1
00:00:00,000 --> 00:00:03,000
scientific research is a strange endeavor.
2
00:00:03,000 --> 00:00:06,000
You spend years working on a project.
3
00:00:06,000 --> 00:00:09,000
You carry out experiment after experiment.
4
00:00:09,000 --> 00:00:12,000
You collect data, you double check all your calculations,
5
00:00:12,000 --> 00:00:15,000
and then package it all up with an introduction and a discussion
6
00:00:15,000 --> 00:00:18,000
into a product called a research paper.
7
00:00:18,000 --> 00:00:20,000
And then you just throw that out into the world,
8
00:00:20,000 --> 00:00:23,000
actively encouraging anyone around the whole world to read it,
9
00:00:23,000 --> 00:00:28,000
to make what they will of it, and to actively try to prove you wrong.
10
00:00:28,000 --> 00:00:32,000
Sometimes, those people go on to become your collaborators.
11
00:00:32,000 --> 00:00:35,000
And sometimes you never ever meet them.
12
00:00:35,000 --> 00:00:41,000
Greg Amendel published his "Vesucher Ubefranzenhebrudin" in 1866,
13
00:00:41,000 --> 00:00:46,000
his most famous publication, and he died in 1884.
14
00:00:46,000 --> 00:00:49,000
It wasn't until almost two decades after his death
15
00:00:49,000 --> 00:00:52,000
that other scientists around Europe and the world began to read his papers
16
00:00:52,000 --> 00:00:56,000
and pick up where he left off to create the field of genetics.
17
00:00:56,000 --> 00:00:58,000
That's not really so unusual.
18
00:00:58,000 --> 00:01:01,000
Research is a process that does not always run smoothly.
19
00:01:01,000 --> 00:01:04,000
Sometimes it can take years for an idea to bounce around
20
00:01:04,000 --> 00:01:07,000
between the right group of people until at some point,
21
00:01:07,000 --> 00:01:09,000
quite to the surprise of everyone involved,
22
00:01:09,000 --> 00:01:14,000
it turns into something new, something shocking that changes the world.
23
00:01:14,000 --> 00:01:17,000
And so it was, in 1987,
24
00:01:17,000 --> 00:01:20,000
a PhD student called Yoshizumi Ishimo,
25
00:01:20,000 --> 00:01:23,000
working at the University of Osaka,
26
00:01:23,000 --> 00:01:27,000
published a paper and sent it off into the world.
27
00:01:27,000 --> 00:01:32,000
The paper was titled "Nuclear Tide Sequence of the IAP gene,
28
00:01:32,000 --> 00:01:35,000
responsible for alkaline phosphatase, isoenzyme conversion,
29
00:01:35,000 --> 00:01:40,000
and escherichia coli, and identification of the gene product."
30
00:01:40,000 --> 00:01:43,000
Not exactly a title that would make you think
31
00:01:43,000 --> 00:01:46,000
a revolution lies in this paper.
32
00:01:46,000 --> 00:01:50,000
Alkaline phosphatases are important in maintaining the phosphate balance of the cell,
33
00:01:50,000 --> 00:01:54,000
and they were the main focus of the research, which is essentially to sequence a gene
34
00:01:54,000 --> 00:01:58,000
encoding an enzyme involved in alkaline phosphatase production in the bacterium
35
00:01:58,000 --> 00:02:02,000
escherichia coli, more commonly known as e-coli.
36
00:02:02,000 --> 00:02:04,000
But in the process of sequencing that gene,
37
00:02:04,000 --> 00:02:06,000
and I will emphasize in the 1980s,
38
00:02:06,000 --> 00:02:10,000
sequencing a gene was a serious undertaking worthy of a PhD.
39
00:02:10,000 --> 00:02:12,000
During the process of sequencing that gene,
40
00:02:12,000 --> 00:02:16,000
Dr Ishimo found something else, something unexpected,
41
00:02:16,000 --> 00:02:19,000
just after the gene, physically in that genome,
42
00:02:19,000 --> 00:02:22,000
he found something that was not what he was interested in,
43
00:02:22,000 --> 00:02:25,000
but he did describe it in the final part of the paper.
44
00:02:25,000 --> 00:02:30,000
He wrote, "An unusual structure was found in the three-prime end,
45
00:02:30,000 --> 00:02:33,000
flanking region of the gene of interest.
46
00:02:33,000 --> 00:02:39,000
Five highly homologous sequences of 29 nuclear tides were arranged as direct repeats
47
00:02:39,000 --> 00:02:43,000
with 32 nuclear tides as spacing.
48
00:02:43,000 --> 00:02:48,000
What he had shared with the world was the first description of an intriguing phenomenon,
49
00:02:48,000 --> 00:02:53,000
that will get picked up and explored by scientists around the globe.
50
00:02:53,000 --> 00:02:57,000
These repeating sequences seemed to be found in most bacteria
51
00:02:57,000 --> 00:03:03,000
and between the repeating units were little pieces of DNA that matched known viruses.
52
00:03:03,000 --> 00:03:07,000
Digging deeper research has found that these repeating DNA elements are managed
53
00:03:07,000 --> 00:03:11,000
and created by a set of bacterial proteins.
54
00:03:11,000 --> 00:03:15,000
The repeating DNA units came to be known as clustered,
55
00:03:15,000 --> 00:03:19,000
but regularly interspaced short palindromic units.
56
00:03:19,000 --> 00:03:25,000
The proteins that managed and and created these clusters were clustered associated proteins or cast proteins.
57
00:03:25,000 --> 00:03:28,000
They turned out to be many of these cast proteins.
58
00:03:28,000 --> 00:03:30,000
They were given numbers as they were discovered,
59
00:03:30,000 --> 00:03:33,000
cast one, cast two, cast three, and so on.
60
00:03:33,000 --> 00:03:39,000
By 2020, we were decades into the existence of this fairly niche field of research
61
00:03:39,000 --> 00:03:43,000
interested by then in what had become to understood as a bacterial immune system,
62
00:03:43,000 --> 00:03:48,000
protecting the cell against viruses by integrating small chunks of viral DNA
63
00:03:48,000 --> 00:03:54,000
into basically a memory bank for the cell to know what to look out for and destroy in future.
64
00:03:54,000 --> 00:04:01,000
The next year in 2011, a small side session was held to discuss these bacterial immune systems
65
00:04:01,000 --> 00:04:06,000
as part of a bigger microbiology conference in San Juan, Puerto Rico.
66
00:04:06,000 --> 00:04:11,000
And at the sidelines of that session, as fringe as could be the side of a side session,
67
00:04:11,000 --> 00:04:14,000
something happened that would change the world.
68
00:04:14,000 --> 00:04:18,000
A French professor in Germany called a manual of Sheponcier
69
00:04:18,000 --> 00:04:25,000
approached an American professor who had grown up in Hilo-Hawaii but worked at UC Berkeley, Jennifer Doudna.
70
00:04:25,000 --> 00:04:31,000
Sheponcier felt that there are two research teams between them had just the right mix of expertise
71
00:04:31,000 --> 00:04:36,000
to tackle a project on her mind, solving the function of a particular protein
72
00:04:36,000 --> 00:04:43,000
that was associated with these clustered, regularly interspersed short palindromic repriets
73
00:04:43,000 --> 00:04:46,000
or CRISPR as they're known.
74
00:04:46,000 --> 00:04:50,000
This protein, Cas9, was little known at that point,
75
00:04:50,000 --> 00:04:55,000
but in just nine years, Doudna and Sheponcier would be accepting a prize awarded
76
00:04:55,000 --> 00:05:00,000
based on one of the easiest decisions of the Nobel Committee for their work to solve the function
77
00:05:00,000 --> 00:05:09,000
of CRISPR Cas9 and ushering in a new age of unlimited possibility to engineer genomes as we please.
78
00:05:09,000 --> 00:05:11,000
Welcome to Modified.
79
00:05:11,000 --> 00:05:16,000
I'm your host, Dr. O'Landedalange, Explants Scientist and Current Science Teacher.
80
00:05:16,000 --> 00:05:20,000
This podcast is my attempt to guide you through the science of GMOs
81
00:05:20,000 --> 00:05:25,000
and hopefully just enough of the history and the social context to give you a path forward in your own thinking.
82
00:05:25,000 --> 00:05:33,000
And if all else fails, I hope you'll enjoy some of the surprising stories and fascinating facts in the world of genetically modified plants.
83
00:05:33,000 --> 00:05:35,000
This is episode 16.
84
00:05:35,000 --> 00:05:40,000
CRISPR changes the world and unlocks GMO 2.0.
85
00:05:40,000 --> 00:05:45,000
If this is the first episode you're listening to, because maybe you saw the word CRISPR and thought,
86
00:05:45,000 --> 00:05:49,000
"Hey, I've heard that, I want to check that out." Well, welcome! Happy to have you here.
87
00:05:49,000 --> 00:05:53,000
You may find that I dive pretty deep into genetics and into the world of GMOs in this episode
88
00:05:53,000 --> 00:05:58,000
and so you may want to go back to episode 1 and enjoy the journey to get here,
89
00:05:58,000 --> 00:06:03,000
because of this is the last episode of season 1 of Modified.
90
00:06:03,000 --> 00:06:08,000
In this episode, I'll explain what CRISPR Cas9 actually is and how it works,
91
00:06:08,000 --> 00:06:16,000
and I'll explain what I mean by GMO 2.0 and why I don't think it's silly to predict a wave of CRISPR derived GMOs in the next five years.
92
00:06:16,000 --> 00:06:21,000
But maybe a good way to tee off the main body of this episode is to hear the words of a Manuarsha Pontea herself,
93
00:06:21,000 --> 00:06:26,000
interviewed in 2022. I will spare you me trying to do a French accent.
94
00:06:26,000 --> 00:06:31,000
She was asked, "What is the greatest benefit to humankind of your research?"
95
00:06:31,000 --> 00:06:35,000
And she had this to say, after first explaining possible applications in medicine.
96
00:06:35,000 --> 00:06:40,000
There are also great implications in plant biology in the production of plant crops,
97
00:06:40,000 --> 00:06:44,000
specifically in the world of global warming and global climate change,
98
00:06:44,000 --> 00:06:48,000
the necessity to be able to engineer crops that can sustain the changes
99
00:06:48,000 --> 00:06:53,000
that are meant to be encountered in the very near future and even in the long term future.
100
00:06:53,000 --> 00:06:58,000
In the life of today, we benefit so much of great well-being thanks to a number of technologies
101
00:06:58,000 --> 00:07:03,000
like improvement of food and medicine. CRISPR has a great impact in this.
102
00:07:03,000 --> 00:07:06,000
That's why it has impacts on humankind.
103
00:07:06,000 --> 00:07:12,000
So CRISPR from the get-go was recognized for its possible applications in crop biotechnology
104
00:07:12,000 --> 00:07:15,000
and I can back that up with some personal experience.
105
00:07:15,000 --> 00:07:22,000
So let's talk a little bit about me. In 2010, I started my PhD at the Ludwig-Mexamillian University of Munich
106
00:07:22,000 --> 00:07:29,000
and the goal handed to me for my first project was test the idea of creating a programmable DNA endonuclease
107
00:07:29,000 --> 00:07:34,000
using the DNA binding domains of the bacterial proteins that our group specialized on.
108
00:07:34,000 --> 00:07:39,000
Those proteins are called tails, T-A-L-E, it's an acronym.
109
00:07:39,000 --> 00:07:43,000
And so we decided to add an N to the end of the acronym to represent the word in nuclease
110
00:07:43,000 --> 00:07:50,000
and then you have T-A-L-E-N, talons. Google Talons and you'll still find plenty of material about them.
111
00:07:50,000 --> 00:07:56,000
Talent technology got so much interest because the idea of a programmable endonuclease was an already incredibly hot idea
112
00:07:56,000 --> 00:08:02,000
in the world of biotechnology and so a lot of people were desperate to see if these new talons could finally deliver
113
00:08:02,000 --> 00:08:08,000
would have been tried since the 1990s with only limited success using different proteins called zinc fingers.
114
00:08:08,000 --> 00:08:14,000
But let me back up, what is a programmable endonuclease that is through that word out there?
115
00:08:14,000 --> 00:08:19,000
An endonuclease is a protein that can cut DNA in the middle of a DNA molecule.
116
00:08:19,000 --> 00:08:24,000
Nuclease is a more general word for any protein that can cut DNA.
117
00:08:24,000 --> 00:08:32,000
You may remember that DNA is an acronym for deoxyribonucleic acid, so nuclease references that nucleic acid part.
118
00:08:32,000 --> 00:08:39,000
The hope is that a programmable endonuclease could allow you to make precision cuts in the DNA of any genome
119
00:08:39,000 --> 00:08:45,000
and that is the first step in then permanently changing the sequence of DNA bases of that genome position.
120
00:08:45,000 --> 00:08:49,000
A process known as gene editing or genome editing.
121
00:08:49,000 --> 00:08:52,000
There are many, many nucleases in the world.
122
00:08:52,000 --> 00:09:00,000
Some just chop up all DNA into pieces randomly. That can still be useful for some sorts of goals, both natural and technological.
123
00:09:00,000 --> 00:09:06,000
And there are other nucleases that are called restriction endonucleases or just restriction enzymes.
124
00:09:06,000 --> 00:09:13,000
They were the ones that drove the birth of recombinant DNA technology in the race to create EMOs back in the 1970s.
125
00:09:13,000 --> 00:09:16,000
They are still very commonly used but they are not programmable.
126
00:09:16,000 --> 00:09:23,000
Each restriction enzyme has a single built-in target DNA sequence that it binds and then cuts of that position.
127
00:09:23,000 --> 00:09:27,000
Generally those target sequences are about six nucleotides long.
128
00:09:27,000 --> 00:09:36,000
Any given six nucleotide sequence would, by pure chance, appear thousands of times in any large complex genome like a plant or animal genome.
129
00:09:36,000 --> 00:09:42,000
So restriction enzymes are not a useful tool if your goal is to try and engineer genomes.
130
00:09:42,000 --> 00:09:51,000
So as the 70s became the 80s and then the 90s, more and better information became available about the DNA sequences of all sorts of interesting genes.
131
00:09:51,000 --> 00:09:58,000
And a lot of that interest was focused on human genes that may play a health role, perhaps mutated versions cause no diseases.
132
00:09:58,000 --> 00:10:02,000
But if you could repair that mutation you could cure the disease.
133
00:10:02,000 --> 00:10:07,000
And there was lots of interest in plant genes that could be tweaked to create improved crop varieties.
134
00:10:07,000 --> 00:10:14,000
So there was a real desire to have some way to go in and do genome engineering.
135
00:10:14,000 --> 00:10:20,000
But this knowledge was somewhat useful because no tools were available that could actually facilitate this.
136
00:10:20,000 --> 00:10:27,000
So for those decades, you know, the 80s and 90s, a lot of money was poured into making use of the tools we did have.
137
00:10:27,000 --> 00:10:35,000
Mainly, namely to piece together engineer DNA molecules in a lab and then randomly insert that DNA somewhere into a target genome.
138
00:10:35,000 --> 00:10:37,000
That is the process we call transgenics.
139
00:10:37,000 --> 00:10:43,000
That is the process used to make all the GMO crops covered so far in this series.
140
00:10:43,000 --> 00:10:52,000
And all of the GMO crops that are currently commercially available, at least outside Japan, to foreshadow the very end of this episode.
141
00:10:52,000 --> 00:10:56,000
But that whole time, so we're talking here about the 80s basically through to the 2010s.
142
00:10:56,000 --> 00:11:00,000
While funding was pouring into academic research to study and use transgenics,
143
00:11:00,000 --> 00:11:05,000
and while investors were happily funding companies to develop products using transgenics,
144
00:11:05,000 --> 00:11:11,000
the dream of a programmable DNA endonuclease and therefore of genome engineering did not die.
145
00:11:11,000 --> 00:11:18,000
And that's why when a new type of DNA binding proteins called tails were first described in a scientific paper in 2009,
146
00:11:18,000 --> 00:11:20,000
a lot of research groups took notice.
147
00:11:20,000 --> 00:11:30,000
I will spare you a lengthy treatise in this podcast, but out of pure vanity, I will link in the show notes a review article about tails, talons, and their uses in plant biotechnology.
148
00:11:30,000 --> 00:11:39,000
And how at the time we tried to spell out the possible advantages the talons still had versus the nascent CRISPR technology since we wrote that article in 2013.
149
00:11:39,000 --> 00:11:42,000
Needless to say, CRISPR won that argument.
150
00:11:42,000 --> 00:11:48,000
The cliff notes are that tails are large proteins, and most of their bulk is a set of repeating units,
151
00:11:48,000 --> 00:11:56,000
and each of those repeating units binds to one DNA base, and there is a single amino acid that differs between each of the repeating units,
152
00:11:56,000 --> 00:12:07,000
and that gives a specific DNA binding preference to that repeat using a specific code that links that amino acid, that variable amino acid, to a specific DNA base.
153
00:12:07,000 --> 00:12:13,000
So that variable amino acid is an iso-lucine, the repeat will bind to an A in the DNA.
154
00:12:13,000 --> 00:12:19,000
If it's a glycine, it will bind to a T. If it's an aspartate, then it will bind to a C like that.
155
00:12:19,000 --> 00:12:25,000
So say you wanted to target a 15 nucleotide DNA sequence that appears in a junior interest in,
156
00:12:25,000 --> 00:12:31,000
you just have to go to your lab freezer where you have stocks of different DNA pieces that encode tail repeats,
157
00:12:31,000 --> 00:12:48,000
each of which have been pre-modified to have molecular cloning sites attached to their ends that allow you to assemble them all into a full length tail gene in a one-pot reaction given the right mix of temperatures and enzymes to assemble everything together into a large DNA vector called a plasmid.
158
00:12:48,000 --> 00:12:53,000
And then you spend about a week sifting through the different reaction products to find one that was correct.
159
00:12:53,000 --> 00:13:04,000
So basically as long as you've done a lot of upfront work to get all the pieces prepared for this process, you could create a totally new designer tail or detail as they're known in about a week.
160
00:13:04,000 --> 00:13:16,000
That was a sensation, that was newsworthy and it was investment worthy with new companies bringing up around the technology and established companies spending, sending teams of lawyers to negotiate a license to use the technology.
161
00:13:16,000 --> 00:13:25,000
But so far what these proteins had evolved to do over millions of years was just to act as transcription factors, not as programmable in their nucleases.
162
00:13:25,000 --> 00:13:35,000
The TA in tail is short for transcription activator. Naturally, tails are delivered by invading bacteria into plant host cells where they then navigate to the nucleus,
163
00:13:35,000 --> 00:13:45,000
and bind to the specific gene that has a DNA sequence matching the pattern of repeats that make up the most of the bulk of the tail.
164
00:13:45,000 --> 00:13:53,000
Then when it has found its target gene, it will recruit RNA polymerase to initiate transcription the first step in gene expression.
165
00:13:53,000 --> 00:13:57,000
In other words, it will activate the gene switching on gene expression.
166
00:13:57,000 --> 00:14:14,000
And the most famous example of this, what it does is help the bacteria by switching on production of a specific gene that encodes a sugar transport protein, the pump sugar out of the plant cell into the surrounding extracellular matrix where the bacteria are feeding and replicating.
167
00:14:14,000 --> 00:14:19,000
Essentially forcing the plant cell to feed the bacteria that are attacking it.
168
00:14:19,000 --> 00:14:25,000
So yeah, tails naturally are transcription activators switching on genes is useful to some extent.
169
00:14:25,000 --> 00:14:31,000
So just being able to rearrange natural tails to make designer tails is useful.
170
00:14:31,000 --> 00:14:37,000
But as I mentioned, it was really programmable and on nucleases that were seen for decades as the real prize.
171
00:14:37,000 --> 00:14:41,000
But now I should explain why. What is actually so useful of being able to cut DNA?
172
00:14:41,000 --> 00:14:57,000
Well, it's not actually the cutting of the DNA that's so interesting. In fact, if you just cut DNA inside a cell, really cut all the way through both strands severing a link in the chain of the long DNA molecule called a chromosome, that is actually fatal for the cell. It cannot be tolerated.
173
00:14:57,000 --> 00:15:09,000
And so cells in all organisms have a DNA repair kit in the form of a range of enzymes that can work together to carry out a few different types of repair, healing that break.
174
00:15:09,000 --> 00:15:17,000
The simplest approach in terms of the number of DNA parts involved is called non-homologous and joining or NHEJ.
175
00:15:17,000 --> 00:15:29,000
This literally means sticking the two ends together and in practice, this involves either chewing back the DNA a bit to create very short matching DNA sequences that can be stuck together.
176
00:15:29,000 --> 00:15:43,000
Or just chewing back one strand to find an A and then you chew back the other to find a T to give you an example. And then you can stitch those together using homologous base pairing, which if you remember is A to T and G to C.
177
00:15:43,000 --> 00:15:50,000
Alternatively, a few DNA bases can be added randomly at the break site in the same way, creating a tiny piece of overlap.
178
00:15:50,000 --> 00:15:59,000
So for example, a G on one strand and a C on the other that can bond together to form a GC bond, which can ultimately heal the entire double strand break.
179
00:15:59,000 --> 00:16:15,000
We describe non-homologous and joining as error prone, meaning that will heal the DNA cut site, but it will almost always change the DNA sequence in the process, introducing short deletions or insertions of random DNA when compared to the original sequence before you make the cut.
180
00:16:15,000 --> 00:16:31,000
That is actually very useful from a biotinology perspective, because if you target a gene with one of those cuts, then the process of healing the cut will introduce mutations into the gene that will in most cases disrupt the function of the gene, essentially breaking the gene.
181
00:16:31,000 --> 00:16:35,000
So let's give a real odd example of why that can be useful.
182
00:16:35,000 --> 00:16:49,000
Sickle cell disease is a brutal genetic disease. It mostly impacts people of West African descent, and it is caused by mutation in the gene encoding hemoglobin, the crucial oxygen carrying protein in our red blood cells.
183
00:16:49,000 --> 00:17:02,000
More specifically, it is a mutation in the gene encoding adult hemoglobin. There is a separate gene in our genomes that encodes a version of hemoglobin, expressed wireria in, while we are in the womb, that's fetal hemoglobin.
184
00:17:02,000 --> 00:17:12,000
In adults, that fetal hemoglobin is not produced because the gene is switched off by a transcriptional repressor protein.
185
00:17:12,000 --> 00:17:22,000
As long as you can get into the blood producing bone marrow cells, you can use a programmable endonuclease to make a cut in the gene that encodes that transcriptional repressor protein.
186
00:17:22,000 --> 00:17:29,000
Then you can let non-homologous end joining, do its thing to mutate the gene and basically break it.
187
00:17:29,000 --> 00:17:37,000
Now, without that transcriptional repressor, the fetal hemoglobin gene switches back on again, even in adult bone marrow cells.
188
00:17:37,000 --> 00:17:44,000
And now the bone marrow cells start producing fetal hemoglobin alongside the dysfunctional adult hemoglobin.
189
00:17:44,000 --> 00:17:50,000
That's not enough to totally remove sickle cell disease symptoms, but enough to vastly improve them.
190
00:17:50,000 --> 00:18:00,000
And that is exactly the mechanism behind the first and only CRISPR-Cas9-based gene therapy currently available in the US, a therapy sold under the brand name Cass Gavie.
191
00:18:00,000 --> 00:18:12,000
And if you think about crop plants, one of the big uses would be to mutate, or as is generally said in plant genetics, knock out a gene that encodes a protein that is targeted as a receptor of pathogens.
192
00:18:12,000 --> 00:18:22,000
It is a common feature of pathogens in all organisms that they need something to latch onto so they can stay attached to the host cell long enough to do their thing.
193
00:18:22,000 --> 00:18:31,000
This is also true of viruses. For example, the SARS-CoV-2 virus that causes COVID-19 latches onto a human protein called ACE2.
194
00:18:31,000 --> 00:18:37,000
The same basic mechanism is crucial for bacterial pathogens, fungal pathogens and so on.
195
00:18:37,000 --> 00:18:45,000
There has to be some host molecule, usually a protein, and usually one that sticks out of the cell membrane for the pathogen to latch onto.
196
00:18:45,000 --> 00:18:56,000
In all cases, these host proteins, what we often refer to as pathogen receptor proteins, did not evolve to offer an attachment site per pathogen that would make no sense.
197
00:18:56,000 --> 00:19:06,000
They have some other natural function for the cell, but from the perspective of a farmer or plant breeder, whatever purpose that pathogen receptor protein actually serves in the normal healthy function of the plant,
198
00:19:06,000 --> 00:19:16,000
is probably outweighed by the damage done by the pathogen being able to use or co-opt that protein to invade the plant's cells and ultimately destroy a field of the crop plant.
199
00:19:16,000 --> 00:19:25,000
So being able to knock out a pathogen receptor protein in plant genomes is a really desirable use of programmable endonucleases.
200
00:19:25,000 --> 00:19:32,000
But so far, I've only talked about one of the possible repair pathways for the cut produced by a DNA endonuclease.
201
00:19:32,000 --> 00:19:42,000
There is another option that is even more exciting, homology-dependent repair or HDR, and this approach, a second DNA molecule is used as a repair template.
202
00:19:42,000 --> 00:19:57,000
And the real trick is that as long as the repair template has a good amount of perfectly matching DNA sequence at its beginning and end, the middle can be different to what was originally in the targeted DNA molecule where you made the cut.
203
00:19:57,000 --> 00:20:08,000
It's kind of a Trojan horse approach, trick the cell into thinking that the repair DNA you've given it is just a perfect match that it can use to close the DNA break that it needs to repair.
204
00:20:08,000 --> 00:20:12,000
But in the process, it is actually introducing a mutation.
205
00:20:12,000 --> 00:20:26,000
But in this case, the mutation would not be random, it would be a specific DNA change designed in a lab and built into the repair template DNA that is delivered into the cell alongside the programmable endonuclease.
206
00:20:26,000 --> 00:20:35,000
This approach, where you can change the DNA sequence however you want to, is called gene editing or done at scale genome editing.
207
00:20:35,000 --> 00:20:44,000
Sometimes gene editing is also used as a term to describe the random targeted mutations, targeted mutations described previously.
208
00:20:44,000 --> 00:20:54,000
But most of the time, if you hear the term gene editing, it means these specific DNA changes introduced at specific sites in a given genome using homology-dependent repair.
209
00:20:54,000 --> 00:21:01,000
These DNA changes are technically mutations because that is a word used for any change to a DNA sequence.
210
00:21:01,000 --> 00:21:10,000
But one of the uses of gene editing that is most exciting in terms of human health is to use it to fix inherited genetic mutations that cause disease.
211
00:21:10,000 --> 00:21:22,000
So to connect back to sickle cell disease, it would be seemingly much better, more elegant and better for the patient if you could just go in and fix the mutation in the adult hemoglobin gene
212
00:21:22,000 --> 00:21:28,000
rather than this convoluted approach of breaking a transcriptional repressor to allow expression of fetal hemoglobin.
213
00:21:28,000 --> 00:21:33,000
So you might think, well why wouldn't you just do that? Why don't just go do gene editing? Why do they use that convoluted approach?
214
00:21:33,000 --> 00:21:43,000
Well here's the catch. Gene editing is actually a lot harder to do because cells generally prefer the approach of non-homologous end joining over homology-dependent repair.
215
00:21:43,000 --> 00:21:52,000
And because at a technical level, it is difficult to get that repair template DNA into the cells where you need it to be in order for gene editing to work.
216
00:21:52,000 --> 00:21:59,000
So that pours a little bit of cold water on the excitement on the new era of gene editing that CRISPR has ushered in.
217
00:21:59,000 --> 00:22:06,000
It is still something very difficult to do even if we have a perfect tool for at least the first step in gene editing.
218
00:22:06,000 --> 00:22:20,000
But back in 2009, even that first step was basically non-existent because we had no programmable endonucleases, apart from the fairly clunky zinc fingers, to be able to make the DNA cuts that were the first step in gene editing.
219
00:22:20,000 --> 00:22:30,000
So that's why when tails appeared on the scene, multiple research teams scrambled to work out how to turn those tails into talons, that's the nuclease version,
220
00:22:30,000 --> 00:22:38,000
and that was done at the DNA level by fusing a non-specific DNA nuclease domain onto the end of the tail gene.
221
00:22:38,000 --> 00:22:45,000
Basically take this thing that doesn't naturally have the ability to cut DNA and add on an extra domain that will allow it to do that.
222
00:22:45,000 --> 00:22:53,000
Much to my dismay, that was successfully demonstrated just a few months into my PhD, but not by me.
223
00:22:53,000 --> 00:23:03,000
Deflated, I moved on to a different project that was more linked to the natural role of tails in causing devastated crop diseases called bacterial spot and bacterial wilt disease.
224
00:23:03,000 --> 00:23:12,000
But just a couple of years later, while I was midway through my new project, I've read a research update that may be very happy that I hadn't hitched my academic success to talons.
225
00:23:12,000 --> 00:23:21,000
That update was a description in probably Nature or Science magazine describing recently published research into a system called CRISPR Cas9.
226
00:23:21,000 --> 00:23:32,000
And after spending just a minute or two reading and then cross checking the actual research article authored by the labs of Jennifer Dowdner and Amantiel Schump, Amantiel Schupanti, I knew that this was the future.
227
00:23:32,000 --> 00:23:35,000
Not just me, really everyone did.
228
00:23:35,000 --> 00:23:42,000
The paper is titled "A programmable dual RNA guided DNA and the nuclease and adaptive bacterial immunity" and I've linked it in the show notes.
229
00:23:42,000 --> 00:23:49,000
It is open access so you don't need to pay to read it, which is good because I think it's an important document for the history of humanity.
230
00:23:49,000 --> 00:23:59,000
If you've listened to many episodes of this podcast, you know I generally shy away from hyperbole and all the smoke and mirrors of the tech world, whether that's biotech or regular silica based tech.
231
00:23:59,000 --> 00:24:03,000
But CRISPR really was and is a world changing technology.
232
00:24:03,000 --> 00:24:07,000
I was not the only one to immediately say how huge this would be.
233
00:24:07,000 --> 00:24:14,000
CRISPR got a massive and rapturous reception from the entire biotechnology community and garnered headlines around the world.
234
00:24:14,000 --> 00:24:20,000
Dowdner and Schupanti got the Nobel Prize just eight years after publication of that paper.
235
00:24:20,000 --> 00:24:29,000
That sort of speed is very unusual and it's a testament to how quickly CRISPR has reshaped the field and for the first time, made gene editing feel within reach.
236
00:24:29,000 --> 00:24:41,000
CRISPR is the shorthand, so often used for CRISPR Cas9, the combination of the Cas9 protein and a guided RNA that together work as a programmable DNA and a nuclease.
237
00:24:41,000 --> 00:24:47,000
The Cas9 protein itself cannot do anything until it is brought into close contact with DNA.
238
00:24:47,000 --> 00:24:52,000
And then when that has happened, it will make a cut at that site in the DNA.
239
00:24:52,000 --> 00:25:02,000
To get the Cas9 protein to the DNA, it partners up with a guide RNA molecule that is a short molecule made of DNA's molecular cousin RNA.
240
00:25:02,000 --> 00:25:20,000
The guide RNA physically embeds within the Cas9 protein, but a 20 nucleotide region of the guide RNA called the spacer is exposed and is basically able to scan along the genome in the target cell until it meets an exactly complementary 20 nucleotide DNA sequence.
241
00:25:20,000 --> 00:25:31,000
At that point, once the guide RNA has found its matching DNA sequence, the Cas9 protein gets stuck in place long enough and securely enough to do its thing and make a difference.
242
00:25:31,000 --> 00:25:42,000
So all you have to do to get Cas9 where you want it is to modify those 20 nucleotides of the gene encoding the guide RNA.
243
00:25:42,000 --> 00:25:54,000
The Cas9 protein remains the same. And since guide RNA genes are so short, they are cheap and easy to chemically synthesize, meaning anyone can sit down and order one for maybe $100.
244
00:25:54,000 --> 00:26:06,000
You'll get your own guide RNA gene in just a few days. You can also use laboratory methods to modify that 20 nucleotide spacer region of the guide RNA if you happen to have a guide RNA gene already lying around the lab.
245
00:26:06,000 --> 00:26:09,000
That is quick and easy taking just a few days.
246
00:26:09,000 --> 00:26:23,000
So that is the huge advantage of CRISPR. It is quick and easy to make a guide gene, a guide RNA gene, that you can package with Cas9 to deliver into a cell to target Cas9's activity to any gene you want.
247
00:26:23,000 --> 00:26:28,000
And that all takes just a matter of days compared to the weeks required to create a talent.
248
00:26:28,000 --> 00:26:33,000
It may seem a minor logistical advantage, but it is one of several that all add up.
249
00:26:33,000 --> 00:26:42,000
Another advantage is that the targeting system is based on RNA, which uses the same molecular matching system as DNA, homologous base pairing.
250
00:26:42,000 --> 00:26:52,000
In practice, if you know what 20 nucleotide DNA sequence in a target genome you want to target, then that 20 nucleotide DNA sequence is actually the exact DNA sequence.
251
00:26:52,000 --> 00:27:12,000
You need to insert into your guide RNA gene as the spacer because that DNA will be transcribed to produce a complementary RNA that then can go search for its complementary DNA, which would be the same sequence as the original DNA used to produce the RNA.
252
00:27:12,000 --> 00:27:24,000
What this means is that it is incredibly easy to design guide RNAs, a lot easier than designing talents, which requires some nuance understanding of the protein to DNA code that underlies the function of talents.
253
00:27:24,000 --> 00:27:40,000
Almost no expertise is required to design guide RNAs. Almost immediately after CRISPR Cas9 was first described, free web tools popped up, the maker as simple as defining what gene you are interested in targeting, and it will spit out for you a whole set of guide RNA sequences you can use.
254
00:27:40,000 --> 00:27:50,000
And even companies offering to just send you the guide RNA gene that would do what you wanted to do as long as you tell it what gene you want to target.
255
00:27:50,000 --> 00:28:03,000
And in the process of designing the guide RNA, these tools will check against a computer database of the whole genome of the target organism to ensure that the guide RNA sequence being designed is totally unique within the genome.
256
00:28:03,000 --> 00:28:18,000
Another advantage is that CRISPR Cas9 just works incredibly well, generally better than talents, meaning it cuts more efficiently and it is about as good at avoiding mistakes where it accidentally cuts a similar but slightly different sequence, what we call an off target.
257
00:28:18,000 --> 00:28:24,000
Those off targets, they do happen and they are a concern. They tend to happen at very low rates.
258
00:28:24,000 --> 00:28:35,000
And after about 15 years of CRISPR being on the scene, modified versions of the Cas9 protein are available that have even better performance in regards to off targets.
259
00:28:35,000 --> 00:28:40,000
Another logistical consideration is that Cas9 is a single protein.
260
00:28:40,000 --> 00:29:00,000
What I have yet to mention is that to get tails to servers nucleases basically to be come talons, you have to rearrange them as a pair of proteins facing each other, each with a nuclease domain that cuts one DNA strand each, together creating the complete break in the DNA molecule.
261
00:29:00,000 --> 00:29:13,000
Each talon protein is big and bulky and it is a logistical problem to get big bulky proteins into cells, especially if you want to deliver that as protein, not as DNA encoding the protein.
262
00:29:13,000 --> 00:29:24,000
The advantage of delivering proteins directly into the cell is that it is then an absolutely zero risk that that endonuclease somehow gets integrated permanently into the cell.
263
00:29:24,000 --> 00:29:35,000
Proteins, by themselves, are inherently unstable molecules within cells. There is no biological mechanism for a protein to become permanently integrated into a cell.
264
00:29:35,000 --> 00:29:49,000
Whereas DNA, depending on the delivery method, may well stably insert itself somewhere randomly into the genome of the target cell, making a permanent DNA change that is just a side effect of the gene edit you are actually trying to make.
265
00:29:49,000 --> 00:29:55,000
Keep that in mind because it is one of the keys to the biggest selling points of using CRISPR in crop engineering.
266
00:29:55,000 --> 00:30:02,000
Cas9 is a fairly bulky protein itself, but it is just one protein that you have to get into the cell, not two.
267
00:30:02,000 --> 00:30:10,000
And there are also modified versions of the Cas9 protein that are smaller and easier to deliver as a protein into cells.
268
00:30:10,000 --> 00:30:24,000
So CRISPR Cas9 was just like a faster, better, stronger version of talons. The felt in reach to really anyone with some basic knowledge of molecular biology, and scientists immediately saw all those advantages, took it and ran with it.
269
00:30:24,000 --> 00:30:34,000
Most of that metaphorical running originally happened in academic research labs using CRISPR Cas9 to modify genes as part of molecular genetic research programs.
270
00:30:34,000 --> 00:30:46,000
Academic researchers were free to take CRISPR Cas9 and run with it because they didn't need a license. As you might expect, CRISPR Cas9 as a tool for gene editing was immediately patented.
271
00:30:46,000 --> 00:30:51,000
However, even though the patents were immediate, they were unfortunately multiple.
272
00:30:51,000 --> 00:31:03,000
So what I haven't mentioned so far was that the Daoudnou Champagne team filed a patent once they published their discovery, but so too did a team at the Broad Institute,
273
00:31:03,000 --> 00:31:10,000
which is affiliated with MIT and led by a biotechnology professor Feng Zhang.
274
00:31:10,000 --> 00:31:21,000
You see, the Zhang lab had published a demonstration of the use of CRISPR Cas9 in eukaryotic cells just a few months after the original work by Daoudnou and Chappontier.
275
00:31:21,000 --> 00:31:28,000
A pretty classic situation of two researchers ending up doing a similar thing and publishing at almost the same time.
276
00:31:28,000 --> 00:31:43,000
Daoudnou and Chappontier were CRISPR Cas experts, but not so much biotechnologists. They did see the potential of their discovery and they tried to demonstrate the use of CRISPR Cas9 as a programmable endonuclease in their original work and they used that to file a patent.
277
00:31:43,000 --> 00:31:55,000
But Zhang and his lab were more experienced in recombinant DNA methods and had seen the possible applications of CRISPR Cas9 based on some research into natural CRISPR systems published earlier on in 2011.
278
00:31:55,000 --> 00:32:10,000
So they were basically working on a very similar idea that didn't really care about the natural CRISPR system, but they saw this idea of how you could use Cas9 as a programmable endonuclease and they published that just a few months after the Daoudnou Champagne paper.
279
00:32:10,000 --> 00:32:24,000
The result of this was a 10 year legal battle that was only resolved in 2017 actually in favor of Zhang and the Broad Institute, a decision that has been reaffirmed multiple times, including as recently as 2026.
280
00:32:24,000 --> 00:32:33,000
And as you might imagine, that legal battle has significantly held back commercial investment in an applications of CRISPR based technology.
281
00:32:33,000 --> 00:32:43,000
Despite that, many companies pressed ahead with R&D, even while they weren't really sure who they would have to pay to get a license to use the technology inherent to their approach.
282
00:32:43,000 --> 00:33:01,000
By now that obstacle is cleared and it's even easier and as I mentioned earlier, multiple variants of Cas9 have now been described from different bacterial species or created through deliberate modification of the Cas9 gene in labs, producing a range of Cas9 variants and alternatives that have their own patents.
283
00:33:01,000 --> 00:33:10,000
So anyone looking to use Cas9 or one of these variants commercially has a range of different options of potential partners to purchase a license from.
284
00:33:10,000 --> 00:33:20,000
So there are a range of options, but what actually are the products that people are trying to work on? Well, I mentioned human gene therapies, but this is a podcast about crops.
285
00:33:20,000 --> 00:33:26,000
What are crop scientists in universities, government research labs and companies trying to do with CRISPR?
286
00:33:26,000 --> 00:33:33,000
First, the general theoretical answer about what you can do with CRISPR that you can't do with general transgenics.
287
00:33:33,000 --> 00:33:46,000
Well, one thing that you can do is go in and alter genes that are in the plant, for example, edit just a few DNA bases to change the part of a gene that encodes the active site of an enzyme that might allow the plant to better tolerate drought or heat stress.
288
00:33:46,000 --> 00:33:55,000
So remember that there's now been decades and decades of what we call molecular genetic research into plants and their genes and genomes.
289
00:33:55,000 --> 00:34:05,000
So we know of all sorts of genes that could be kind of very powerful if you made a slight DNA change to them. Maybe you know that there's a wild relative that's more drought resistant.
290
00:34:05,000 --> 00:34:14,000
And if you could modify the version of the corresponding gene in the crop plan to mimic the version that's in that wild relative, you could probably get some boosted drought resistance.
291
00:34:14,000 --> 00:34:22,000
So there's all sorts of ideas like that floating around just waiting to be implementable with the right tools.
292
00:34:22,000 --> 00:34:41,000
That is not the sort of thing you could do by just inserting a new gene. You would just be layering something new on top of what was already exists inside the crop, which sometimes might be useful, but sometimes having both versions simultaneously expressed and functioning will kind of ruin the effect you're trying to achieve.
293
00:34:41,000 --> 00:34:48,000
But there is another very significant reason that crop scientists were immediately excited about CRISPR-Cas9, probably the biggest reason.
294
00:34:48,000 --> 00:35:04,000
If you can get the Cas9 and the Guida RNA, so that's as protein and RNA, not as DNA encoding those things, get them into Calis cells. Those are those uninduced, sorry, those are those induced plant stem cells that are then used to regenerate your whole plant.
295
00:35:04,000 --> 00:35:24,000
So if you can get that Cas9 protein and the Guida RNA into Calis cells and the Cas9 protein and Guida RNA can work together to get to a gene of interest, break that gene, or even better if you can get the Cas9 Guida RNA and a template DNA molecule into the cells and let them work together to edit the gene of interest.
296
00:35:24,000 --> 00:35:40,000
Well, after that gene edit, the cells will continue to divide. After a few hours, the Cas9 protein and the Guida RNA will be destroyed because proteins and RNA are for the most part short-lived molecules and cells are full of enzymes that will destroy RNA and proteins.
297
00:35:40,000 --> 00:35:52,000
One new ones are turned out to take their place, except of course the cells doesn't have the genes to make the Cas9 protein and the Guida RNA because they didn't come from the plant, they were literally injected or inserted into the cells.
298
00:35:52,000 --> 00:36:02,000
And the template DNA, it will also be destroyed because in general the template DNA that we use in this sort of experiment is a short linear DNA molecule.
299
00:36:02,000 --> 00:36:08,000
So that's kind of naked DNA, not the huge long protein package DNA molecules we call chromosomes.
300
00:36:08,000 --> 00:36:18,000
And those short linear DNA molecules, similar to RNA and proteins are inherently unstable in cells, they will be chewed up and destroyed within one or two cell generations.
301
00:36:18,000 --> 00:36:28,000
It's very unlikely, but in rare occasions that DNA template, sorry, that template DNA might get inserted into the genome of the plant cell.
302
00:36:28,000 --> 00:36:42,000
But that would happen very rarely and it would be very easy for the crop scientists to check for this and exclude any such rare occurrences from what would be the tens or hundreds of developing calli being created in parallel to be sure of getting at least one good one.
303
00:36:42,000 --> 00:37:02,000
So as the callus develops into an adult plant, the Cas9 protein, the Guida RNA and the template DNA are all destroyed. But what does remain is the edit in the target gene in the actual genome of the crop plant that is a permanent change that is passed down through each generation of cells and the developing plant and then to all the progeny of that plant.
304
00:37:02,000 --> 00:37:10,000
And what to make of those progeny of the gene editing experiment, at the genetic level, they contain no outside DNA.
305
00:37:10,000 --> 00:37:32,000
No genes that were not generated from within the plant itself as part of its evolutionary journey, just slightly tweaked versions of those genes, not really any different to the general genetic variation that can occur in the course of evolution and the more accelerated process of artificial selection that has been used for millennia to produce crop plants.
306
00:37:32,000 --> 00:37:41,000
Would that be considered a genetically modified organism? Well, many companies are banking on the answer to that question being no.
307
00:37:41,000 --> 00:37:52,000
What that would mean in practice is a huge reduction in costs as these crop varieties would be treated no differently to any other crop variety developed using non genome genome methods.
308
00:37:52,000 --> 00:38:05,000
That is to say, they would be assumed safe and require no regulatory approval at all before being sold. And they would be allowed to be sold in jurisdictions that currently totally banned the sale of GMOs such as most countries in the EU.
309
00:38:05,000 --> 00:38:16,000
Well, unfortunately, if you're hoping into a deep dive into all the regulation around CRISPR crops, you'll have to wait until season 2. This episode is already long enough, but I can give you a brief summary.
310
00:38:16,000 --> 00:38:35,000
In the last five years, countries around the world have taken this question very seriously and most have basically landed somewhere in the middle agreeing to relax regulations on crops created using CRISPR Cas9 genetic scene, especially if it's done in a way that leaves no trace of the Cas9 or Guida on agines in the plant as I was just describing.
311
00:38:35,000 --> 00:38:52,000
In the USA, the most relevant regulatory agency, the USDA, actually created a totally new regulatory framework that would have wiped away most regulation from crop varieties developed using CRISPR, but a judge in 2025 vacated that regulatory framework based on process issues.
312
00:38:52,000 --> 00:39:16,000
So that has added further confusion and delay that has to the original confusion delay caused by the patent battles. And so it's maybe not surprising that despite the immediate rush of excitement to make use of CRISPR as a tool going back more than a decade now, what I've been calling GMO 2.0, which is CRISPR gene editing or crops varieties created using CRISPR gene editing have not really hit the market.
313
00:39:16,000 --> 00:39:31,000
Despite that, so as far as I can work out, there are zero CRISPR genetic crop varieties for sale in the USA. There are several in the works though and there is a CRISPR gene editing crop for sale in one country, Japan.
314
00:39:31,000 --> 00:39:46,000
The Sicilian Rouge High Gabba Tomato, created and sold by Sun Attack Life Science, much like the purple tomatoes covered last episode, this is a variety modified to produce higher concentrations of chemicals with purported health benefits.
315
00:39:46,000 --> 00:40:02,000
So maybe to wrap up this episode, in episode 11, we met the first ever GMO crop, a tomato, and we can end here with I believe what is the first commercially available GMO 2.0 CRISPR gene edited crop, another tomato.
316
00:40:02,000 --> 00:40:21,000
That is it for season 1. I've had a blast putting it together and I've learned a lot, not just about the history of science and GMOs, filling in a lot of gaps in my knowledge as I researched each episode, but also about how to do audio editing and video editing as I've also started to do more video content, and how to generally create a podcast.
317
00:40:21,000 --> 00:40:32,000
I feel very far away from the standards that I would hope one day to achieve and I have a vague notion that some point in future will come back and redo this season, but really stepping everything up a notch.
318
00:40:32,000 --> 00:40:36,000
In the meantime, we will keep moving forward into new territory.
319
00:40:36,000 --> 00:40:50,000
Season 2 will be putting the basic science of GMOs a little bit more in the background and bringing the economic and social impacts of GMOs more into the spotlight, but there will still be a mixture of both, so don't worry science nerds, I won't let you down.
320
00:40:50,000 --> 00:41:00,000
But the thing is, I will have a lot of research to do before I feel ready to start writing out the episodes and so I'm going to give myself a pretty long break to do that.
321
00:41:00,000 --> 00:41:16,000
For now, I'm setting a target date of October 3rd, 2026, to premiere season 2. I'll aim to release some short episodes ever summer to let you know what I'm working on, and I'll be experimenting with more video content, so make sure to follow me on YouTube at ModifiedPod.
322
00:41:16,000 --> 00:41:23,000
So before I say goodbye for now, here are your five take home messages for this episode.
323
00:41:23,000 --> 00:41:30,000
Number 1. CRISPR-Cast systems are found in most bacteria and are immune systems that defend against viruses.
324
00:41:30,000 --> 00:41:41,000
Number 2. Cas9, together with a guide RNA, acts as a programmable DNA and a nucleus, which can make DNA cuts at target sites based on a 20 nucleotide spacer region in the guide RNA.
325
00:41:41,000 --> 00:41:54,000
Number 3. CRISPR as a tool for gene editing was first demonstrated in 2012 and it immediately generated excitement, however for a mix of reasons commercial products are now only just arriving on the market.
326
00:41:54,000 --> 00:42:08,000
Number 4. A DNA break will in most cases result in random mutations at the break site, but if a DNA repair template is present, it will sometimes result in the template sequence being inserted at the break site, this is called gene editing.
327
00:42:08,000 --> 00:42:18,000
Number 5. Gene edited crops facing uncertain regulatory environment currently, but in many countries they are more likely regulated than conventional transgenic crops.
328
00:42:18,000 --> 00:42:25,000
Well, season one has been a blast and it's time for me to get right to work on season 2. Bye!
329
00:42:25,000 --> 00:42:45,140
[ Silence ]
00:00:00,000 --> 00:00:03,000
scientific research is a strange endeavor.
2
00:00:03,000 --> 00:00:06,000
You spend years working on a project.
3
00:00:06,000 --> 00:00:09,000
You carry out experiment after experiment.
4
00:00:09,000 --> 00:00:12,000
You collect data, you double check all your calculations,
5
00:00:12,000 --> 00:00:15,000
and then package it all up with an introduction and a discussion
6
00:00:15,000 --> 00:00:18,000
into a product called a research paper.
7
00:00:18,000 --> 00:00:20,000
And then you just throw that out into the world,
8
00:00:20,000 --> 00:00:23,000
actively encouraging anyone around the whole world to read it,
9
00:00:23,000 --> 00:00:28,000
to make what they will of it, and to actively try to prove you wrong.
10
00:00:28,000 --> 00:00:32,000
Sometimes, those people go on to become your collaborators.
11
00:00:32,000 --> 00:00:35,000
And sometimes you never ever meet them.
12
00:00:35,000 --> 00:00:41,000
Greg Amendel published his "Vesucher Ubefranzenhebrudin" in 1866,
13
00:00:41,000 --> 00:00:46,000
his most famous publication, and he died in 1884.
14
00:00:46,000 --> 00:00:49,000
It wasn't until almost two decades after his death
15
00:00:49,000 --> 00:00:52,000
that other scientists around Europe and the world began to read his papers
16
00:00:52,000 --> 00:00:56,000
and pick up where he left off to create the field of genetics.
17
00:00:56,000 --> 00:00:58,000
That's not really so unusual.
18
00:00:58,000 --> 00:01:01,000
Research is a process that does not always run smoothly.
19
00:01:01,000 --> 00:01:04,000
Sometimes it can take years for an idea to bounce around
20
00:01:04,000 --> 00:01:07,000
between the right group of people until at some point,
21
00:01:07,000 --> 00:01:09,000
quite to the surprise of everyone involved,
22
00:01:09,000 --> 00:01:14,000
it turns into something new, something shocking that changes the world.
23
00:01:14,000 --> 00:01:17,000
And so it was, in 1987,
24
00:01:17,000 --> 00:01:20,000
a PhD student called Yoshizumi Ishimo,
25
00:01:20,000 --> 00:01:23,000
working at the University of Osaka,
26
00:01:23,000 --> 00:01:27,000
published a paper and sent it off into the world.
27
00:01:27,000 --> 00:01:32,000
The paper was titled "Nuclear Tide Sequence of the IAP gene,
28
00:01:32,000 --> 00:01:35,000
responsible for alkaline phosphatase, isoenzyme conversion,
29
00:01:35,000 --> 00:01:40,000
and escherichia coli, and identification of the gene product."
30
00:01:40,000 --> 00:01:43,000
Not exactly a title that would make you think
31
00:01:43,000 --> 00:01:46,000
a revolution lies in this paper.
32
00:01:46,000 --> 00:01:50,000
Alkaline phosphatases are important in maintaining the phosphate balance of the cell,
33
00:01:50,000 --> 00:01:54,000
and they were the main focus of the research, which is essentially to sequence a gene
34
00:01:54,000 --> 00:01:58,000
encoding an enzyme involved in alkaline phosphatase production in the bacterium
35
00:01:58,000 --> 00:02:02,000
escherichia coli, more commonly known as e-coli.
36
00:02:02,000 --> 00:02:04,000
But in the process of sequencing that gene,
37
00:02:04,000 --> 00:02:06,000
and I will emphasize in the 1980s,
38
00:02:06,000 --> 00:02:10,000
sequencing a gene was a serious undertaking worthy of a PhD.
39
00:02:10,000 --> 00:02:12,000
During the process of sequencing that gene,
40
00:02:12,000 --> 00:02:16,000
Dr Ishimo found something else, something unexpected,
41
00:02:16,000 --> 00:02:19,000
just after the gene, physically in that genome,
42
00:02:19,000 --> 00:02:22,000
he found something that was not what he was interested in,
43
00:02:22,000 --> 00:02:25,000
but he did describe it in the final part of the paper.
44
00:02:25,000 --> 00:02:30,000
He wrote, "An unusual structure was found in the three-prime end,
45
00:02:30,000 --> 00:02:33,000
flanking region of the gene of interest.
46
00:02:33,000 --> 00:02:39,000
Five highly homologous sequences of 29 nuclear tides were arranged as direct repeats
47
00:02:39,000 --> 00:02:43,000
with 32 nuclear tides as spacing.
48
00:02:43,000 --> 00:02:48,000
What he had shared with the world was the first description of an intriguing phenomenon,
49
00:02:48,000 --> 00:02:53,000
that will get picked up and explored by scientists around the globe.
50
00:02:53,000 --> 00:02:57,000
These repeating sequences seemed to be found in most bacteria
51
00:02:57,000 --> 00:03:03,000
and between the repeating units were little pieces of DNA that matched known viruses.
52
00:03:03,000 --> 00:03:07,000
Digging deeper research has found that these repeating DNA elements are managed
53
00:03:07,000 --> 00:03:11,000
and created by a set of bacterial proteins.
54
00:03:11,000 --> 00:03:15,000
The repeating DNA units came to be known as clustered,
55
00:03:15,000 --> 00:03:19,000
but regularly interspaced short palindromic units.
56
00:03:19,000 --> 00:03:25,000
The proteins that managed and and created these clusters were clustered associated proteins or cast proteins.
57
00:03:25,000 --> 00:03:28,000
They turned out to be many of these cast proteins.
58
00:03:28,000 --> 00:03:30,000
They were given numbers as they were discovered,
59
00:03:30,000 --> 00:03:33,000
cast one, cast two, cast three, and so on.
60
00:03:33,000 --> 00:03:39,000
By 2020, we were decades into the existence of this fairly niche field of research
61
00:03:39,000 --> 00:03:43,000
interested by then in what had become to understood as a bacterial immune system,
62
00:03:43,000 --> 00:03:48,000
protecting the cell against viruses by integrating small chunks of viral DNA
63
00:03:48,000 --> 00:03:54,000
into basically a memory bank for the cell to know what to look out for and destroy in future.
64
00:03:54,000 --> 00:04:01,000
The next year in 2011, a small side session was held to discuss these bacterial immune systems
65
00:04:01,000 --> 00:04:06,000
as part of a bigger microbiology conference in San Juan, Puerto Rico.
66
00:04:06,000 --> 00:04:11,000
And at the sidelines of that session, as fringe as could be the side of a side session,
67
00:04:11,000 --> 00:04:14,000
something happened that would change the world.
68
00:04:14,000 --> 00:04:18,000
A French professor in Germany called a manual of Sheponcier
69
00:04:18,000 --> 00:04:25,000
approached an American professor who had grown up in Hilo-Hawaii but worked at UC Berkeley, Jennifer Doudna.
70
00:04:25,000 --> 00:04:31,000
Sheponcier felt that there are two research teams between them had just the right mix of expertise
71
00:04:31,000 --> 00:04:36,000
to tackle a project on her mind, solving the function of a particular protein
72
00:04:36,000 --> 00:04:43,000
that was associated with these clustered, regularly interspersed short palindromic repriets
73
00:04:43,000 --> 00:04:46,000
or CRISPR as they're known.
74
00:04:46,000 --> 00:04:50,000
This protein, Cas9, was little known at that point,
75
00:04:50,000 --> 00:04:55,000
but in just nine years, Doudna and Sheponcier would be accepting a prize awarded
76
00:04:55,000 --> 00:05:00,000
based on one of the easiest decisions of the Nobel Committee for their work to solve the function
77
00:05:00,000 --> 00:05:09,000
of CRISPR Cas9 and ushering in a new age of unlimited possibility to engineer genomes as we please.
78
00:05:09,000 --> 00:05:11,000
Welcome to Modified.
79
00:05:11,000 --> 00:05:16,000
I'm your host, Dr. O'Landedalange, Explants Scientist and Current Science Teacher.
80
00:05:16,000 --> 00:05:20,000
This podcast is my attempt to guide you through the science of GMOs
81
00:05:20,000 --> 00:05:25,000
and hopefully just enough of the history and the social context to give you a path forward in your own thinking.
82
00:05:25,000 --> 00:05:33,000
And if all else fails, I hope you'll enjoy some of the surprising stories and fascinating facts in the world of genetically modified plants.
83
00:05:33,000 --> 00:05:35,000
This is episode 16.
84
00:05:35,000 --> 00:05:40,000
CRISPR changes the world and unlocks GMO 2.0.
85
00:05:40,000 --> 00:05:45,000
If this is the first episode you're listening to, because maybe you saw the word CRISPR and thought,
86
00:05:45,000 --> 00:05:49,000
"Hey, I've heard that, I want to check that out." Well, welcome! Happy to have you here.
87
00:05:49,000 --> 00:05:53,000
You may find that I dive pretty deep into genetics and into the world of GMOs in this episode
88
00:05:53,000 --> 00:05:58,000
and so you may want to go back to episode 1 and enjoy the journey to get here,
89
00:05:58,000 --> 00:06:03,000
because of this is the last episode of season 1 of Modified.
90
00:06:03,000 --> 00:06:08,000
In this episode, I'll explain what CRISPR Cas9 actually is and how it works,
91
00:06:08,000 --> 00:06:16,000
and I'll explain what I mean by GMO 2.0 and why I don't think it's silly to predict a wave of CRISPR derived GMOs in the next five years.
92
00:06:16,000 --> 00:06:21,000
But maybe a good way to tee off the main body of this episode is to hear the words of a Manuarsha Pontea herself,
93
00:06:21,000 --> 00:06:26,000
interviewed in 2022. I will spare you me trying to do a French accent.
94
00:06:26,000 --> 00:06:31,000
She was asked, "What is the greatest benefit to humankind of your research?"
95
00:06:31,000 --> 00:06:35,000
And she had this to say, after first explaining possible applications in medicine.
96
00:06:35,000 --> 00:06:40,000
There are also great implications in plant biology in the production of plant crops,
97
00:06:40,000 --> 00:06:44,000
specifically in the world of global warming and global climate change,
98
00:06:44,000 --> 00:06:48,000
the necessity to be able to engineer crops that can sustain the changes
99
00:06:48,000 --> 00:06:53,000
that are meant to be encountered in the very near future and even in the long term future.
100
00:06:53,000 --> 00:06:58,000
In the life of today, we benefit so much of great well-being thanks to a number of technologies
101
00:06:58,000 --> 00:07:03,000
like improvement of food and medicine. CRISPR has a great impact in this.
102
00:07:03,000 --> 00:07:06,000
That's why it has impacts on humankind.
103
00:07:06,000 --> 00:07:12,000
So CRISPR from the get-go was recognized for its possible applications in crop biotechnology
104
00:07:12,000 --> 00:07:15,000
and I can back that up with some personal experience.
105
00:07:15,000 --> 00:07:22,000
So let's talk a little bit about me. In 2010, I started my PhD at the Ludwig-Mexamillian University of Munich
106
00:07:22,000 --> 00:07:29,000
and the goal handed to me for my first project was test the idea of creating a programmable DNA endonuclease
107
00:07:29,000 --> 00:07:34,000
using the DNA binding domains of the bacterial proteins that our group specialized on.
108
00:07:34,000 --> 00:07:39,000
Those proteins are called tails, T-A-L-E, it's an acronym.
109
00:07:39,000 --> 00:07:43,000
And so we decided to add an N to the end of the acronym to represent the word in nuclease
110
00:07:43,000 --> 00:07:50,000
and then you have T-A-L-E-N, talons. Google Talons and you'll still find plenty of material about them.
111
00:07:50,000 --> 00:07:56,000
Talent technology got so much interest because the idea of a programmable endonuclease was an already incredibly hot idea
112
00:07:56,000 --> 00:08:02,000
in the world of biotechnology and so a lot of people were desperate to see if these new talons could finally deliver
113
00:08:02,000 --> 00:08:08,000
would have been tried since the 1990s with only limited success using different proteins called zinc fingers.
114
00:08:08,000 --> 00:08:14,000
But let me back up, what is a programmable endonuclease that is through that word out there?
115
00:08:14,000 --> 00:08:19,000
An endonuclease is a protein that can cut DNA in the middle of a DNA molecule.
116
00:08:19,000 --> 00:08:24,000
Nuclease is a more general word for any protein that can cut DNA.
117
00:08:24,000 --> 00:08:32,000
You may remember that DNA is an acronym for deoxyribonucleic acid, so nuclease references that nucleic acid part.
118
00:08:32,000 --> 00:08:39,000
The hope is that a programmable endonuclease could allow you to make precision cuts in the DNA of any genome
119
00:08:39,000 --> 00:08:45,000
and that is the first step in then permanently changing the sequence of DNA bases of that genome position.
120
00:08:45,000 --> 00:08:49,000
A process known as gene editing or genome editing.
121
00:08:49,000 --> 00:08:52,000
There are many, many nucleases in the world.
122
00:08:52,000 --> 00:09:00,000
Some just chop up all DNA into pieces randomly. That can still be useful for some sorts of goals, both natural and technological.
123
00:09:00,000 --> 00:09:06,000
And there are other nucleases that are called restriction endonucleases or just restriction enzymes.
124
00:09:06,000 --> 00:09:13,000
They were the ones that drove the birth of recombinant DNA technology in the race to create EMOs back in the 1970s.
125
00:09:13,000 --> 00:09:16,000
They are still very commonly used but they are not programmable.
126
00:09:16,000 --> 00:09:23,000
Each restriction enzyme has a single built-in target DNA sequence that it binds and then cuts of that position.
127
00:09:23,000 --> 00:09:27,000
Generally those target sequences are about six nucleotides long.
128
00:09:27,000 --> 00:09:36,000
Any given six nucleotide sequence would, by pure chance, appear thousands of times in any large complex genome like a plant or animal genome.
129
00:09:36,000 --> 00:09:42,000
So restriction enzymes are not a useful tool if your goal is to try and engineer genomes.
130
00:09:42,000 --> 00:09:51,000
So as the 70s became the 80s and then the 90s, more and better information became available about the DNA sequences of all sorts of interesting genes.
131
00:09:51,000 --> 00:09:58,000
And a lot of that interest was focused on human genes that may play a health role, perhaps mutated versions cause no diseases.
132
00:09:58,000 --> 00:10:02,000
But if you could repair that mutation you could cure the disease.
133
00:10:02,000 --> 00:10:07,000
And there was lots of interest in plant genes that could be tweaked to create improved crop varieties.
134
00:10:07,000 --> 00:10:14,000
So there was a real desire to have some way to go in and do genome engineering.
135
00:10:14,000 --> 00:10:20,000
But this knowledge was somewhat useful because no tools were available that could actually facilitate this.
136
00:10:20,000 --> 00:10:27,000
So for those decades, you know, the 80s and 90s, a lot of money was poured into making use of the tools we did have.
137
00:10:27,000 --> 00:10:35,000
Mainly, namely to piece together engineer DNA molecules in a lab and then randomly insert that DNA somewhere into a target genome.
138
00:10:35,000 --> 00:10:37,000
That is the process we call transgenics.
139
00:10:37,000 --> 00:10:43,000
That is the process used to make all the GMO crops covered so far in this series.
140
00:10:43,000 --> 00:10:52,000
And all of the GMO crops that are currently commercially available, at least outside Japan, to foreshadow the very end of this episode.
141
00:10:52,000 --> 00:10:56,000
But that whole time, so we're talking here about the 80s basically through to the 2010s.
142
00:10:56,000 --> 00:11:00,000
While funding was pouring into academic research to study and use transgenics,
143
00:11:00,000 --> 00:11:05,000
and while investors were happily funding companies to develop products using transgenics,
144
00:11:05,000 --> 00:11:11,000
the dream of a programmable DNA endonuclease and therefore of genome engineering did not die.
145
00:11:11,000 --> 00:11:18,000
And that's why when a new type of DNA binding proteins called tails were first described in a scientific paper in 2009,
146
00:11:18,000 --> 00:11:20,000
a lot of research groups took notice.
147
00:11:20,000 --> 00:11:30,000
I will spare you a lengthy treatise in this podcast, but out of pure vanity, I will link in the show notes a review article about tails, talons, and their uses in plant biotechnology.
148
00:11:30,000 --> 00:11:39,000
And how at the time we tried to spell out the possible advantages the talons still had versus the nascent CRISPR technology since we wrote that article in 2013.
149
00:11:39,000 --> 00:11:42,000
Needless to say, CRISPR won that argument.
150
00:11:42,000 --> 00:11:48,000
The cliff notes are that tails are large proteins, and most of their bulk is a set of repeating units,
151
00:11:48,000 --> 00:11:56,000
and each of those repeating units binds to one DNA base, and there is a single amino acid that differs between each of the repeating units,
152
00:11:56,000 --> 00:12:07,000
and that gives a specific DNA binding preference to that repeat using a specific code that links that amino acid, that variable amino acid, to a specific DNA base.
153
00:12:07,000 --> 00:12:13,000
So that variable amino acid is an iso-lucine, the repeat will bind to an A in the DNA.
154
00:12:13,000 --> 00:12:19,000
If it's a glycine, it will bind to a T. If it's an aspartate, then it will bind to a C like that.
155
00:12:19,000 --> 00:12:25,000
So say you wanted to target a 15 nucleotide DNA sequence that appears in a junior interest in,
156
00:12:25,000 --> 00:12:31,000
you just have to go to your lab freezer where you have stocks of different DNA pieces that encode tail repeats,
157
00:12:31,000 --> 00:12:48,000
each of which have been pre-modified to have molecular cloning sites attached to their ends that allow you to assemble them all into a full length tail gene in a one-pot reaction given the right mix of temperatures and enzymes to assemble everything together into a large DNA vector called a plasmid.
158
00:12:48,000 --> 00:12:53,000
And then you spend about a week sifting through the different reaction products to find one that was correct.
159
00:12:53,000 --> 00:13:04,000
So basically as long as you've done a lot of upfront work to get all the pieces prepared for this process, you could create a totally new designer tail or detail as they're known in about a week.
160
00:13:04,000 --> 00:13:16,000
That was a sensation, that was newsworthy and it was investment worthy with new companies bringing up around the technology and established companies spending, sending teams of lawyers to negotiate a license to use the technology.
161
00:13:16,000 --> 00:13:25,000
But so far what these proteins had evolved to do over millions of years was just to act as transcription factors, not as programmable in their nucleases.
162
00:13:25,000 --> 00:13:35,000
The TA in tail is short for transcription activator. Naturally, tails are delivered by invading bacteria into plant host cells where they then navigate to the nucleus,
163
00:13:35,000 --> 00:13:45,000
and bind to the specific gene that has a DNA sequence matching the pattern of repeats that make up the most of the bulk of the tail.
164
00:13:45,000 --> 00:13:53,000
Then when it has found its target gene, it will recruit RNA polymerase to initiate transcription the first step in gene expression.
165
00:13:53,000 --> 00:13:57,000
In other words, it will activate the gene switching on gene expression.
166
00:13:57,000 --> 00:14:14,000
And the most famous example of this, what it does is help the bacteria by switching on production of a specific gene that encodes a sugar transport protein, the pump sugar out of the plant cell into the surrounding extracellular matrix where the bacteria are feeding and replicating.
167
00:14:14,000 --> 00:14:19,000
Essentially forcing the plant cell to feed the bacteria that are attacking it.
168
00:14:19,000 --> 00:14:25,000
So yeah, tails naturally are transcription activators switching on genes is useful to some extent.
169
00:14:25,000 --> 00:14:31,000
So just being able to rearrange natural tails to make designer tails is useful.
170
00:14:31,000 --> 00:14:37,000
But as I mentioned, it was really programmable and on nucleases that were seen for decades as the real prize.
171
00:14:37,000 --> 00:14:41,000
But now I should explain why. What is actually so useful of being able to cut DNA?
172
00:14:41,000 --> 00:14:57,000
Well, it's not actually the cutting of the DNA that's so interesting. In fact, if you just cut DNA inside a cell, really cut all the way through both strands severing a link in the chain of the long DNA molecule called a chromosome, that is actually fatal for the cell. It cannot be tolerated.
173
00:14:57,000 --> 00:15:09,000
And so cells in all organisms have a DNA repair kit in the form of a range of enzymes that can work together to carry out a few different types of repair, healing that break.
174
00:15:09,000 --> 00:15:17,000
The simplest approach in terms of the number of DNA parts involved is called non-homologous and joining or NHEJ.
175
00:15:17,000 --> 00:15:29,000
This literally means sticking the two ends together and in practice, this involves either chewing back the DNA a bit to create very short matching DNA sequences that can be stuck together.
176
00:15:29,000 --> 00:15:43,000
Or just chewing back one strand to find an A and then you chew back the other to find a T to give you an example. And then you can stitch those together using homologous base pairing, which if you remember is A to T and G to C.
177
00:15:43,000 --> 00:15:50,000
Alternatively, a few DNA bases can be added randomly at the break site in the same way, creating a tiny piece of overlap.
178
00:15:50,000 --> 00:15:59,000
So for example, a G on one strand and a C on the other that can bond together to form a GC bond, which can ultimately heal the entire double strand break.
179
00:15:59,000 --> 00:16:15,000
We describe non-homologous and joining as error prone, meaning that will heal the DNA cut site, but it will almost always change the DNA sequence in the process, introducing short deletions or insertions of random DNA when compared to the original sequence before you make the cut.
180
00:16:15,000 --> 00:16:31,000
That is actually very useful from a biotinology perspective, because if you target a gene with one of those cuts, then the process of healing the cut will introduce mutations into the gene that will in most cases disrupt the function of the gene, essentially breaking the gene.
181
00:16:31,000 --> 00:16:35,000
So let's give a real odd example of why that can be useful.
182
00:16:35,000 --> 00:16:49,000
Sickle cell disease is a brutal genetic disease. It mostly impacts people of West African descent, and it is caused by mutation in the gene encoding hemoglobin, the crucial oxygen carrying protein in our red blood cells.
183
00:16:49,000 --> 00:17:02,000
More specifically, it is a mutation in the gene encoding adult hemoglobin. There is a separate gene in our genomes that encodes a version of hemoglobin, expressed wireria in, while we are in the womb, that's fetal hemoglobin.
184
00:17:02,000 --> 00:17:12,000
In adults, that fetal hemoglobin is not produced because the gene is switched off by a transcriptional repressor protein.
185
00:17:12,000 --> 00:17:22,000
As long as you can get into the blood producing bone marrow cells, you can use a programmable endonuclease to make a cut in the gene that encodes that transcriptional repressor protein.
186
00:17:22,000 --> 00:17:29,000
Then you can let non-homologous end joining, do its thing to mutate the gene and basically break it.
187
00:17:29,000 --> 00:17:37,000
Now, without that transcriptional repressor, the fetal hemoglobin gene switches back on again, even in adult bone marrow cells.
188
00:17:37,000 --> 00:17:44,000
And now the bone marrow cells start producing fetal hemoglobin alongside the dysfunctional adult hemoglobin.
189
00:17:44,000 --> 00:17:50,000
That's not enough to totally remove sickle cell disease symptoms, but enough to vastly improve them.
190
00:17:50,000 --> 00:18:00,000
And that is exactly the mechanism behind the first and only CRISPR-Cas9-based gene therapy currently available in the US, a therapy sold under the brand name Cass Gavie.
191
00:18:00,000 --> 00:18:12,000
And if you think about crop plants, one of the big uses would be to mutate, or as is generally said in plant genetics, knock out a gene that encodes a protein that is targeted as a receptor of pathogens.
192
00:18:12,000 --> 00:18:22,000
It is a common feature of pathogens in all organisms that they need something to latch onto so they can stay attached to the host cell long enough to do their thing.
193
00:18:22,000 --> 00:18:31,000
This is also true of viruses. For example, the SARS-CoV-2 virus that causes COVID-19 latches onto a human protein called ACE2.
194
00:18:31,000 --> 00:18:37,000
The same basic mechanism is crucial for bacterial pathogens, fungal pathogens and so on.
195
00:18:37,000 --> 00:18:45,000
There has to be some host molecule, usually a protein, and usually one that sticks out of the cell membrane for the pathogen to latch onto.
196
00:18:45,000 --> 00:18:56,000
In all cases, these host proteins, what we often refer to as pathogen receptor proteins, did not evolve to offer an attachment site per pathogen that would make no sense.
197
00:18:56,000 --> 00:19:06,000
They have some other natural function for the cell, but from the perspective of a farmer or plant breeder, whatever purpose that pathogen receptor protein actually serves in the normal healthy function of the plant,
198
00:19:06,000 --> 00:19:16,000
is probably outweighed by the damage done by the pathogen being able to use or co-opt that protein to invade the plant's cells and ultimately destroy a field of the crop plant.
199
00:19:16,000 --> 00:19:25,000
So being able to knock out a pathogen receptor protein in plant genomes is a really desirable use of programmable endonucleases.
200
00:19:25,000 --> 00:19:32,000
But so far, I've only talked about one of the possible repair pathways for the cut produced by a DNA endonuclease.
201
00:19:32,000 --> 00:19:42,000
There is another option that is even more exciting, homology-dependent repair or HDR, and this approach, a second DNA molecule is used as a repair template.
202
00:19:42,000 --> 00:19:57,000
And the real trick is that as long as the repair template has a good amount of perfectly matching DNA sequence at its beginning and end, the middle can be different to what was originally in the targeted DNA molecule where you made the cut.
203
00:19:57,000 --> 00:20:08,000
It's kind of a Trojan horse approach, trick the cell into thinking that the repair DNA you've given it is just a perfect match that it can use to close the DNA break that it needs to repair.
204
00:20:08,000 --> 00:20:12,000
But in the process, it is actually introducing a mutation.
205
00:20:12,000 --> 00:20:26,000
But in this case, the mutation would not be random, it would be a specific DNA change designed in a lab and built into the repair template DNA that is delivered into the cell alongside the programmable endonuclease.
206
00:20:26,000 --> 00:20:35,000
This approach, where you can change the DNA sequence however you want to, is called gene editing or done at scale genome editing.
207
00:20:35,000 --> 00:20:44,000
Sometimes gene editing is also used as a term to describe the random targeted mutations, targeted mutations described previously.
208
00:20:44,000 --> 00:20:54,000
But most of the time, if you hear the term gene editing, it means these specific DNA changes introduced at specific sites in a given genome using homology-dependent repair.
209
00:20:54,000 --> 00:21:01,000
These DNA changes are technically mutations because that is a word used for any change to a DNA sequence.
210
00:21:01,000 --> 00:21:10,000
But one of the uses of gene editing that is most exciting in terms of human health is to use it to fix inherited genetic mutations that cause disease.
211
00:21:10,000 --> 00:21:22,000
So to connect back to sickle cell disease, it would be seemingly much better, more elegant and better for the patient if you could just go in and fix the mutation in the adult hemoglobin gene
212
00:21:22,000 --> 00:21:28,000
rather than this convoluted approach of breaking a transcriptional repressor to allow expression of fetal hemoglobin.
213
00:21:28,000 --> 00:21:33,000
So you might think, well why wouldn't you just do that? Why don't just go do gene editing? Why do they use that convoluted approach?
214
00:21:33,000 --> 00:21:43,000
Well here's the catch. Gene editing is actually a lot harder to do because cells generally prefer the approach of non-homologous end joining over homology-dependent repair.
215
00:21:43,000 --> 00:21:52,000
And because at a technical level, it is difficult to get that repair template DNA into the cells where you need it to be in order for gene editing to work.
216
00:21:52,000 --> 00:21:59,000
So that pours a little bit of cold water on the excitement on the new era of gene editing that CRISPR has ushered in.
217
00:21:59,000 --> 00:22:06,000
It is still something very difficult to do even if we have a perfect tool for at least the first step in gene editing.
218
00:22:06,000 --> 00:22:20,000
But back in 2009, even that first step was basically non-existent because we had no programmable endonucleases, apart from the fairly clunky zinc fingers, to be able to make the DNA cuts that were the first step in gene editing.
219
00:22:20,000 --> 00:22:30,000
So that's why when tails appeared on the scene, multiple research teams scrambled to work out how to turn those tails into talons, that's the nuclease version,
220
00:22:30,000 --> 00:22:38,000
and that was done at the DNA level by fusing a non-specific DNA nuclease domain onto the end of the tail gene.
221
00:22:38,000 --> 00:22:45,000
Basically take this thing that doesn't naturally have the ability to cut DNA and add on an extra domain that will allow it to do that.
222
00:22:45,000 --> 00:22:53,000
Much to my dismay, that was successfully demonstrated just a few months into my PhD, but not by me.
223
00:22:53,000 --> 00:23:03,000
Deflated, I moved on to a different project that was more linked to the natural role of tails in causing devastated crop diseases called bacterial spot and bacterial wilt disease.
224
00:23:03,000 --> 00:23:12,000
But just a couple of years later, while I was midway through my new project, I've read a research update that may be very happy that I hadn't hitched my academic success to talons.
225
00:23:12,000 --> 00:23:21,000
That update was a description in probably Nature or Science magazine describing recently published research into a system called CRISPR Cas9.
226
00:23:21,000 --> 00:23:32,000
And after spending just a minute or two reading and then cross checking the actual research article authored by the labs of Jennifer Dowdner and Amantiel Schump, Amantiel Schupanti, I knew that this was the future.
227
00:23:32,000 --> 00:23:35,000
Not just me, really everyone did.
228
00:23:35,000 --> 00:23:42,000
The paper is titled "A programmable dual RNA guided DNA and the nuclease and adaptive bacterial immunity" and I've linked it in the show notes.
229
00:23:42,000 --> 00:23:49,000
It is open access so you don't need to pay to read it, which is good because I think it's an important document for the history of humanity.
230
00:23:49,000 --> 00:23:59,000
If you've listened to many episodes of this podcast, you know I generally shy away from hyperbole and all the smoke and mirrors of the tech world, whether that's biotech or regular silica based tech.
231
00:23:59,000 --> 00:24:03,000
But CRISPR really was and is a world changing technology.
232
00:24:03,000 --> 00:24:07,000
I was not the only one to immediately say how huge this would be.
233
00:24:07,000 --> 00:24:14,000
CRISPR got a massive and rapturous reception from the entire biotechnology community and garnered headlines around the world.
234
00:24:14,000 --> 00:24:20,000
Dowdner and Schupanti got the Nobel Prize just eight years after publication of that paper.
235
00:24:20,000 --> 00:24:29,000
That sort of speed is very unusual and it's a testament to how quickly CRISPR has reshaped the field and for the first time, made gene editing feel within reach.
236
00:24:29,000 --> 00:24:41,000
CRISPR is the shorthand, so often used for CRISPR Cas9, the combination of the Cas9 protein and a guided RNA that together work as a programmable DNA and a nuclease.
237
00:24:41,000 --> 00:24:47,000
The Cas9 protein itself cannot do anything until it is brought into close contact with DNA.
238
00:24:47,000 --> 00:24:52,000
And then when that has happened, it will make a cut at that site in the DNA.
239
00:24:52,000 --> 00:25:02,000
To get the Cas9 protein to the DNA, it partners up with a guide RNA molecule that is a short molecule made of DNA's molecular cousin RNA.
240
00:25:02,000 --> 00:25:20,000
The guide RNA physically embeds within the Cas9 protein, but a 20 nucleotide region of the guide RNA called the spacer is exposed and is basically able to scan along the genome in the target cell until it meets an exactly complementary 20 nucleotide DNA sequence.
241
00:25:20,000 --> 00:25:31,000
At that point, once the guide RNA has found its matching DNA sequence, the Cas9 protein gets stuck in place long enough and securely enough to do its thing and make a difference.
242
00:25:31,000 --> 00:25:42,000
So all you have to do to get Cas9 where you want it is to modify those 20 nucleotides of the gene encoding the guide RNA.
243
00:25:42,000 --> 00:25:54,000
The Cas9 protein remains the same. And since guide RNA genes are so short, they are cheap and easy to chemically synthesize, meaning anyone can sit down and order one for maybe $100.
244
00:25:54,000 --> 00:26:06,000
You'll get your own guide RNA gene in just a few days. You can also use laboratory methods to modify that 20 nucleotide spacer region of the guide RNA if you happen to have a guide RNA gene already lying around the lab.
245
00:26:06,000 --> 00:26:09,000
That is quick and easy taking just a few days.
246
00:26:09,000 --> 00:26:23,000
So that is the huge advantage of CRISPR. It is quick and easy to make a guide gene, a guide RNA gene, that you can package with Cas9 to deliver into a cell to target Cas9's activity to any gene you want.
247
00:26:23,000 --> 00:26:28,000
And that all takes just a matter of days compared to the weeks required to create a talent.
248
00:26:28,000 --> 00:26:33,000
It may seem a minor logistical advantage, but it is one of several that all add up.
249
00:26:33,000 --> 00:26:42,000
Another advantage is that the targeting system is based on RNA, which uses the same molecular matching system as DNA, homologous base pairing.
250
00:26:42,000 --> 00:26:52,000
In practice, if you know what 20 nucleotide DNA sequence in a target genome you want to target, then that 20 nucleotide DNA sequence is actually the exact DNA sequence.
251
00:26:52,000 --> 00:27:12,000
You need to insert into your guide RNA gene as the spacer because that DNA will be transcribed to produce a complementary RNA that then can go search for its complementary DNA, which would be the same sequence as the original DNA used to produce the RNA.
252
00:27:12,000 --> 00:27:24,000
What this means is that it is incredibly easy to design guide RNAs, a lot easier than designing talents, which requires some nuance understanding of the protein to DNA code that underlies the function of talents.
253
00:27:24,000 --> 00:27:40,000
Almost no expertise is required to design guide RNAs. Almost immediately after CRISPR Cas9 was first described, free web tools popped up, the maker as simple as defining what gene you are interested in targeting, and it will spit out for you a whole set of guide RNA sequences you can use.
254
00:27:40,000 --> 00:27:50,000
And even companies offering to just send you the guide RNA gene that would do what you wanted to do as long as you tell it what gene you want to target.
255
00:27:50,000 --> 00:28:03,000
And in the process of designing the guide RNA, these tools will check against a computer database of the whole genome of the target organism to ensure that the guide RNA sequence being designed is totally unique within the genome.
256
00:28:03,000 --> 00:28:18,000
Another advantage is that CRISPR Cas9 just works incredibly well, generally better than talents, meaning it cuts more efficiently and it is about as good at avoiding mistakes where it accidentally cuts a similar but slightly different sequence, what we call an off target.
257
00:28:18,000 --> 00:28:24,000
Those off targets, they do happen and they are a concern. They tend to happen at very low rates.
258
00:28:24,000 --> 00:28:35,000
And after about 15 years of CRISPR being on the scene, modified versions of the Cas9 protein are available that have even better performance in regards to off targets.
259
00:28:35,000 --> 00:28:40,000
Another logistical consideration is that Cas9 is a single protein.
260
00:28:40,000 --> 00:29:00,000
What I have yet to mention is that to get tails to servers nucleases basically to be come talons, you have to rearrange them as a pair of proteins facing each other, each with a nuclease domain that cuts one DNA strand each, together creating the complete break in the DNA molecule.
261
00:29:00,000 --> 00:29:13,000
Each talon protein is big and bulky and it is a logistical problem to get big bulky proteins into cells, especially if you want to deliver that as protein, not as DNA encoding the protein.
262
00:29:13,000 --> 00:29:24,000
The advantage of delivering proteins directly into the cell is that it is then an absolutely zero risk that that endonuclease somehow gets integrated permanently into the cell.
263
00:29:24,000 --> 00:29:35,000
Proteins, by themselves, are inherently unstable molecules within cells. There is no biological mechanism for a protein to become permanently integrated into a cell.
264
00:29:35,000 --> 00:29:49,000
Whereas DNA, depending on the delivery method, may well stably insert itself somewhere randomly into the genome of the target cell, making a permanent DNA change that is just a side effect of the gene edit you are actually trying to make.
265
00:29:49,000 --> 00:29:55,000
Keep that in mind because it is one of the keys to the biggest selling points of using CRISPR in crop engineering.
266
00:29:55,000 --> 00:30:02,000
Cas9 is a fairly bulky protein itself, but it is just one protein that you have to get into the cell, not two.
267
00:30:02,000 --> 00:30:10,000
And there are also modified versions of the Cas9 protein that are smaller and easier to deliver as a protein into cells.
268
00:30:10,000 --> 00:30:24,000
So CRISPR Cas9 was just like a faster, better, stronger version of talons. The felt in reach to really anyone with some basic knowledge of molecular biology, and scientists immediately saw all those advantages, took it and ran with it.
269
00:30:24,000 --> 00:30:34,000
Most of that metaphorical running originally happened in academic research labs using CRISPR Cas9 to modify genes as part of molecular genetic research programs.
270
00:30:34,000 --> 00:30:46,000
Academic researchers were free to take CRISPR Cas9 and run with it because they didn't need a license. As you might expect, CRISPR Cas9 as a tool for gene editing was immediately patented.
271
00:30:46,000 --> 00:30:51,000
However, even though the patents were immediate, they were unfortunately multiple.
272
00:30:51,000 --> 00:31:03,000
So what I haven't mentioned so far was that the Daoudnou Champagne team filed a patent once they published their discovery, but so too did a team at the Broad Institute,
273
00:31:03,000 --> 00:31:10,000
which is affiliated with MIT and led by a biotechnology professor Feng Zhang.
274
00:31:10,000 --> 00:31:21,000
You see, the Zhang lab had published a demonstration of the use of CRISPR Cas9 in eukaryotic cells just a few months after the original work by Daoudnou and Chappontier.
275
00:31:21,000 --> 00:31:28,000
A pretty classic situation of two researchers ending up doing a similar thing and publishing at almost the same time.
276
00:31:28,000 --> 00:31:43,000
Daoudnou and Chappontier were CRISPR Cas experts, but not so much biotechnologists. They did see the potential of their discovery and they tried to demonstrate the use of CRISPR Cas9 as a programmable endonuclease in their original work and they used that to file a patent.
277
00:31:43,000 --> 00:31:55,000
But Zhang and his lab were more experienced in recombinant DNA methods and had seen the possible applications of CRISPR Cas9 based on some research into natural CRISPR systems published earlier on in 2011.
278
00:31:55,000 --> 00:32:10,000
So they were basically working on a very similar idea that didn't really care about the natural CRISPR system, but they saw this idea of how you could use Cas9 as a programmable endonuclease and they published that just a few months after the Daoudnou Champagne paper.
279
00:32:10,000 --> 00:32:24,000
The result of this was a 10 year legal battle that was only resolved in 2017 actually in favor of Zhang and the Broad Institute, a decision that has been reaffirmed multiple times, including as recently as 2026.
280
00:32:24,000 --> 00:32:33,000
And as you might imagine, that legal battle has significantly held back commercial investment in an applications of CRISPR based technology.
281
00:32:33,000 --> 00:32:43,000
Despite that, many companies pressed ahead with R&D, even while they weren't really sure who they would have to pay to get a license to use the technology inherent to their approach.
282
00:32:43,000 --> 00:33:01,000
By now that obstacle is cleared and it's even easier and as I mentioned earlier, multiple variants of Cas9 have now been described from different bacterial species or created through deliberate modification of the Cas9 gene in labs, producing a range of Cas9 variants and alternatives that have their own patents.
283
00:33:01,000 --> 00:33:10,000
So anyone looking to use Cas9 or one of these variants commercially has a range of different options of potential partners to purchase a license from.
284
00:33:10,000 --> 00:33:20,000
So there are a range of options, but what actually are the products that people are trying to work on? Well, I mentioned human gene therapies, but this is a podcast about crops.
285
00:33:20,000 --> 00:33:26,000
What are crop scientists in universities, government research labs and companies trying to do with CRISPR?
286
00:33:26,000 --> 00:33:33,000
First, the general theoretical answer about what you can do with CRISPR that you can't do with general transgenics.
287
00:33:33,000 --> 00:33:46,000
Well, one thing that you can do is go in and alter genes that are in the plant, for example, edit just a few DNA bases to change the part of a gene that encodes the active site of an enzyme that might allow the plant to better tolerate drought or heat stress.
288
00:33:46,000 --> 00:33:55,000
So remember that there's now been decades and decades of what we call molecular genetic research into plants and their genes and genomes.
289
00:33:55,000 --> 00:34:05,000
So we know of all sorts of genes that could be kind of very powerful if you made a slight DNA change to them. Maybe you know that there's a wild relative that's more drought resistant.
290
00:34:05,000 --> 00:34:14,000
And if you could modify the version of the corresponding gene in the crop plan to mimic the version that's in that wild relative, you could probably get some boosted drought resistance.
291
00:34:14,000 --> 00:34:22,000
So there's all sorts of ideas like that floating around just waiting to be implementable with the right tools.
292
00:34:22,000 --> 00:34:41,000
That is not the sort of thing you could do by just inserting a new gene. You would just be layering something new on top of what was already exists inside the crop, which sometimes might be useful, but sometimes having both versions simultaneously expressed and functioning will kind of ruin the effect you're trying to achieve.
293
00:34:41,000 --> 00:34:48,000
But there is another very significant reason that crop scientists were immediately excited about CRISPR-Cas9, probably the biggest reason.
294
00:34:48,000 --> 00:35:04,000
If you can get the Cas9 and the Guida RNA, so that's as protein and RNA, not as DNA encoding those things, get them into Calis cells. Those are those uninduced, sorry, those are those induced plant stem cells that are then used to regenerate your whole plant.
295
00:35:04,000 --> 00:35:24,000
So if you can get that Cas9 protein and the Guida RNA into Calis cells and the Cas9 protein and Guida RNA can work together to get to a gene of interest, break that gene, or even better if you can get the Cas9 Guida RNA and a template DNA molecule into the cells and let them work together to edit the gene of interest.
296
00:35:24,000 --> 00:35:40,000
Well, after that gene edit, the cells will continue to divide. After a few hours, the Cas9 protein and the Guida RNA will be destroyed because proteins and RNA are for the most part short-lived molecules and cells are full of enzymes that will destroy RNA and proteins.
297
00:35:40,000 --> 00:35:52,000
One new ones are turned out to take their place, except of course the cells doesn't have the genes to make the Cas9 protein and the Guida RNA because they didn't come from the plant, they were literally injected or inserted into the cells.
298
00:35:52,000 --> 00:36:02,000
And the template DNA, it will also be destroyed because in general the template DNA that we use in this sort of experiment is a short linear DNA molecule.
299
00:36:02,000 --> 00:36:08,000
So that's kind of naked DNA, not the huge long protein package DNA molecules we call chromosomes.
300
00:36:08,000 --> 00:36:18,000
And those short linear DNA molecules, similar to RNA and proteins are inherently unstable in cells, they will be chewed up and destroyed within one or two cell generations.
301
00:36:18,000 --> 00:36:28,000
It's very unlikely, but in rare occasions that DNA template, sorry, that template DNA might get inserted into the genome of the plant cell.
302
00:36:28,000 --> 00:36:42,000
But that would happen very rarely and it would be very easy for the crop scientists to check for this and exclude any such rare occurrences from what would be the tens or hundreds of developing calli being created in parallel to be sure of getting at least one good one.
303
00:36:42,000 --> 00:37:02,000
So as the callus develops into an adult plant, the Cas9 protein, the Guida RNA and the template DNA are all destroyed. But what does remain is the edit in the target gene in the actual genome of the crop plant that is a permanent change that is passed down through each generation of cells and the developing plant and then to all the progeny of that plant.
304
00:37:02,000 --> 00:37:10,000
And what to make of those progeny of the gene editing experiment, at the genetic level, they contain no outside DNA.
305
00:37:10,000 --> 00:37:32,000
No genes that were not generated from within the plant itself as part of its evolutionary journey, just slightly tweaked versions of those genes, not really any different to the general genetic variation that can occur in the course of evolution and the more accelerated process of artificial selection that has been used for millennia to produce crop plants.
306
00:37:32,000 --> 00:37:41,000
Would that be considered a genetically modified organism? Well, many companies are banking on the answer to that question being no.
307
00:37:41,000 --> 00:37:52,000
What that would mean in practice is a huge reduction in costs as these crop varieties would be treated no differently to any other crop variety developed using non genome genome methods.
308
00:37:52,000 --> 00:38:05,000
That is to say, they would be assumed safe and require no regulatory approval at all before being sold. And they would be allowed to be sold in jurisdictions that currently totally banned the sale of GMOs such as most countries in the EU.
309
00:38:05,000 --> 00:38:16,000
Well, unfortunately, if you're hoping into a deep dive into all the regulation around CRISPR crops, you'll have to wait until season 2. This episode is already long enough, but I can give you a brief summary.
310
00:38:16,000 --> 00:38:35,000
In the last five years, countries around the world have taken this question very seriously and most have basically landed somewhere in the middle agreeing to relax regulations on crops created using CRISPR Cas9 genetic scene, especially if it's done in a way that leaves no trace of the Cas9 or Guida on agines in the plant as I was just describing.
311
00:38:35,000 --> 00:38:52,000
In the USA, the most relevant regulatory agency, the USDA, actually created a totally new regulatory framework that would have wiped away most regulation from crop varieties developed using CRISPR, but a judge in 2025 vacated that regulatory framework based on process issues.
312
00:38:52,000 --> 00:39:16,000
So that has added further confusion and delay that has to the original confusion delay caused by the patent battles. And so it's maybe not surprising that despite the immediate rush of excitement to make use of CRISPR as a tool going back more than a decade now, what I've been calling GMO 2.0, which is CRISPR gene editing or crops varieties created using CRISPR gene editing have not really hit the market.
313
00:39:16,000 --> 00:39:31,000
Despite that, so as far as I can work out, there are zero CRISPR genetic crop varieties for sale in the USA. There are several in the works though and there is a CRISPR gene editing crop for sale in one country, Japan.
314
00:39:31,000 --> 00:39:46,000
The Sicilian Rouge High Gabba Tomato, created and sold by Sun Attack Life Science, much like the purple tomatoes covered last episode, this is a variety modified to produce higher concentrations of chemicals with purported health benefits.
315
00:39:46,000 --> 00:40:02,000
So maybe to wrap up this episode, in episode 11, we met the first ever GMO crop, a tomato, and we can end here with I believe what is the first commercially available GMO 2.0 CRISPR gene edited crop, another tomato.
316
00:40:02,000 --> 00:40:21,000
That is it for season 1. I've had a blast putting it together and I've learned a lot, not just about the history of science and GMOs, filling in a lot of gaps in my knowledge as I researched each episode, but also about how to do audio editing and video editing as I've also started to do more video content, and how to generally create a podcast.
317
00:40:21,000 --> 00:40:32,000
I feel very far away from the standards that I would hope one day to achieve and I have a vague notion that some point in future will come back and redo this season, but really stepping everything up a notch.
318
00:40:32,000 --> 00:40:36,000
In the meantime, we will keep moving forward into new territory.
319
00:40:36,000 --> 00:40:50,000
Season 2 will be putting the basic science of GMOs a little bit more in the background and bringing the economic and social impacts of GMOs more into the spotlight, but there will still be a mixture of both, so don't worry science nerds, I won't let you down.
320
00:40:50,000 --> 00:41:00,000
But the thing is, I will have a lot of research to do before I feel ready to start writing out the episodes and so I'm going to give myself a pretty long break to do that.
321
00:41:00,000 --> 00:41:16,000
For now, I'm setting a target date of October 3rd, 2026, to premiere season 2. I'll aim to release some short episodes ever summer to let you know what I'm working on, and I'll be experimenting with more video content, so make sure to follow me on YouTube at ModifiedPod.
322
00:41:16,000 --> 00:41:23,000
So before I say goodbye for now, here are your five take home messages for this episode.
323
00:41:23,000 --> 00:41:30,000
Number 1. CRISPR-Cast systems are found in most bacteria and are immune systems that defend against viruses.
324
00:41:30,000 --> 00:41:41,000
Number 2. Cas9, together with a guide RNA, acts as a programmable DNA and a nucleus, which can make DNA cuts at target sites based on a 20 nucleotide spacer region in the guide RNA.
325
00:41:41,000 --> 00:41:54,000
Number 3. CRISPR as a tool for gene editing was first demonstrated in 2012 and it immediately generated excitement, however for a mix of reasons commercial products are now only just arriving on the market.
326
00:41:54,000 --> 00:42:08,000
Number 4. A DNA break will in most cases result in random mutations at the break site, but if a DNA repair template is present, it will sometimes result in the template sequence being inserted at the break site, this is called gene editing.
327
00:42:08,000 --> 00:42:18,000
Number 5. Gene edited crops facing uncertain regulatory environment currently, but in many countries they are more likely regulated than conventional transgenic crops.
328
00:42:18,000 --> 00:42:25,000
Well, season one has been a blast and it's time for me to get right to work on season 2. Bye!
329
00:42:25,000 --> 00:42:45,140
[ Silence ]
