not perfect and allow for mutations to be carried on to following generations, with critical implications on the process of evolution.
Genome editing is tricky because it needs to take into account and circumvent or take advantage of DNA repair systems. Yeasts and mosses, for example, employ very efficient homologous recombination machineries and easily incorporate heterologous sequences into their genomes and perform in vivo genome assemblies – when there is sufficient sequence homology. On the contrary, the Escherichia coli strains typically used for cloning have their repair protein RecA gene inactivated; this improves the stability of insertion vectors by reducing gene rearrangements.
Examples of how to use different repair mechanisms for genome editing. Non-homologous End-joining (NHEJ) joins double stranded DNA pieces after a break occurs. Foreign DNA pieces can be incorporated when present (A and B). Homologous Recombination (HR) relies on sequence homology and can insert or rearrange sequences in the genome. Image by Voytas and Cao, 2014, PLOS Biology (CC BY 4.0)
CRISPR-based genome editing systems require the cooperation of the native DNA repair systems to finalize the genetic modifications. The nuclease, usually CRISPR Cas9, makes a DNA lesion in a very precise manner. The repair enzymes fix the damage, and they may incorporate the heterologous DNA (which needs to be co-infiltrated into the cell) either by non-homologous end joining or homologous recombination (as seen in the image below).
Sequence insertion using CRISPR/Cas9. Figure by Khadotia et al, 2016, Frontiers in Plant Science (CC BY 4.0)
This cutting and rejoining is largely responsible for the low efficiency rate and the mutations introduced by CRISPR techniques (the other important factor being non-specific CRISPR targeting of other DNA sequences). It is possible to use CRISPR to do base editing without cutting. However, this technique relies on DNA base substitution and is not suitable for DNA insertions or larger-scale edits.
Performing DNA insertions or other modifications without double-stranded cuts can enhance the CRISPR techniques. And two American groups tried to do just that, by combining CRISPR with transposable elements.
Transposons, aka jumping genes
Transposable elements, or transposons (also known as “jumping genes”) are DNA sequences that have the unique property of moving themselves into different places within the genome. They were first observed by Barbara McClintock in maize plants, a discovery that granted her the Nobel Prize in Physiology or Medicine in 1983.
Transposons are worthy of a Nobel Prize because they contribute to the evolution of species. They cause gene duplications, deletions, and insertions, and they can be a means of horizontal gene transfer. They are an invaluable tool for geneticists, as transposons libraries can help determine essential genes or genetic and environmental fitness of particular genetic elements. In synthetic biology, transposon experiments are the starting point for constructing artificial chromosomes and minimal genomes.
There are many types of transposons, and they integrate themselves using different mechanisms. A transposon may be using a cut-and-paste manner or it may employ reverse translation, it may be autonomous or non-autonomous (based on whether the element contains the genes required for its own transposition), and it may either be species-specific or work in different organisms. The Mariner-type transposon illustrated below was first observed in the fruit fly Drosophila but can be integrated into the genome of several different organisms, including humans.
Structure of a Type II transposable element and mechanism of its transposition. The enzymes transposases recognize the flanking sequences, excise the element, and insert it into a new genetic locus. Image by Mariuswalter (CC BY-SA 4.0 )
CRISPR and transposons
In 2017, Joseph Peters and his collaborators discovered that some transposable elements – Tn7-like bacterial transposons in particular – are associated with the CRISPR system of their bacterial hosts. The hypothesis is that these transposable elements hijack the CRISPR proteins to facilitate their insertion into the genome. Using this, would it be possible to generate a programmable insertion system using CRISPR and engineered transposons?
Two research groups seem to say yes. Samuel Stenger’s group from Columbia University reported in Nature a study which combines transposon insertion with CRISPR precision. They call their system
INTEGRATE (INsert Transposable Elements by Guide RNA-Assisted Targeting). Sanne Klompe and her collaborators identified a transposable element from Vibrio cholera, Tn6677. This jumping gene squeezes itself into the chromosome, taking advantage of disturbances on the DNA structure caused by the CRISPR binding. The researchers characterized the system and developed a customizable system where a transposon can be inserted into different genetic loci of E. coli.
At the same time, a second group from MIT reported in Science another insertion system that uses the same principle. Jonathan Strecker and his collaborators in Feng Zheng’s lab found a compact CRISPR-associated transposon (CAST) in the cyanobacterium Scytonema crispum. The researchers characterized the system and modified it to make it programmable. They were able to insert DNA sequences of various lengths into different loci of E. coli. The reported efficiency reached 80%, as compared to 1% of typical Cas9-mediated insertions.
The cyanobacterium Scytonema crispum (left) and the bacterium Vibrio cholerae (right) (Source: Wikimedia commons)
Is the DNA insertion problem solved?
Both research groups showed that it is possible to precisely insert DNA sequences in a desired chromosome location, but only in bacteria. As CRISPR is a bacterial system, there shouldn’t be CRISPR-associated transposable elements in eukaryotes. It will be interesting to see whether INTEGRATE or CAST systems can function as they are in other organisms, or if the eukaryotic transposons can be engineered to associate with a non-cutting CRISPR protein.
There are several genetic engineering applications that may require chromosome insertions of long DNA sequences, such as gene therapy and crop improvement. Developing a cut-free insertion system is definitely a step towards more versatile, safer, and less laborious CRISPR applications.4