Biofuels’ Saviour: Genetically Modified Algae

In 2019, I began my quest on how to create more efficient biofuels; I came across a research paper that made my dreams possible.

Okay, let’s be real here. Biofuels aren’t exactly the saviour we all thought it was going to be. It has limitations, and at this point in time, it’s way cheaper to continue to use traditional forms of energy (ie., fossil fuels) than it is to use biofuels.

A large-scale algae farm (Source: Symbiosis In Development)

But humour me for a few moments, if you will. Let’s imagine what the world would look like if we were to use biofuels as our primary source of energy instead of fossil fuels.

First, we would need a source of biomass to turn into biofuels. Enter stage right: algae! Algae is a great candidate because:

1. It doesn’t take that much space to grow;
2. It’s really efficient at absorbing carbon dioxide;
3. It produces a lot of lipids (hint: that the component that we need to make biofuels)

“So, great! We have a candidate for biofuels! Let’s start using it now — “

I’m going to have to pull you back to reality there. One of the major limitations of algae fuels is that with our current technology, it is financially unsustainable to produce algae fuels for lots of consumers.

In other words, it’s still too expensive to make a lot of it.

But, we humans are stubborn creatures. What do we do if we can’t get what we want? We bend the rules a little. So, let’s bend the rules of nature…just a little.

Researchers found that when a specific species of algae called Nannochloropsis gaditana grows in a nitrogen starved environment, it produces twice as much lipid as its wild type, which is pretty wild (bu dum tisssss). What this tells them is that there is some sort of connection between how much lipid is made and the environment it grows in.

There’s one issue though: growing large amounts of algae in a nitrogen starved environment is immensely difficult, not to mention expensive. Nitrogen is the most abundant gas in out atmosphere, so getting rid of it a large scale is virtually impossible.

So, what’s our next option? Well, in the spirit of the twenty-first century, the golden age of tech, we could hack it! Hack its DNA, to be more precise.

(Spoiler: this is what the researchers did!)

We know based on the algae that was grown in the N-starved environment that there is obviously a genetic factor (aka, a gene), that controls how much lipids are being produced. Researchers found that if they used CRISPR-Cas9, currently the most popular gene editing technique, to knock-out the genes that regulate lipid production, the algae would produce twice as many lipids in their regular environment.

(if you know nothing about CRISPR, don’t worry! I made this 5-minute video on the basics.)

In the spirit of excitement, I desperately wanted to see this in action. Albeit, I did not have the money to conduct this experiment and it’s the dead of winter where I live (January’s in Canada, am I right?), so it was very unlikely I could physically replicate this at home.

So, I did the next best thing.

CRISPR + algae = a whole lotta LIPIDS!

I had worked with an online-based simulator before called Benchling to help me simulate another one of my projects. (To those who are interested, I created a new and lab-accurate E. coli detection kit with CRISPR!) I thought Benchling would be the perfect candidate to recreate the research on. With the original research paper as a guide, I created a plan:

1. Locate the lipid regulating genes on the algae genome;
2. Locate the target DNA (or which specific genes I wanted to knock out);
3. Find a suitable PAM (the site where the Cas9 protein will cleave);
4. Design the gRNA (the guide RNA that will guide Cas9 to the target site)

Full disclosure! I’m about to get technical, so if you’d like a simplified overview of my process, skip to the next section :)

According to the research paper, the genes which encode ZnCys, specifically Zn(II)2Cys6 were the ones responsible for double the production of lipids. There are 20 genes total, but scientists found that knocking out 18 was enough for lipid production to double.

I’ll be completely honest, locating the genes was no easy task.

This is the genome for nannochloropsis gaditana. It was hundreds of thousands of base pairs long, and needless to say I was overwhelmed.

After uploading the algae genome to Benchling, it took me several days to find where exactly the ZnCys gene was. Luckily, I was able to find a database with the locations for all the genes I was looking for.

The next challenge was finding where the genes were on the actual algae genomes. Unfortunately, I did not have access to the genome browser that the researchers had access to (unfortunately, the site [link] has been down the weeks I have been trying to replicate this experiment), so I embarked on the tedious task of cross-referencing each and every ZnCys gene myself until I found a match on the algae genome.

Unfortunately, I was not able to locate the specific ZnCys genes directly. However, I was able to locate the nuclear localization sequence (NLS). The NLS is a short sequence of base pairs after the gene that will help guide the encoded protein nucleus. Because I know that the ZnCys is about 2.2 kilobases long and the NLS directly follows it, I deduced the approximate location of a ZnCys gene. Finally, I had my target DNA.

Using the nuclear localiztion sequence as a guide, I was able to make a rough estimate of where the ZnCys gene was located and identified my target DNA.

Next, I needed to find a suitable protospacer adjacent motif (PAM). The PAM is often a very short sequence after the target DNA, but varies depending on which bacterium the Cas9 protein is taken from. One of the most common Cas9 proteins taken from the bacteria S. pyogenes, where ’N’ is any nucleotide followed by two guanine (G) bases.

The PAM is quite a few base pairs downstream of the target DNA, but this is the first instance where the Cas9 protein would be able to identify the cleave site.

In my case, the PAM would be downstream from the target DNA and would have the sequence ‘AATTGG’. Notice how the final two base pairs are both G? That’s how Cas9 knows where to cut.

Finally, with all the steps in the necessary components, I could finally design my gRNA. As a little refresher to how DNA bonds, base pairs can only bond with their complementary bases. In other words, A can only bond with T and C can only bond with G. So in order for the gRNA to bond onto the target DNA, it must be the inverse of the target DNA (running from 3’ to 5’).

The gRNA encompasses the entirety of the target DNA just until the PAM.

Note that the gRNA should not bond to the PAM, because if the gRNA was bonded to the PAM, Cas9 would be unable to make the required cleave.

Et voila…kinda. This is what the experiment would look like with one ZnCys gene. According to the researchers, in order for the algae to produce double the amount of lipids, 18 out of the 20 ZnCys genes would need to be knocked out. However, if there comes a day where I will have more expansive resources and I could legally replicate this experiment (there’s a pending patent for this technology, so to avoid being sued I’m not going to copy it in its entirety), I will definitely retry this replication.

So, once and for all… how do you knock out algae genes?

I know I said there will be a simple explanation, but this process was very far from simple. Still, I’ll explain it in the easiest way possible:

1. Gather all your required genomes from online databases. You can’t simulate this experiment unless you have the actual genetic code of each of the actors. This includes the genomes of the algae (specifically, Nannochloropsis gaditana), the ZnCys gene you are choosing to knock out, and the PAM).
2. Using an online simulator, open the algae genome. You are going to be harassed with hundreds of thousands of A’s, T’s, C’s, and G’s. Don’t be alarmed. This is the genetic code that makes algae.
3. Search the algae genome for the ZnCys gene you plan on knocking out. Highlight and annotate this gene.
4. Locate the PAM. Depending on what bacteria the Cas9 you plan to use comes from, it may very. A common Cas9 protein comes from S. pyogenes, where the PAM looks like ‘NGG’.
5. Finally, make your gRNA. Do this by selecting the genetic code that encompasses the target DNA and stops right before the PAM.

And there you have it! All the necessary components to knocking out a single ZnCys gene. Of course, we would need to repeat this process 17 other times with 17 other ZnCys genes, but the process remains valid.

A big step for biofuels

Remember how I said that producing biofuels is not financially sustainable, especially when compared to the rate we are currently consuming fossil fuels? This research may not the be-all, end-all solution, but this is a step in the right direction.

I spoke about the potential that genetically modified algae biofuels could have at the December 2019 IBM Townhall.

While I was working on this project, I came across a lot of articles and people who had given up on biofuels and just ignored out impending climate change doom.

That only fueled me to do this project more; I wanted to prove that I wasn’t over for biofuels and for renewable resources as a replacement for biofuels. Innovations like these — the product of combining two seemingly different fields — are the reason big steps towards a better future could be made in the first place.

Hi! My name’s Eliza and I’m a 16-year-old Innovator and The Knowledge Society (TKS). I’m super passionate about gene editing and global sustainability. I’m currently doing research on how we can use gene editing techniques to solve climate change. Connect with me on Linkedin or subscribe to my monthly newsletter!