Curing disease with a flip of a switch
--
Imagine if curing diseases was as simple as getting up and flipping your light switch off (or for the lazy people out there, telling your Alexa to turn off the lights for you). Sadly, the version of life we’re all currently running doesn’t support that feature. But there’s a new update coming soon, the epigenetics update, that would allow us to do just that.
Epigenetics is a very sexy topic, one if the sexiest topics in all of genomics if I may say so myself.
It goes beyond individual parts of the genome and instead studies what makes some genes more active than others.
Unsurprisingly, this version of epigenetics has some bugs in it that need to be patched. In fact, some of humanities worst diseases are caused by flaws in gene expression. Diseases like solid tumour Cancers, leukemia and a wide range of intellectual disabilities, all which affect millions of people.
If we’re able to hack our epigenetics and alter epigenetic pathways, we could cure some of the biggest diseases in the world, at the flip of a switch!
Using a research paper by Tomasz P Jurkowski, Mirunalini Ravichandran, and Peter Stepper I’ve summarized the process of how we can take control of our epigenetics.
Step one — Understand how gene expression works
Silencing genes so they stop producing the proteins they code for is called methylation. It involves adding the molecule methyl (CH3) to the base cytosine in CpG islands of gene promoters. Demethylation, on the other hand, is the exact opposite and it involves removing methyl groups so genes are expressed more.
CpG islands are areas of a gene longer than 300 base pairs with a CG concentration higher than 55%. When a methyl group is added to these cytosine’s it stops transcription factors from binding to a gene so that it can’t be copied, essentially inactivating the gene.
Step two — Figuring out a way to target the gene you want
Finding a way to direct the methylation/demethylation to the specific part of the genome we want to impact is very important.
There’s no telling what would happen if you accidentally de/methylate the wrong gene, but side effects can range anywhere from no effect at all (phew), all the way to death(woah that took a dramatic turn).
To avoid that we’ll need something known as DNA Binding Domain (DBD) for targeting.
A domain is a subunit within a protein that can function independently from that protein. Since we’re trying to target a DNA sequence to methylate it, we need a type of domain that can specifically bind to DNA.
For all the smart people who couldn’t make the connection, that’s why they’re called DNA binding domains.
Choosing the right DBD that fits YOUR needs
Your list of DBDs to choose from is about as extensive as your list of food options when you’re ordering off the dollar menu in McDonald's. For those of you who can’t relate, it’s not a very big list.
That being said, the most common DBDs scientists work with are Zinc fingers (ZFNs) and TALENs. These are special because they allow connections between proteins and nucleic acids on the genome, two things that are usually not compatible.
It’s like trying to fit a square peg into a round hole, you need something to give it a little extra push, a hammer maybe. But in this case, the hammer is actually an amino acid chain stabilized by an ion.
Unlike the name suggests, zinc fingers aren’t actually fingers at all (I was shocked too). Each zinc finger unit is made up of 30 amino acid residues in a compact structure. This structure is stabilized by an ion made of a metal bound to two histidines and two cysteine residues.
I wonder what metal though…. hmmm. Oh yeah, it’s Zinc (did you see that coming?). Every module of zinc fingers can bind to 3 bases on the genome so they’re usually set up in arrays to try and target more and more base pairs.
There’s also a new kid on the DBD block.
Deactivated Cas9 (dcas9) is a version of Cas9 with deactivated nucleases (the scissor parts don’t cut anymore). So instead of cutting a gene, it just goes and sits there… doing nothing…not very useful on its own huh. Now, dCas9 hasn’t been as extensively used in labs but scientists say it has potential (don’t we all).
dCas9 is way simpler in that it mainly involves a single guide RNA (sgRNA) that has a gene sequence complementary to the one being targeted. That sequence is then attached to a dCas9 protein and guide’s it to the area of the genome it needs to go.
The only caveat is that for dCas9 to effectively bind to a target sequence, it needs a PAM sequence at the 3′-end of it. This PAM sequence is a series of nucleotides, NGG (where N is any nucleotide) that helps the Cas protein find its target. This is a problem because it limits the area of the genome we can target.
Below I’ve made an infographic to compare between commonly used Zinc fingers and dCas9.
Step three — Decide on your effector domain
Once you’ve chosen the DBD that finds your target gene, it’s time to pick another domain that does the de/methylating. This domain is called the Epigenetic modification domain and it’s what causes the intended effect of the whole process (the de/methylation).
If we focus in on methylation to silence an oncogene causing cancer, it narrows down the list of domains to choose from.
One of the most popular and common domains in labs is called a Krüppel associated box domain, KRAB domain for short. The KRAB domain doesn’t only represent a time when scientists finally gave something a good name, but it’s also one of the strongest repressors found in the human genome.
Another great methyltransferase(an enzyme that methylates) is Dnmt3a. Dnmt3a is special because it has a unique ability to multimerize with other similar enzymes. This means that even when one Dnmt3a is targeted to a gene, it can cause other Dnmt3a molecules to non-covalently bond to it and make a long strand of Dnmt3a’s down the gene. This, in turn, extends the effect of the methylation along a gene.
Step four — Link your DBD and effector domain
Yeah so just link them together with a linking domain. There’s a lot of them and the most important thing is to choose the right length as it will determine where exactly your effector domain methylates.
Step five — Deliver your methylation complex
The delivery method will depend on the size and charge of your domain.
Zinc fingers are small and cell permeable so they have the ability to move through the cell wall with their domain. This makes their delivery quite simple since no other modifications need to be made.
dCas9 is a larger protein and for it to go through the cell wall, it’ll need some help. The help comes in the form of cell-penetrating peptides (CPPs) which have the ability to translocate through a cell wall in multiple ways.
How do CPPs work?
Direct penetration: The most straightforward method of translocation which sees the CPP disrupting the cell wall through electrostatic interactions. There’s no real agreement on how exactly this mechanism works though.
Endocytosis-mediated translocation: Endocytosis is the process by which a cell absorbs outside material by wrapping the cell wall around it (almost looks like it’s ingesting it.)
Here’s the truth with CPPs though, no one is exactly sure how they work. Different studies using different methods show different results but it’s most likely a combination of some of the methods discussed above
Epigenetic editing to silence EpCam promoter in breast cancer
This research study by Nunna S1, Reinhardt R2, Ragozin S1, Jeltsch A1. Outlines the use of epigenetic editing to treat ovarian cancer.
The EpCAM gene is one that’s overexpressed in a lot of cancer patients and research shows that EpCAM overexpression is correlated with decreased patient survival.
The goal of this paper was to achieve targeted methylation of the EpCAM gene. In their methylation complex design, the researchers used a zinc finger DBD linked to a Dnmt3a effector domain. The complex was then targeted in SKOV3 cells (ovarian cells). The researchers were able to see an 80% reduction in the expression of EpCAM and cell growth rate decreased by 40%-60% which opens up the door for EpCAM silencing to be used therapeutically.
This is just one of the many examples of epigenetics being used right now to treat a variety of diseases out there. Each time, researchers are finding new methods and components that are making it easier to control gene expression.
As with all updates though, it takes time, but hopefully, in the future, we’ll see diseases that are being treated with as much ease as the flip of a switch.
Takeaways
Gene expression plays a very important role in different diseases. Gaining control would allow us to treat those diseases. This done through de/methylation.
To de/methylate a gene, a DNA binding domain is needed for targeting. This includes Zinc fingers and deactivated Cas9. DBDs are linked with effector domains (e.g KRAB, Dnmt3a) to produce a de/methylation complex.
CPPs allow for the delivery of large molecules into cells. Potential applications in drug delivery as well. Targeted epigenetic modifications are being used in therapies for a variety of diseases. However, there are still many improvements to the field.
