I cured breast cancer as a 16-year-old using gene editing 🧬💉…
Creating a Novel plasmid capable of tumour suppression using Benchling
I like to call myself something of a Youtube connoisseur. Although I hate to admit it, I love binging Youtube videos (who doesn’t!). During one of my usual youtube binge sessions after school one afternoon, I stumbled onto a video titled: Genetic Engineering Will Change Everything Forever — CRISPR’.
Now as a 13-year-old at the time, my interest was piqued. Despite knowing nothing about genetics, I sat through the full 16 mins of the video listening intently about some revolutionary technology called CRISPR. It seemed cool, but that was it, just another Youtube video about a cool topic that I would forget the next day.
That was until the same video appeared in my recommendations earlier in 2022. I rewatched the video, mainly for nostalgia’s sake. But since I now knew a thing or two about CRISPR, I decided to dive a little bit deeper.
After many rabbit holes, I concluded that I wanted to use CRISPR. This sounds absurd coming from a 16-year-old without access to a million-dollar lab, but I figured that wouldn’t stop me.
By some convenient miracle, I was introduced to Benchling later that week. Benchling is a modelling application used by real scientists, allowing them to simulate experiments and understand concepts without access to a lab.
I wanted to use my knowledge of CRISPR to solve a real problem: cancer. Often referred to as humanity’s greater foe, 10 million people die from this terrible disease every year. What was interesting was that our body has in-built mechanisms known as proto-oncogenes specifically designed to stop cancer. However, cancer is caused when these genes malfunction.
Using Benchling and CRISPR, with some help from some really helpful tutorials (linked at the end of the article), I designed a plasmid and guide RNA that targets a specific proto-oncogene (p21) and turns it into a possible solution to solve breast cancer, one of the most common forms of cancer. Get ready to learn a lot of words you can’t pronounce (I promise medical terms are fun to learn😅) as we embark on a journey to fight humanity’s biggest foe.
Feel free to also check out the videos below where I walk you step by step how to create the plasmid.
If you want to see the finished Benchling model, skip to the end.
Before we start, we have to get a better understanding of how our body naturally fights off cancer, and how cell division works. I don’t dive deep into what exactly cancer is in this article, but feel free to check this article I wrote or this one by the National Cancer Institute (they definately know their stuff).
Cells in your body are constantly multiplying. They do so in an organized pattern, following what’s known as a cell cycle. The cell cycle dictates when a cell should multiply and when it shouldn’t. When your cell is not multiplying during the M phase, it is growing and getting ready to multiply.
The cell cycle is propagated by ‘checkpoints’. Think of it like a road with border security that crosses several countries (it’s a really long road). The border security checks to make sure you aren’t doing anything illegal, and you have to pay a tourist tax to get through. If you don’t pay, you can’t go forward. But since there’s a lot of traffic behind, you can’t go back either. You’re just stuck. Since you are driving across different countries, you have to pay with a different currency each time.
The checkpoints in cells ensure the cell is healthy and functioning as normal. With cells, the ‘currency’ by which you pay is cyclin which is specific to the checkpoint, and the border security operator is CDK (Cyclin-dependent kinases). Once the cyclin binds to a CDK, it releases a growth factor. Once a threshold of these ‘green lights’ is reached, the cell cycle progress forward. However taking the border-crossing analogy to an extreme, if a cell fails to meet a checkpoint they undergo apoptosis, known as cellular suicide.
Continuing to the previous analogy, think of cancer like a sneaky criminal. Under normal circumstances, the criminal would be arrested. However, cancer has a trick up its sleeve. It can produce its only cyclin, regardless of the fact it is clearly not healthy. The CDKs are easily fooled (I mean can you blame them, they don’t even have eyes!) and allow the cancerous cell to progress and eventually multiply. Not only do cancer cells illegally bypass the CDKs, but they drive extremely fast. In other words, they go through the cell cycle at an accelerated rate faster than normal cells usually do.
You are probably thinking that the immune system (in the case of the analogy, think of it as the police) would notice all the commotion, and you would be right. However, cancer cells are able to ‘bribe’ the police and thus get the immune system to instead protect the tumour. The mechanism by which this works is still unknown, but it is important to note.
Tired of my elaborate analogies yet? Don’t worry, we’re just about to start diving into the technical details behind it all. Scientists have identified CDK 4 & 6 and Cyclin D & E are the culprits behind breast cancer, a type of cancer that starts in the breast tissue. However, recall how I said your body has in-built mechanisms to detect cancer? Well, there are two main ones. The first one is fairly simple to understand. CDK4/6 inhibitors are activated by protein 21 (p21). The way it does this is by phosphorylating (adding a phosphate ion) a tumour suppressor protein called retinoblastoma (Rb). In other words, it adds the final ingredient to Rb for it to begin its work, which is to stop the cell cycle. More p21 = More CDK4/6 inhibitors = less CDK = stop to cell cycle and cell death.
The second method is a little more complicated. Remember our trusty old friend the immune system. Well to be activated, two criteria must be a must.
- An antigen must bind to the cancer cell, ‘marking’ it for death by T-cell (a type of cell in the immune system). It is fairly easy as there are a lot of antigens roaming around your body
- A costimulator is needed, which essentially is just verifying that the cell is indeed a cancer cell. However, our body does not produce that many costimulators. Cytokines are an example of the most prevalent costimulator and are also the costimulator with the most research surrounding them, so we will focus on them.
Remember when I said the immune system could be ‘bribed’? Well, when an abundance of immune cells come together, they are capable of overpowering cancer. This response is not natural in your body, which is why we will artificially stimulate it.
Cytokines are produced by a group of genes known as interferons (IFN). Although our body has various copies of IFNs already, they are susceptible to being mutated, which is why we will insert another copy reducing the possibility of this occurring tenfold. Additional copies may result in the over-production of cytokines would make our immune cells attack healthy cells, which we don’t want. The IFN we will use today is mIFN-α (murine interferon-alpha), as this is a naturally occurring gene found in mice. Since mice and humans share very similar anatomy and physiology, it makes a perfect fit for the experiment.
If you’re a little bit confused about all these proteins or how proteins are made, check out my article (last shameless plug I promise) where I explain how genes encode proteins. I have also created a flowchart to simplify things:
The last thing we need to understand are what plasmids are. Plasmids are circulator pieces of DNA found in bacteria. They offer the unique ability to edit the DNA sequence, and thus we can alter the role of the bacteria. A regular E.coli cell can be edited to glow green with the simple insertion of GFG (green fluorescent gene). These genes can be artificially crafted by humans or exist in other cells. We can then take out the section we want to use and insert it into a human cell, via CRISPR and a guide RNA.
So that’s all you need to know, which admittedly is a mouthful. But don’t worry, I’m going to break down the process in a step-by-step process, and explain some things further along the way🚀.
There are 2 main steps to the process.
- Inserting a copy of the p21 copy into the mIFN plasmid, using the mIFN as the template plasmid, thus maximizing the effectiveness of the plasmid
- Designing the guide RNA necessary to insert the plasmid into the human genome.
Retrieving the Plasmid
First, open a new project in Benchling and import a new DNA/RNA Sequence. The plasmid we will be using is found here. The plasmid has a couple of useful qualities, namely several protein promoter sites as well as Neomycin (NeoR). Protein promoter sites are essentially places where RNA polymerase (what copies the gene and ‘implements it’) can begin transcription, moving in the 3'–5' direction (5’ to 3’). Think of them as detectors. Tumours release immense amounts of signalling molecules. Once a specific molecule is detected, the promoter site activates IFN. Without this activation, IFN will not replicate and thus have no effect. For example, a correlation between CMV promoter activation and cancer has been observed. Similarly, the SV40 promoter is activated in the presence of malignant tumours. Furthermore, Neomycin is a selection marker. Selection markers allow us to verify that our gene is working successfully, and allow scientists to monitor the effects of the edit.
The mIFN gene looks as follows:
The left showcases how the gene will look in Benchling, while the right image shows us what part of the gene corresponds to which mechanism. The star of the show is the IFN gene (IFNg). We want to preserve this gene sequence, We can see the schematic on the right tells us where this gene is placed. It is between restriction enzyme sites NheI and NotI. In order to ensure we don’t lose or write over IFNg, let’s label it (right-click). Notice how the promoter site precedes the IFNg. This is further evidence of how promoter sites ‘activate’ genes
Note: Restriction enzymes allow how to cut in specific locations, and operate similarly to how passwords work. There is a specific recognition sequence that a specific enzyme recognizes. After recognizing the sequence, it cuts at that location.
Inserting p21
Now we have to import another plasmid (following the same procedure) that encodes for p21. The plasmid we are using is here.
Once we have imported the plasmid, simply select both the p21 annotation and the hGH poly(A) signal and right-click to copy. The poly(A) (also known as polyadenylation) signal is the terminator sequence. In other words, this tells RNA polymerase to stop transcribing the gene. Note that one gene can encode multiple proteins, that’s why a terminator sequence is necessary. Moreover, RNA polymerase will infinitely transcribe in the 3’-5’ direction until it is given a stop signal.
If you want to learn more, read this awesome explanation by Khan Academy: A poly(A) is a sequence of nucleotides that marks where an RNA transcript should end. The polyadenylation signal is recognized by an enzyme that cuts the RNA transcript nearby, releasing it from RNA polymerase. Oddly enough, RNA polymerase continues transcribing after the transcript is released, often making 500–2000 more nucleotides’ worth of RNA. Eventually, it detaches from the DNA through mechanisms that are not yet fully understood. The extra RNA is not usually translated and seems to be a wasteful byproduct of transcription.
Now we have to find a suitable location to insert the gene as we can’t simply copy-paste (if only it were that simple). To do so, we have to find the right enzyme site. This process is known as a restriction enzyme digest. After a little bit of searching, we can see a gap between the IFNg and the promoter sites to the right of the gene (after opening the Linear map). Now if you’re confused about why there are promoter sites on both sides, then it is important to note that the DNA is circular. Therefore, transcription can occur in either direction, hence it is important to have promoter sites in 5’ direction (left) of gene.
Both NotI and EagI share the same recognition site, but once we double-click each enzyme (to view more information) we can see that EagI makes 2 cuts while NotI makes 1 cut. This is important, as we don’t want to remove an entire segment of DNA but rather just make a nick.
Before I continue, notice the black jagged line. This denotes where exactly the enzyme will cut, which in our cases forms a sticky end. A sticky end is simply an overhang of bases that remain after the cut. Contrary to sticky ends are blunt ends, which are straight-line cuts.
To better distinguish, I have included a diagram of two different enzymes and the cuts they make. However, the difference is unimportant in our current simulation.
In a real lab setting, the enzyme must be applied to the plasmid to cut at the recognition site and then insert the p21 gene. However, to simulate this on Benchling, we simply hover over the NotI label and highlight the recognition sequence, ensuring we don’t highlight any necessary bases. You will notice we cut one base off from IFNg, but this is fine as our body has self-repair mechanisms (such as DNA polymerase)which will fix the single-base deletion caused by our enzyme. Once we have highlighted the recognition sequence, simply paste the p21 and poly(a) sequence (this will delete the recognition site). The annotations will be copied over, so we can see the resulting plasmid
Adding Kozak Sequences
Hold your hats folks as we still have one more thing to do. We still need a start codon that begins translating the DNA. But wait, I thought that was the role of the promoters. Well yes and no. Promoters initiate transcription while start-codons initiate translation. Below is a helpful way of thinking about it:
“Transcription is the synthesis of RNA from a DNA template where the code in the DNA is converted into a complementary RNA code. Translation is the synthesis of a protein from an mRNA template where the code in the mRNA is converted into an amino acid sequence in a protein”.
Another analogy to use is to think of mRNA are the messenger (hence the name) that goes and gets the job, while DNA is the lazy king that likes to boss everyone around and tell them what to do. The promoter is the king telling the messenger to start writing down what he wants. The start codon is the first instruction that the king tells the messenger.
There are several start codons, also known as Kozak sequences, ranging in application and effectiveness. We will use the GCCGCCACC sequence. We must insert a Kozak sequence corresponding to the p21 gene (as the IFNg already contains a different Kozak sequence). To do this, we perform another restriction enzyme digest.
Identifying the AccI as the most appropriate enzyme, we again delete the recognition site and replace it the with Kozak sequence above. Benchling will auto-fill the complementary strand. Then we label it as the Kozak sequence. Similar to before, DNA polymerase will auto-correct the deleted bases toward the end of IFNg.
Final Plasmid
After adding p21 and a Kozak sequence, our final plasmid will look like this (see below). Before we move on, I want to explain the role of some of the labels which we didn’t utilize. Most notable are the ‘ori’ sites, which are simply the origin of replication. This allows the plasmid to replicate inside the bacteria, allowing us to have several copies of the plasmid. This is because plasmids are very expensive to buy, so this helps save us some costs. Secondary are the introns. Since we are going to insert it inside a eukaryotic cell, we have to take into account the fact that eukaryotic cells don’t just go from transcription to translating, there are post-transcription modifications to protein/mRNA. Using our king and messenger analogy again, you can think of introns as ministers who check the king’s instructions for any flaws and correct them as necessary. Lastly is the label denoting ‘AmpR’. Ampicillin is another selection marker similar to NeoR, although it is not applicable in our case as we are dealing with it. Before we end, let’s give our plasmid a name.
How to implement our plasmid into our body
Now our plasmid seems cool and all, but how could we take this a step further and insert it into the human genome? To do this, we can leverage a neat tool known as CRISPR.
However, to use CRISPR we must identify several components.
- The spacer RNA is the sequence we are inserting. In our case, we are inserting our entire plasmid
- The gRNA, or the molecule that tells us where our sequence will be inserted.
To identify the gRNA, we have to first pick a gene to insert our plasmid into. The gene we have chosen is CDK2NA. This gene is a member of the alpha interferon gene cluster on chromosome 9 and is activated in response to viral infection as a key part of the innate immune response. We know CDK2NA already serves a prevalent role in tumour suppression so by inserting our plasmid, we’re gonna give it a huge boost. In reality, this would make the spread of cancer significantly slower rather than cure it altogether, but it will allow doctors to make a diagnosis while the tumour is in earlier stages (with less critical symptoms).
First, we import a CRISPR guide for CDK2NA. After searching for the gene, go to advanced settings and open up all the transcripts/exons, then select all (via ctrl A or command A). This ensures our guide will be able to target all transcripts/exons, maximizing effectiveness
Transcripts, also known as exons, as essentially binding sites for gRNA. The more binding sites that are available, the lesser the probability to miss. One exon has 1–5 binding sites throughout the gene, with each gene having a dozen exons.
We need to find all the exon 2s or 3s. The reason why we are targeting the second and third exons is that the targeting site must be later on in the gene in order to knock in (synonymous with insert) a plasmid at the end of the existing gene.
Since the linear map showcases the majority of exon 2s to be between 24kb and 24.5 kb, we will insert here. That also means we cannot target the exons that do not fall in this range.
We must determine a specific target region. For our gRNA, our target region is the large exon on the top. We use this as all the other exons fall within the range of the large exon and therefore we can encompass all of them. Then, we press ‘design CRISPR’ and the plus button confirming our target region.
This will then give you a list of guides. It will also show on-target scores and off-target scores. What on-target represents is the pharmacologic side-effects of a successful test. Off-target refers to adverse effects as a result of modulation of other un-intended targets. The higher the score for on-target, the better the biological symptoms. The higher the score for off-target, the less damage it may cause if it goes awry. Therefore the scores explain the theoretical success of the guide. We want the values both to be as close to 100 as possible.
After some analysis, we have selected the guide(s) below:
Note that usually we pick 2–3 guides to ensure certainty in knocking in the gene (based on the score, they have varying levels of efficiency). We picked 3 guides, and you can see the CRISPR recognition sites for each other respectively (3rd one got cut off).
Then we simply right-click on the recognition site to ‘copy’. Then it will give you the sequence.
After you have derived the necessary guide(s) you can assemble a guide RNA (gRNA) in a wet lab. This is done by adding the recognition site to the gRNA and inserting the plasmid we created as the spacer DNA. Since we select single-stranded CRISPR, what the Cas9 protein would do is make a cut on one strand, and DNA polymerase will ‘fill’ in the complementary strand.
Conclusion
And that’s it. We have successfully created a plasmid capable of suppressing a tumour to remove several harmful side effects, and made it possible to enhance diagnostic capabilities immensely via the selection marker. Of course, this plasmid would only work in theory and additional testing/refinement is needed before human trials. But who knows maybe scientists will be using the muIFN-but-better plasmid to cure cancer.
Not only does this article demonstrate the incredible capabilities of Benchling but also that it’s possible for anymore with a laptop and access to the Internet to create the solution to the world’s largest problems!
Link to Benchling schematic → Benchling Model
