Split-Cas9 design and its implications for AAV-mediated CRISPR delivery

Wright AV et al, PNAS 2015,
www.pnas.org/cgi/doi/10.1073/pnas.1501698112, copyright: PNAS

In their paper, Wright et al. separated the Cas9 enzyme into two distinct parts, a nuclease lobe and an α-helicase lobe. The two different polypeptides were shown to be brought together by the guide RNA, reconstituting the active CRISPR complex. One of the most important applications of this system is that two smaller regions of Cas9 could be cloned into AAV, overcoming capacity limitations of this vector. Moreover, there would be more room for larger or inducible promoters to spatiotemporally limit Cas9 expression. Inducible dimerization domains may also be applied to regulate expression. Readers should be aware that this split Cas9 sytem seems to be less effective than wild-type Cas9, WT Cas9 generated indels in HEK293T cells with around 22% frequency, but this was only 0.6% with split-Cas9 (synchronized cells showed slightly increased indel formation). This warrants further optimization, but having the Cas9 on two different polypeptides definitely has advantages. The lower indel rate compared to WT Cas9 is in accordance with the split Cas9 from the Zetsche B (http://goo.gl/J5Ogqu) paper. The difference in the Wright and the Zetsche paper is that in the latter, the Cas9 is cut in half (to obtain an N- and a C-terminal domain), whereas in the Wright paper the nuclease lobe contains a short N-terminal part and a long C-terminal part (with a short linker in between) and the α-helicase lobe is consisted of 'middle' amino acids. The Wright strategy split Cas9 is able to dimerize spontaneously in the presence of guide RNA, but the Zetsche split Cas9 needs chemically inducible dimerization domains for function.  

Original paper: Wright AV, Sternberg SH, Taylor DW, Staahl BT, Bardales JA, Kornfeld JE, Doudna JA. Rational design of a split-Cas9 enzyme complex. Proc Natl Acad Sci U S A. 2015 Feb 23. pii: 201501698a, http://www.pnas.org/content/early/2015/02/18/1501698112.long

The development of immunological memory in bacteria 

CRISPR mediated bacterial immunity, with permission
from NPG, from Yosef I et al, Nature 2015

Two papers published last week in Nature describe the amazing development of bacterial immunological memory. Previously it has been thought that in contrast to vertebrates, bacteria can not remember previous infections. Now it is obvious that the original function of CRISPR is to fight against invading viruses by cleaving specific parts of the viral genome. During an infection, part of the viral genome is stored in the bacterial genome as a library, from which a matching sequence can be utilized upon subsequent infection. How this happens remained elusive until recently, now it is shown that components of the CRISPR system (Cas1, Cas2, Csn1, Cas9 and tracrRNA) act in concert to find appropriate target regions in the viral genome, and insert a copy into the bacterial genome. Finally it makes sense, why there is a so-called PAM site. This is a very short motif adjacent to the recognition region, which is an absolute sequence requirement of the system (e.g. this is NGG for the S. pyogenes CRISPR system). Most importantly, when pasted into the bacterial genome, there is no PAM site next to the sequence. So that is why bacteria do not attack themselves, since the PAM site is only in the virus, leading to cleavage. That is how bacteria evade autoimmunity. Smart!

Check out the original papers here: http://goo.gl/kz2X1V

Colon carcinoma development in the dish 

With permission from NPG, licence number:  3575540372115

Researchers from Japan were able to model the development of colon cancer in intestinal organoids. CRISPR was used to disrupt APC, SMAD4 and TP53 tumor suppressor genes and to introduce tumor specific mutations into KRAS and PIK3CA in intestinal organoids isolated from human patients. Most interestingly engineered normal tissue was not converted to highly invasive tumor. On the contrary, when the engineering was done in chromosome instabile adenomas, the organoids formed macrometastatic colonies in mice. This study shows the versatility of genome editing to model a highly complex disease process.

Matano M, Date S, Shimokawa M, Takano A, Fujii M, Ohta Y, Watanabe T, Kanai T,
Sato T. Modeling colorectal cancer using CRISPR-Cas9-mediated engineering of
human intestinal organoids. Nat Med. 2015 Feb 23.

http://www.nature.com/nm/journal/vaop/ncurrent/full/nm.3802.html

MPN Genetics Announces Research Grants with focus on CRISPR

Funding option is available for researchers developing genome editing therapies for myeloproliferative diseases. A patient advocacy group called MPN (Myeloproliferative Neoplasms) Genetics and the Leukemia & Lymphoma Society announced "The 2015 MPN Challenge" grant program, one priority area being genome editing in MPN diseases. Deadline is April 1st. Check it out here: http://www.mpnresearchfoundation.org/2015-Request-for-Research-Proposals

Genome editing and the CRISPR

Genome editing: break and repair

In the modern era of genetics, it is simple to read the human genome, determine every single nucleotide in your cells can be done in a few days at an relatively low cost. However it is extremely challenging to change something in the DNA, e.g. to correct a disease causing mutation. This molecule is extremely stable and well-protected from any interventions, so until recently it was hardly possible to ’edit’ the genome. Editing means everything that writers or publishers do with a written text: cutting, pasting  or changing letters.  Genome editing is similarly done with the four  nucleotides (A, G, C, T) in the DNA. Genome editing is not a completely new technolgy researchers have been using other enyzmes (like the so-called zinc  finger nucleases and the TALENs) to perform genome editing. But there is a huge differnce between the these and the novel technology, CRISPR (the abbreviation is very complicated: Clustered Regularly Interspaced Short Palindromic Repeats) The former technologies use peptide decoders to find the certain part in the long DNA that has to be modified and therefore it is challenging to figure out what peptide is the one that exclusively binds to my favourite sequence. But CRISPR is a real gift from the nature: it uses a complimentary RNA, which will just pair with one strand of the DNA. As you may know, the whole genome has been already deciphered, we know the sequence of the human genome. If it was written on sheets, the tower of books would be 169 meter high!

So, if the genome sequence is known, then it is simple to design the RNA (this will be called the guide RNA, as this guides the editing enzyme to the genomic region) that is complimentary to the DNA. As you know: adenin [A] pairs with timine [T], guanine [G] with cytosine [C] and vice versa. Let’s see: here is a diseases gene sequence: AGCTGTGCTGTCGATGC. I want to edit this. So first I have to direct the editing enzyme (this is called the Cas9) to this sequence with the guide RNA, which has the sequence of: TCGACACGACAGCTACG. So look at this!

The gene to be edited: AGCTGTGCTGTCGATGC

The guide:                       TCGACACGACAGCTACG

These are pairs! So to be simple: the guide RNA can select the region of interest from 3 billion base pairs quickly and effectively.

OK, so we have something that selects a gene of interest in the genome. The next question is: how this part of the genome will be edited? Basicly, the Cas9 protein will cleave the DNA and create a double strand break (so it cuts the DNA to two parts). Cells doesn’t like this, a break in the DNA is a nightmare for them, so they want to utilize one of many repair mechanisms to rejoin the injured DNA. Among these is the so-called NHEJ – non-homologus end joining. This repair is an error-prone repair, which will create mutations (some difference compared to the original), otherwise the Cas9 enzymewill cut it again and again. So it has to be some change in the sequence so that the RNA guide will not recognize it anymore. This NHEJ can kick out a gene and inactivate it. So let’s say there is a mutation in a gene that leads to an abnormal protein that has abnormal function, like a tumor formation. We have two copies of each gene (one from the mother and one from the father), but even if we have ONE normal copy, there is that bad gene that drives tumor growth. This is called the DOMINANT effect, with one bad copy, there is still disease. But using CRISPR this bad gene coud be knocked out, so the normal copy will take over and the disease can be cured! Forever since it the DNA is modified!

But you can say, where is the editing? Where is the real correction? Fortunatley this is also possible. If you happen to deliver a repair template to the cells (from which the DNA could be repaired) you can insert anything into the genome, exactly at the cut site, where the DNA is broken. This is called the homology directed repair (HDR). So if someone has a so-called RECESSIVE disorder (when both copies of the gene is mutated), the gene can be corrected with this method.

And applications are indefinite: it is feasible to do it in plants, bacteria, or other animals to cure diseases, or to create even new lifestyles. This system can be also used as a novel technology to study the function of genes. I will tell you even more, if you come back to this site!