Written by TKS Toronto Student, Maggie Li (email: ziyu.lili.maggie@gmail.com)

Imagine driving a car without a brake, at 150 km an hour. And then suddenly, you hit a bump.

Game over. Literally.

Similar to a car, CRISPR (which stands for clustered regularly interspaced short palindromic repeats), associated with a protein such as Cas9, is a pair a genetic ‘molecular scissors’ that will run nonstop. Until it’s stopped.


CRISPR without brakes: off-target effects

In a cell, that means these genetic scissors could be also cutting away elsewhere along the DNA (in addition to the sequence specified by the guide RNA in the CRISPR mechanism).

Known as off-target effects, this could have unforeseen consequences that are potentially deadly mutations in your cells:

  1. Point mutations: you get 1 faulty base pair in your DNA. This type of mutation is the cause of genetic diseases like sickle cell anemia.
  2. Deletions: an entire DNA sequence is deleted. That obviously isn’t great either — important proteins that would be transcribed from this sequence are now gone.
  3. Insertions: new DNA sequences now govern the proteins produced and cell function. The effect? Unknown, but could be deadly.
  4. Inversions: your DNA sequence loses its original meaning. It could still produce proteins, but likely nonfunctional or deadly ones(e.g. ATTGC becomes CGTTA)
  5. Translocations: a little more complicated but in a nut shell, the DNA sequence moves into another DNA sequence from another chromosome.

That’s the CRISPR mechanism in a nutshell. That running nonstop = not good.

Lots of bio words, but these are all mutations caused by the binding of CRISPR at off-target locations (because the protein doesn’t just get ‘turned off’ after it finishes it’s job).

Example: CRISPR can be used to target genetic mutations in muscle cells that cause muscular dystrophy. But the virus (a delivery mechanism for the CRISPR-Cas system) that is known to bind effectively to muscle cells also binds to liver cells. CRISPR activity in these other unintended cells is at best useless. At its worst, it’s a safety risk.

If you were driving a car without brakes, you would probably lose your driver’s license. Only thing is, there are no ‘CRISPR police’.

So, after Jennifer Doudna’s and Emmanuelle Charpentier’s teams discovered the CRISPR mechanism in 2012 (literally one of the biggest advances in molecular biology for over a decade), naturally, the next step was to find the brake for CRISPR.

Alan Davidson, Karen Maxwell, and Erik J. Sontheimer discovered anti-CRISPR in 2016. Their research identified natural CRISPR inhibitors in the bacterial defense system.

Here’s how it works and what the future of anti-CRISPR holds for science.


A quick crash course on how CRISPR works:

I will only outline how it roughly works (without going into the details, but here’s an awesome visualization), so we can move on to the fun stuff (anti-crispr!):

  • The CRISPR-Cas 9 complex (could be associated with another protein like Cas12), needs to bind to the DNA sequence. It does so by locating the PAM sequence, which is common to all genomes.
  • The complex unwinds the DNA, based on the guide RNA, which targets a very specific sequence in the DNA.
  • Cas9 now cuts the DNA, resulting in a break in the double strand.Following this, the cell will try to repair the cut.

The Anti-CRISPR Mechanism

In the fascinating world of little organisms, there are viruses called phagesthat kill bacteria cells by infecting them with viral DNA that will eventually cause the poor bacteria cell to explode and die.

That’s a phage (minus the sinister mustache).

Naturally, bacteria needed a defense mechanism so they evolved to take advantage of the CRISPR mechanism (which was also discovered in bacteria). In the defense mechanism, CRISPR uses phage DNA sequences to recognize phages when they attack.

These sequences form complexes with the Cas9 portein when phages attack and effectively destroys the phage DNA, preventing infection.

However, as a co-evolutionary mechanism (meaning the phage was also fighting back), phages evolved phage proteins that could bind to the Cas9 protein and protect the phage from the bacterial immune system.

The result: the phage with the sinister mustache wins!

How it works (with proper terms):

CRISPR-Cas systems defend bacteria against sequence-specific phage DNA material or plasmids (circular DNA). It’s a co-evolutionary mechanism that evolved along with the phage infection mechanism.

Small CRISPR RNA (crRNA) molecules guide these CRISPR-Cas systems to recognize and destroy foreign sequences that are complementary (in base pairs) to the crRNA.

Similar to how humans develop immunity, every time a bacterial cell with an active CRISPR-Cas system encounters a new mobile genetic element(MGE), aka foreign DNA material, it will incorporate the spacer (a small piece of the foreign DNA) into its CRISPR array. This way, next time it encounters the same MGE, the system can easily identify and degrade it.

However, these MGEs can evade detection by CRISPR-Cas systems via accumulations of mutations in their photospacer-adjacent motif (PAM). If you can recall from quick crash course on CRISPR above, that’s the sequence that (most) CRISPR-Cas systems need to bind to and destory the MGE.

Effectively, the MGE has evaded the bacterial immune system.

The first active inhibitors of CRISPR-Cas systems were discovered in a strain of Pseudomonas spp. phages.

Interesting fact: even though this phage DNA had photospacer sequences that CRISPR-Cas (specifically, type I-F) should have targeted, the phage easily evaded the Cas protein!

The Cas9 inhibitor binds to the CRISPR-Cas9 protein to prevent the destruction of the phage.

Sequencing analysis (DNA of the inhibitors) was then performed to identify the individual genes involved. 5 distinct proteins were then identified to be inactivators (AcrF1, AcrF2, AcrF3, AcrF4, and AcrF5).

Then, in a followup study following this initial discovery, 4 other distinctproteins were identified as inhibitors of type 1-E CRISPR-Cas systems (AcrE1, AcrE2, AcrE3 and AcrE4).

Effective evasion of CRISPR-Cas via Acr proteins. Original Paper.

What stood out to researchers about these anti-CRISPR proteins was a putative transcriptional regulator known as Aca1 (anti-CRISPR associated 1). This was immediately downstream of the Acr gene in the phage’s DNA. The regulator determined whether or not the Acr proteins were produced.

It was determined that the anti-CRISPR gene (which produces the Acr protein) and the aca1 form a single operon and the expression of the anti-CRISPR gene (aka the protein gets transcribed) is regulated by the aca1gene, which produces a regulatory protein.

When the Aca protein is produced, it ‘turns-on’ the acr gene. When the DNA sequence is transcribed to mRNA by RNA polymerase (which then is read to ribosomes to produce proteins), the acr gene is read by RNA polymerase and thus, produces Acr proteins.

However, when the Aca protein is not produced, the acr gene is off. And thus, no Acr protein is procued. In a nutshell, this is how it works. But, in reality, the specific mechanisms of an operon are a lot more complex than this.

The production of AcrF1 and AcrF3 proteins that inhibit the activity of crRNA Cas protein complexes, by preventing the destruction of external phage DNA (mobile genetic element). Original Paper.

Following this, there were 2 key discoveries:

  • Similar anti-CRISPR genes were observed upstream of the aca1 gene that was distinct from Aca1. This led to the discovery of Aca2 proteins. From here, 5 new families of type 1-F anti-CRISPR proteins were discovered (AcrF6, AcrF7, AcrF8, AcrF9 and AcrF10).
  • Using this guilt-by association method, type II anti-CRISPR proteins that could inhibit CRISPR-Cas9 were discovered (AcrIIC1, AcrIIC2, AcrIIC3).

Sofar, there have been 20+ unique families of anti-CRISPR proteins that have been found as inhibitors of type I and II CRISPR-Cas systems. They are usually quite small and between 50–150 amino acids.

No common features are conserved among all these protein sequences and none of these families have sequence similarities to other known proteins.

How the anti-CRISPR protein exactly inhibits (structurally and chemically) the CRISPR-Cas system is yet to be determined. There are been numerous studies that indicate the anti-CRISPR protein induces some conformational change (change in protein structure) of the CRISPR-Cas system.

Conformational changes such as dimerization obstruct the active site of the Cas9 protein (aka where it binds to the target DNA), effectively preventing any CRISPR mechanisms from taking effect.

Questions that remain unanswered:

  • The exact function of anti-CRISPR protein: how does it work in targeting the CRISPR-Cas system?
  • How were anti-CRISPR genes evolved?
  • Do all bacteria and archaea possess some mechanism to inhibit or inactivate anti-CRISPR activity?
  • How do anti-CRISPR proteins influence bacterial physiology and ultimately, human health?
  • What are the effects of anti-CRISPR genes on horizontal gene transfer (in co-evolving bacterial communities)?

An incredible amount of questions (well beyond this list) remain unanswered. The most promising and a very probably outcome of anti-CRISPR research is the potential to repurpose CRISPR-Cas systems into gene regulatory mechanisms (much like how methylation of DNA naturally occurs) to silence or activate gene expression in order to cure diseases.

It would also mean more precise CRISPR-Cas systems to allow increased control over the gene-editing mechanism, reducing off target effects to prevent undesired mutations.

Before we get there, step 1 would be to find more anti-CRISPR genes and proteins so that a larger database of genetic and structural protein information is available for analysis, in order to unlock new insights.


2020 and Beyond: Bioinformatics for Identification of Anti-CRISPR Loci

Check out the original paper here.

A team of scientists developed a bioinformatics pipeline to identify genomic loci (location of genes on a chromosome) containing Acr homologs through 3 computational approaches:

  1. Homology: comparing matches between 2 different samples of amino acid sequences in proteins or DNA sequences then, assigning a system of point values to identical/similar matches that occur in alignments.
  2. Guilt by association: used to infer the function of a poorly characterized gene from the known function(s) of the well-described genes that are co-expressed.
  3. Self-targeting spacers: new Acr subtypes based on known Acr protein families are identified by on investigating these mutations in the Cas protein’s CRISPR array.

The homology search found thousands of Acr homologs in bacterial and viral genomes, most of which are homologous to AcrIIA7 and AcrIIA9.

A key insight revealed was that only a small percentage (23.0% in bacteria dn 8.2% in viruses) have neighboring aca genes that form Acr-Aca operons. Additionally, of the Acr-Aca loci found in bacterial genomes, many did not have self-targeting spacers or a complete CRISPR-Cas system.

‘Based on these findings, we conclude that the discovery of new anti-CRISPRs should not be restricted to genomes with selftargeting spacers and loci with Acr homologs.’

By investigating acr genes and their loci co-existing with self-targeting spacers in the same genomes, 5 new subtypes (I-B, III-A, III-B, IV-A, and V-U4) were inferred. The additional data compiled throughout the study will be incredibly useful in guiding future development of new CRISPR-Cas regulators.

This was their research pipeline:

Image source

This is incredibly promising research that could help automate the research process by increasing the rate at which Acr genes and their respective proteins can be identified before being tested in a lab.

These amazing scientists have just scratched the surface in unlocking a phenomenal level of insight into the world of CRISPR and CRISPR regulators.

It’s an incredibly promising field with a myriad of possibilities and applications, with the potential to solve some really complex problems in the field of genetic engineering.