Let's talk about GMOs, part one: What techniques are used?
Introduction
In my previous blog post, I discussed what DNA is and how it relates to phenotypical change through to central dogma of molecular biology. In this post, we will discuss techniques currently in use.
The techniques I want to discuss are not only those that have been used in genetic engineering, such as those used commercially to engineer the Bt. crops. I also want to discuss those that are currently in use, particularly Crispr. The latter is a very recent technique that has revolutionised the way genetic engineering works.The goal is not to give an extensive review of techniques, but rather to give readers some insight into common techniques.
Producing Insulin
Synthetic biology doesn't make nice stories. The amount of different tools (enzymes) used and their abbreviated names obfuscate the essence of any story.I want to discuss a few specific tools.
Let's first discuss the bacterium; E. coli. This bacterium can be found in the lower intestine of most warm-blooded organisms and is a common model bacterium in biological research. It has a circular segment of DNA carrying its genetic information; this literally means a circle of DNA.
In 1972, researchers devised how to make recombinant DNA by using two tools (Wiki). The first tool was a restriction enzyme. Restriction enzymes are very specific enzymes that can cut DNA at a specific code (wiki). For instance, the Hind-iii restriction enzyme cuts DNA at the sequence 'AAGCTT' (wiki). The second tool is a DNA ligase (wiki), which is an enzyme that fuses two large pieces of DNA.
But that doesn't put the DNA inside a host cell's genetic material, such as E. coli. What is used for that is a bacteriophage. Bacteriophages are actually bacteria-viruses. As most viruses, they're quite simple - a protein 'vessel' that transports genetic material. A bacteriophage floats around until it comes near a bacterium. It then bonds to the wall of the bacterium, where it breaks down a part of it and empties into the bacterium - the host cell.
As I said, they're fairly simple. Simplicity is not always ideal, but leaves an opening for genetic engineers. Bacteriophages can be 'loaded' with DNA of another origin. In that way, a genetic engineer might make DNA that he or she wants to insert into E. coli - such as the DNA that produces Insulin. The method used by the researchers in 1972 can be combined with PCR (1983) to create plasmids of your choice. PCR is a technique that thermally separates double-stranded DNA into single-stranded DNA. You then add DNA Polymerase, which is an enzyme that copies DNA. This gives you double stranded DNA again, after which you start the next cycle by thermal separation. This chain of reproduction cycles is called the Polymerase Chain Reaction, or PCR. The following picture give an illustration. The difference with the picture is that when you use to create the DNA 'primer' (short piece) that you want is that the red building block actually extends over the edge of the green one. PCR used for DNA sequencing only adds the nucleotides (=single letters) so that it copies the DNA many times.
After you have the primer, you use the 1972 method to create a plasmid. Plasmids can be easily migrated into E. coli cells, which then have their own circular DNA floating around and the added plasmids. Some biophages will target the plasmid rather than the E. coli genetic material. Their 'goal' is that they switch their own biophage material with a piece of the E. coli genetic material. They do this by having a 'tail', which matches a part of the host material. Then follows the biophage material - instructions for making biophages - and finally another tail. Sometimes, it messes up - it inserts instructions to make more biophages with the DNA already in place, rather than the DNA of the biophage. The trick is to get the biophage to take the DNA primer we inserted.
You now have 'loaded' phages and 'wild' phages. A common way to make the distinction in the end was to use e.g. Insulin plus an antibiotic resistance gene. If you succeed, you can purge the 'wild' population with antibiotics.
The bacteriophages are added to a solution with E. coli, and they insert their DNA packages into the host cells. However, we want to have the 'insertion DNA' added to the genetic material of the host cell. Adding restriction enzymes, which break the DNA of the E. coli cells would be a good start. Adding a ligase would then randomly incorporate the 'insertion DNA' into the host genetic material. It is also possible to remove the antibiotic resistance gene. For this, you need to put code sequences on both side of the antibiotic code, called FRT regions. By then infecting the E. coli cell with a flippase, a plasmid of baker's yeast, you can remove the antibiotic gene. Flippase will sit between the two FRT sites, and twist - causing a loop. It then connects the two FRT regions - cutting off the remainder, which is antibiotic code in our example.
A lot of phages come with their own enzymes and ligases, which greatly simplifies the process. For E. coli, a common choice is the lambda phage (wiki).That is the basis of phage engineering. To summarise:
This succeeded in 1978 (wiki), after just six short years (with a different technique to make the primer). This alone is a good indicator of how well these techniques worked. And don't forget the advances in techniques such as DNA sequencing, which allows you to check whether you succeeded in your engineering. For that, you'd add enough food that your genetically-enhanced E. coli multiplied their numbers, then separate part of their population and sequence their DNA.
Let's take one step back. Earlier, I mentioned DNA tails on the phage DNA that had to match the host. Why? The keyword here is affinity. A lot of molecular biology and biochemistry turns into the question of whether it is likely to happen. For something to be likely, we need to know the energy of something not happening versus that of it happening. Technically, we also need the temperature. The required concept is called a Boltzmann factor, and is part of the foundation of the Atomic theory. Before you run off, that's just the theory of there being atoms. Using Boltzmann statistics, we can derive most of the thermodynamic quantities discovered during the industrial age. Boltzmann statistics work well on the smaller scales, too. You can predict the natural shape of a large protein, for instance, based on Boltzmann statistics. If you do that for DNA, you find that something is needed to store our DNA (~2m length) in our cells.
The tails are the complementary code of a region of code on the DNA. Because they are complimentary, they have a very high 'affinity' for that region - the energy of bonding is very, very low. Most principles in nature (most of the physics you were taught in high school!) ultimately boil down to minimisation of energy. So it is for chemistry - affinity means minimal energy, and therefore far more likely. The higher the affinity, the lower the energy of the combined parts, and the more likely it is for them to combine.
With a gene gun, you can take plant cells and deliver DNA-coated particles to them (wiki) . The DNA-coated particles are often just gold particles covered in plasmids - which we know you can design. This way, plasmids are delivered into a plant cell, a process that is called Transfection. Sometimes, these plasmids are incorporated into the host genetic material. This appears to be a lucky event rather than a targeted one. The same concept is used with plant cells as with E. coli; you use antibiotics or herbicide to kill the cells without the new code. Before you ask, antibiotics do kill other cells; it depends on the antibiotic. Some, for instance, just destroy cell walls - be they plant or otherwise. For a fully developed organism, this is unlikely and dosage is low. For a plant cell near a researcher trying to find the successful transfections (stable transfection), dosage is extremely high. Hence, antibacterial agents work.
Cetus corporation started out with automated breeding methods to select greater amounts of chemical feedstock, antibiotics and vaccine components. It entered the new fields of biotechnology. Later it merged with another company to create Agracetus, which created the first Roundup Ready crops using a gene gun. Five years later, this company was bought by Monsanto, creating the Monsanto Agracetus campus. That's right - Monsanto didn't first create that technology. And this wasn't a single corporation going into the fields of agriculture. The potential was realised, the race was on. To quote from Sir Terry Pratchett's works, it was steam engine time. Health organisations, such as the WHO, were already wondering about policy to evaluate these new, exciting products while not giving in to technological enthusiasm.
There are other methods of editing plants - such as using the Ti Plasmid, also called the natural genetic engineer. However, the above should give you some insight into how this is done.
After cleaving, it is possible to insert DNA.This is done by hijacking the repair pathways. In that way, the cleavage introduced by Cas9 can be filled up with the code that you wanted to insert. That story is slightly more specific, and I will leave it to the interested reader to search for it. There are two possibilities. The breaks introduced by Cas9 can be repaired by non-homologous end joining (NHEJ) or by homology-directed repair (HDR) pathways. See also the image to the right.
One other benefit of Crispr/Cas9 is that it can be done in vitro. That's pretty powerful, and can be used for all sorts of things. Let me put down a hopeful hypothesis. Gene-therapy for cancer. Take a sample from the tumor and from healthy tissue, and find code sequences that are different. Use those to target Cas9 to cancer cells, introducing a terminator gene. That's exactly what it sounds - it terminates the cell, by making it mess up its chemistry. Not only do you hit the tumor you sampled from, but also the smaller tumors it seeded. Can this become reality? Honestly, I don't know - my knowledge of the subject matter is broad, not specific. However, I do hope this can become reality!
We've now discussed what DNA is and how we can change it. In the next post, I'll try to give you an overview of current uses of genetic engineering.
Let's first discuss the bacterium; E. coli. This bacterium can be found in the lower intestine of most warm-blooded organisms and is a common model bacterium in biological research. It has a circular segment of DNA carrying its genetic information; this literally means a circle of DNA.
 |
| Typical bacteriophage (wiki). |
But that doesn't put the DNA inside a host cell's genetic material, such as E. coli. What is used for that is a bacteriophage. Bacteriophages are actually bacteria-viruses. As most viruses, they're quite simple - a protein 'vessel' that transports genetic material. A bacteriophage floats around until it comes near a bacterium. It then bonds to the wall of the bacterium, where it breaks down a part of it and empties into the bacterium - the host cell.
As I said, they're fairly simple. Simplicity is not always ideal, but leaves an opening for genetic engineers. Bacteriophages can be 'loaded' with DNA of another origin. In that way, a genetic engineer might make DNA that he or she wants to insert into E. coli - such as the DNA that produces Insulin. The method used by the researchers in 1972 can be combined with PCR (1983) to create plasmids of your choice. PCR is a technique that thermally separates double-stranded DNA into single-stranded DNA. You then add DNA Polymerase, which is an enzyme that copies DNA. This gives you double stranded DNA again, after which you start the next cycle by thermal separation. This chain of reproduction cycles is called the Polymerase Chain Reaction, or PCR. The following picture give an illustration. The difference with the picture is that when you use to create the DNA 'primer' (short piece) that you want is that the red building block actually extends over the edge of the green one. PCR used for DNA sequencing only adds the nucleotides (=single letters) so that it copies the DNA many times.
 |
| PCR reaction.The red primer (and blue complementary chain) will outnumber the green strands in a few cycles. As the added nucleotides (letters) are just molecules, the product of the PCR reaction can truly be called artificial. It's not (e.g.) plant DNA; it is a sequence equal to that present in plants. |
After you have the primer, you use the 1972 method to create a plasmid. Plasmids can be easily migrated into E. coli cells, which then have their own circular DNA floating around and the added plasmids. Some biophages will target the plasmid rather than the E. coli genetic material. Their 'goal' is that they switch their own biophage material with a piece of the E. coli genetic material. They do this by having a 'tail', which matches a part of the host material. Then follows the biophage material - instructions for making biophages - and finally another tail. Sometimes, it messes up - it inserts instructions to make more biophages with the DNA already in place, rather than the DNA of the biophage. The trick is to get the biophage to take the DNA primer we inserted.
You now have 'loaded' phages and 'wild' phages. A common way to make the distinction in the end was to use e.g. Insulin plus an antibiotic resistance gene. If you succeed, you can purge the 'wild' population with antibiotics.
The bacteriophages are added to a solution with E. coli, and they insert their DNA packages into the host cells. However, we want to have the 'insertion DNA' added to the genetic material of the host cell. Adding restriction enzymes, which break the DNA of the E. coli cells would be a good start. Adding a ligase would then randomly incorporate the 'insertion DNA' into the host genetic material. It is also possible to remove the antibiotic resistance gene. For this, you need to put code sequences on both side of the antibiotic code, called FRT regions. By then infecting the E. coli cell with a flippase, a plasmid of baker's yeast, you can remove the antibiotic gene. Flippase will sit between the two FRT sites, and twist - causing a loop. It then connects the two FRT regions - cutting off the remainder, which is antibiotic code in our example.
A lot of phages come with their own enzymes and ligases, which greatly simplifies the process. For E. coli, a common choice is the lambda phage (wiki).That is the basis of phage engineering. To summarise:
- Use synthetic biology (PCR) to create the DNA segment you want to insert. Add the code for resistance to an (old) antibiotic and FRT regions around it. Turn it into a plasmid.
- Infect E. coli cells with your plasmid, then load phages. The phages multiply, and some get loaded with material from your plasmid.
- Add the phages to a population of E. coli bacteria.
- Purge the E. coli so that only those that have the segment remain.
- Remove the antibiotic gene.
- Feed them, nourish them and farm the Insulin they produce.
This succeeded in 1978 (wiki), after just six short years (with a different technique to make the primer). This alone is a good indicator of how well these techniques worked. And don't forget the advances in techniques such as DNA sequencing, which allows you to check whether you succeeded in your engineering. For that, you'd add enough food that your genetically-enhanced E. coli multiplied their numbers, then separate part of their population and sequence their DNA.
Improving the quality, quantity or availability of food.
Genetically engineering E. coli and yeast is all good, but can we do something else with it? Can we change the DNA of a plant so that it can withstand bugs, droughts, colds, rains and heat flares? That is one of many questions one can consider when hearing of genetic engineering.Let's take one step back. Earlier, I mentioned DNA tails on the phage DNA that had to match the host. Why? The keyword here is affinity. A lot of molecular biology and biochemistry turns into the question of whether it is likely to happen. For something to be likely, we need to know the energy of something not happening versus that of it happening. Technically, we also need the temperature. The required concept is called a Boltzmann factor, and is part of the foundation of the Atomic theory. Before you run off, that's just the theory of there being atoms. Using Boltzmann statistics, we can derive most of the thermodynamic quantities discovered during the industrial age. Boltzmann statistics work well on the smaller scales, too. You can predict the natural shape of a large protein, for instance, based on Boltzmann statistics. If you do that for DNA, you find that something is needed to store our DNA (~2m length) in our cells.
The tails are the complementary code of a region of code on the DNA. Because they are complimentary, they have a very high 'affinity' for that region - the energy of bonding is very, very low. Most principles in nature (most of the physics you were taught in high school!) ultimately boil down to minimisation of energy. So it is for chemistry - affinity means minimal energy, and therefore far more likely. The higher the affinity, the lower the energy of the combined parts, and the more likely it is for them to combine.
With a gene gun, you can take plant cells and deliver DNA-coated particles to them (wiki) . The DNA-coated particles are often just gold particles covered in plasmids - which we know you can design. This way, plasmids are delivered into a plant cell, a process that is called Transfection. Sometimes, these plasmids are incorporated into the host genetic material. This appears to be a lucky event rather than a targeted one. The same concept is used with plant cells as with E. coli; you use antibiotics or herbicide to kill the cells without the new code. Before you ask, antibiotics do kill other cells; it depends on the antibiotic. Some, for instance, just destroy cell walls - be they plant or otherwise. For a fully developed organism, this is unlikely and dosage is low. For a plant cell near a researcher trying to find the successful transfections (stable transfection), dosage is extremely high. Hence, antibacterial agents work.
Cetus corporation started out with automated breeding methods to select greater amounts of chemical feedstock, antibiotics and vaccine components. It entered the new fields of biotechnology. Later it merged with another company to create Agracetus, which created the first Roundup Ready crops using a gene gun. Five years later, this company was bought by Monsanto, creating the Monsanto Agracetus campus. That's right - Monsanto didn't first create that technology. And this wasn't a single corporation going into the fields of agriculture. The potential was realised, the race was on. To quote from Sir Terry Pratchett's works, it was steam engine time. Health organisations, such as the WHO, were already wondering about policy to evaluate these new, exciting products while not giving in to technological enthusiasm.
There are other methods of editing plants - such as using the Ti Plasmid, also called the natural genetic engineer. However, the above should give you some insight into how this is done.
Crispr/Cas9
 |
| Illustration of Crispr/Cas9. See text for explanation. Source: Nature |
This is the new technique. The novel method that is revolutionising Academia - I can't tell you how often the abbreviation features on posters around the biophysics building (my workplace). The variety and enthusiasm with which people have grabbed onto this new technique since its introduction around 2012 is astonishing. In theoretical physics, a Nobel Prize was just given away for a field founded by a paper that was forgotten and ignored for thirty years or more (topological insulators). Not so for biologists - they grabbed onto Crispr/Cas and immediately started working with it.
To explain the system, I will use some images from a Nature Biotechnology review. The illustration of Crispr/Cas9 is quite clear. Cas9 is a nuclease, a sort of molecular scissor. Cas9 is very particular; you give it a piece of Guide RNA (remember, the letters match with DNA - the backbone is slightly different). This way, the Cas9 only cuts at that particular code, called the specificity of the nuclease. In general, it has very high specificity, meaning that it only seldom (less than 1%) cuts at the wrong location. In the next image, a slightly more complicated systematic illustration is shown. In this figure, we have naturally occuring (a) Crispr systems that incorporate foreign DNA. This is the natural defence system of numerous bacteria - cutting up infections. In (b) we find the engineering approach, where the researcher provides the Guide RNA and the Cas9 enzyme cleaves (cuts) at that code. In (c) we see the natural guide RNA (top) and the engineered guide RNA (gRNA). As you can see, these are not that different.
To explain the system, I will use some images from a Nature Biotechnology review. The illustration of Crispr/Cas9 is quite clear. Cas9 is a nuclease, a sort of molecular scissor. Cas9 is very particular; you give it a piece of Guide RNA (remember, the letters match with DNA - the backbone is slightly different). This way, the Cas9 only cuts at that particular code, called the specificity of the nuclease. In general, it has very high specificity, meaning that it only seldom (less than 1%) cuts at the wrong location. In the next image, a slightly more complicated systematic illustration is shown. In this figure, we have naturally occuring (a) Crispr systems that incorporate foreign DNA. This is the natural defence system of numerous bacteria - cutting up infections. In (b) we find the engineering approach, where the researcher provides the Guide RNA and the Cas9 enzyme cleaves (cuts) at that code. In (c) we see the natural guide RNA (top) and the engineered guide RNA (gRNA). As you can see, these are not that different.
After cleaving, it is possible to insert DNA.This is done by hijacking the repair pathways. In that way, the cleavage introduced by Cas9 can be filled up with the code that you wanted to insert. That story is slightly more specific, and I will leave it to the interested reader to search for it. There are two possibilities. The breaks introduced by Cas9 can be repaired by non-homologous end joining (NHEJ) or by homology-directed repair (HDR) pathways. See also the image to the right.
 |
| Repair of cleaved DNA with code supplied by researcher (blue). |
Another interesting application of Cas9 is inactivation. There's such a thing as dead Cas9, which has been modified so it can't actually cut. It will still move toward the targeted code, so it can block other molecular machines from sitting there. In this way, sites can be deactivated and the effect of that can be studied.
One other benefit of Crispr/Cas9 is that it can be done in vitro. That's pretty powerful, and can be used for all sorts of things. Let me put down a hopeful hypothesis. Gene-therapy for cancer. Take a sample from the tumor and from healthy tissue, and find code sequences that are different. Use those to target Cas9 to cancer cells, introducing a terminator gene. That's exactly what it sounds - it terminates the cell, by making it mess up its chemistry. Not only do you hit the tumor you sampled from, but also the smaller tumors it seeded. Can this become reality? Honestly, I don't know - my knowledge of the subject matter is broad, not specific. However, I do hope this can become reality!
Conclusion
Genetic engineering turned out to be a set of powerful techniques. The different techniques are different in simplicity of use, of concept and of specificity. But they are quite specific, and we have the technology to compare wild (unedited) DNA to target (edited) DNA to see if we succeeded. And we usually do.
Crispr/Cas9 is mostly making the rounds because of simplicity, broad applicability (many organisms) and because it also has extra options. It's also likely to get a Nobel prize, likely shared between a large number of people. And that's not only the prize for biology, but also that for medicine. And because food issues can lie at the root of conflict (war), it might even be the one for peace.
Yes, the technique has me excited. It's even more specific than the previous techniques, it has a high success rate, a high specificity, a broad applicability and so much more. Honestly, I can't wait to see what this will turn up. And by that I mean not only in commerce, but also in medicine, in agriculture, in helping developing agriculture, in understanding of nature and things I can't even imagine.
Crispr/Cas9 is mostly making the rounds because of simplicity, broad applicability (many organisms) and because it also has extra options. It's also likely to get a Nobel prize, likely shared between a large number of people. And that's not only the prize for biology, but also that for medicine. And because food issues can lie at the root of conflict (war), it might even be the one for peace.
Yes, the technique has me excited. It's even more specific than the previous techniques, it has a high success rate, a high specificity, a broad applicability and so much more. Honestly, I can't wait to see what this will turn up. And by that I mean not only in commerce, but also in medicine, in agriculture, in helping developing agriculture, in understanding of nature and things I can't even imagine.
We've now discussed what DNA is and how we can change it. In the next post, I'll try to give you an overview of current uses of genetic engineering.
