Writer’s note: this paper was co-written with Matis Ringdal in 2016 during my Master’s degree in Innovation Management. It is published following the 2020 Nobel Prize in Chemistry for Jennifer Doudna and Emmanuelle Charpentier.
Abstract
CRISPR-Cas9 is a revolutionary genome editing tool. It allows scientists to genetically modify any organism with unprecedented ease. Unlike its predecessors, it is relatively inexpensive and is less time-consuming. The CRISPR-Cas9 system is a two-component tool which is composed of a protein — Cas9 — that can cut DNA, and a guide molecule which can target specific regions of DNA that need to be cleaved or replaced. The potential applications of this tool are multifold. Firstly, it is an extraordinary system that can help scientists make considerable advances in biomedical research. For instance, it can help us better understand genetic diseases, thereby making it easier to find a cure or treatment. In addition, it can also help in the development of novel drugs. Besides academia and medicine, CRISPR can also be used in agriculture in order to increase both the yield and quality of crops, without the need for foreign DNA. It can also help to create new biofuels by modifying particular species of algae or other microorganisms. However, like with any biotechnological breakthroughs, CRISPR has raised some concerns regarding its use in humans. Indeed, many prominent scientists and innovators have warned about its potential use as a eugenics tool. Understandably, the fear is that this technology could be used to design a totally new generation of humans that would be so different from what we now know that it would disrupt our lifestyle and create a dystopian society.
I. Introduction
Earlier in his life, when he was deciding what career he wanted to pursue, Elon Musk — arguably the most important innovative entrepreneur of the 21st century — asked himself a simple, yet profound, question: “What will most affect the future of humanity?” He came up with five answers: space exploration (and colonization), sustainable energy, the internet, artificial intelligence, and the ability to reprogram the human genetic code, or genome. Musk hasn’t rested on his laurels since he identified these answers. Indeed, he has founded successful companies in all of these fields but one. Today, it is virtually impossible to nurture an interest in innovation without stumbling upon SpaceX, SolarCity, Tesla Motors, PayPal, or OpenAI — all of which were founded by Musk. Although the latter initially voiced some concern about artificial intelligence (AI), likening it to “summoning the demon”, he decided that the benefits of AI would outweigh its downsides. And this is how OpenAI, a non-profit research company, was born in December 2015. Given Musk’s interest in all the aforementioned fields, it leaves us begging the question: why, then, has he decided not to get involved in genetic engineering? When Tim Urban — who is the main contributor on Wait But Why — asked him the question, Musk responded that it wasn’t so much of a technical battle that was hindering him from getting into it, but rather a moral one. He calls it the Hitler problem. As he puts it, “Hitler was all about creating the Übermensch and genetic purity, and it’s like — how do you avoid the Hitler Problem? I don’t know.”
Today, many scientists around the world are facing the Hitler problem. Thanks to a new genome editing tool named CRISPR, a new era is now underway in molecular biology, and indeed in science as a whole. CRISPR allows scientists to target specific regions of DNA, and then replace or delete those regions with unprecedented ease. These regions are usually genes, which can therefore be turned on/off, or simply be replaced if they are dysfunctional. In other words, CRISPR is akin to a tool that could scan an entire document, look for a specific letter sequence (“xyz” for example), and then delete this sequence, or replace it with a sequence that has been chosen beforehand (say “abc”). In essence, it is like a search-and-replace function of a word processor.
CRISPR is composed of two components: the first one, a small molecule called RNA, is responsible for recognizing a specific sequence (it is thus called a guide RNA, or gRNA); the other one is an endonuclease called Cas9, a protein that is responsible for cutting the sequence. As a result, CRISPR is able to read the entire genome, which is 3 billion base pairs long in humans, and which resides in the 23 pairs of chromosomes within the nucleus of each of our cells.
Therefore, the enormous possibilities that this new tool offers should not be understated: its application could potentially have repercussions on various industries such as agriculture and transport, and is thus not limited to biomedical research (CRISPR could be the holy grail that modern medicine has been waiting for). It is undeniable that since Watson and Crick first described the double-helix structure of DNA in 1953, considerable advances have been made in the field of molecular biology. However, no discovery since then has had such an enormous impact as CRISPR has during these past few years — that is, since its first use as a genome editing tool in 2012. Yet, as with artificial intelligence, genetic engineering comes with its drawbacks. Elon Musk has pointed out the most obvious one, but there are others as well. Nonetheless, the substantial benefits that could be reaped from this scientific endeavor are far too important to ignore.
II. What is genetic engineering?
Before we address genetic engineering, it seems appropriate to focus on genes first.
Every single living organism has to carry information about its own phenotypic traits (i.e. its features) inherited by its parents, and that can be transmitted to its offspring. The information is encoded by specific sequences of molecules strung together called nucleotides, the basic units of DNA. Nucleotides could be considered as the alphabet making up an instruction booklet called the genome, which contains all the information stored in DNA. There are 5 main nucleotides: adenine (A), cytosine (C), guanine (G), thymine (T), and uracil (U). Three nucleotides strung together constitute a codon — which could be analogous to a word, punctuation, or space. In humans, DNA forms a double-helix (as shown below) and is therefore formed of two complementary strands. The complementarity occurs thanks to the pairing of nucleotides on opposite strands — A pairs with T or U, and C pairs with G. Two nucleotides paired together are known as base pairs, or nucleotide pairs. The human genome contains about 3.2 billion nucleotide pairs. The specific order of codons defines genetic information, like the specific order of words makes up a sentence (more specifically an instruction in this case). A specific region of DNA coding for a specific protein is called gene. A gene is therefore a set of instructions (codons that make sense and which are delimited by ‘punctuations’) that is used to make RNA, the intermediate molecule that forms the template for the production of proteins.
In essence, genetic engineering refers to a technique that enables scientists to modify the genome, and more specifically the genes.
For many years, mankind has been able to use biology in order to manipulate his environment for food production, agriculture and medicine. Thanks to the emergence of new fields in the 20th century, in particular molecular biology and genetics, considerable progress was made in biotechnology. For example, new techniques such as gene knockout were developed. Gene knockout, as the name suggests, involves targeting a particular gene in order to inactivate it. It is a typical example of genetic engineering. Genetic engineering, then, mostly involves artificial means of modifying the genome of organisms in order to obtain a desired characteristic. Though such techniques helped make progress in biomedical research, they were time-consuming, cumbersome, and mostly ineffective.
However, it all changed recently, thanks to a bacterium known as Streptococcus pyogenes. Indeed, we have good reasons to think that we are now living in the Golden Age of genetic engineering, for better or worse. The biggest leap in biotechnology was made: CRISPR-Cas9 was born.
III. What is CRISPR-Cas9?
CRISPR stands for clustered regularly interspaced short palindromic repeats and refers to a part of DNA containing short repetitions of base sequences, essentially found in prokaryotes (mostly bacteria and archaea). Cas refers to an endonuclease enzyme associated with CRISPR regions. This type of enzyme is able to recognize a specific DNA sequence and to cut it precisely. Therefore, we call this specific sequence a restriction site. The ability to cleave DNA is a well-known biological mechanism used by living systems in DNA repair when mutations arise. Cas9 is a subtype of the Cas protein family, and is an RNA-guided DNA endonuclease, which in simpler terms means that this protein will be escorted by a guide molecule (gRNA) to a targeted restriction site on a DNA molecule. The CRISPR/Cas9 system is an immune system used by some archaea and bacteria to fight against foreign DNA intrusions (usually viral DNA). Therefore, it is a system responsible for acquired immunity. Indeed, the system will cut and integrate foreign DNA inside the prokaryote’s chromosome, conferring an inheritable immunity to daughter cells. Therefore, daughter cells will be able to adapt themselves to the rapid changes triggered by a foreign genome. For instance, if a virus infects a microbe, the latter will quickly recognize the viral DNA via the guide RNA and it will cleave it with Cas9.
IV. How does the system work?
As we mentioned in the previous section, S. pyogenes uses the CRISPR-Cas9 system to detect and handle foreign DNA. It performs this detection by rolling down the foreign DNA and checking whether there is a complementarity with the 20 base pairs contained in the guide RNA. If there is a match between the guide and the viral DNA, Cas9 becomes involved and cleaves the invasive DNA.
By tinkering with the CRISPR-Cas9 system, scientists have been able to develop a powerful tool that can target specific regions of DNA and then cut those regions (usually genes). The gRNA — which has been modified to contain the complementary gene sequence — does the identification while Cas9 does the cutting.
The ruptures in DNA lead to gene inactivation or gene reparation. Indeed, there is a natural mechanism in living cells that help repair DNA. If the process goes well, it leads to gene reparation, but if an error occurs, it causes gene inactivation. Therefore, there is some kind of randomness involved as it all depends on the natural repair mechanism.
As we will see in the next section, such tools already existed before. However, the efficiency, rapidity and low cost of CRISPR-Cas9 create all new possibilities for mainstream use in genetic engineering.
The advantage of CRISPR-Cas9 lies in the ease with which scientists can choose and replace the guide RNA, the molecule responsible for targeting a specific gene. Subsequently, it can allow the replacement of a defective gene by a full-functioning one.
The CRISPR-Cas9 system is usually transfected to a target cell by a plasmid — a little circular DNA molecule separated from the chromosomal DNA and able to replicate by itself.
Once the plasmid carrying Cas9 has been injected into the cell, the guide RNA will permit a single or double strand break in the DNA. The break can also allow the insertion of a specific DNA sequence at the desired location (or locus) in the DNA if scientists also insert the corresponding nucleotides. The new nucleotides are incorporated into the genome, becoming a part of the cell’s genetic material.
V. What was used before?
TALENs (transcription activator-like effector nucleases) are artificial restriction enzymes used to cleave targeted sequences of DNA. They are the result of a fusion between a DNA binding domain, TALE, and an endonuclease.
A TALE domain, which is a protein obtained from Xanthomonas oryzae, is easily made and is able to approximately recognize any DNA sequence. Associated with an endonuclease, the construction can be injected into a cell in order to modify the genome. TALEN has been used to produce embryonic stem cells and induced pluripotent stem cells to deactivate some genes.
TALENs are rather controversial mostly because of the lack of precision involved, hence causing unwanted cleavages. These kinds of undesired cuts in DNA can lead to several problems, ranging from the inactivation of a gene to the dysfunction of cellular activity — which can potentially cause the death of the cell.
ZFN (zinc-finger nuclease) is another type of artificial restriction enzyme, also obtained from the fusion of a “zinc-finger” DNA binding domain and an endonuclease. Again, one of the main concern about this construction is the lack of precision regarding the cleavage. Indeed, the number of off-target cuts is relatively high. ZFN has been used for genetic engineering on stem cells and immunity cells.
TALEN and ZFN are both two artificial constructions which have been useful in genetic studies during the past decade. However, their acceptance by the scientific community was limited by the high costs associated, the lack of precision, and the high number of attempts required to obtain a correct mutation.
VI. Applications of CRISPR
Before CRISPR-Cas9, an entire PhD thesis could be written on the replacement of a single gene. Today, a gene can be replaced in a matter of minutes and with great precision thanks to the CRISPR system. Some of the most impressive applications that we can expect, or already that are underway, include immunization from pathogenic bacteria, virus inactivation, drug discovery or even biofuel production.
Biomedical research
The CRISPR-Cas9 tool allows a new and very efficient way to alter gene function. With this technology, many genes can be turned off and their role evaluated in a single experiment (in biology, we usually infer the role of a gene by measuring the effects of its inactivation). Actually, drug screening with CRISPR-Cas9 is highly efficient, especially when it comes to the understanding of drug mechanisms and patient stratification.
The genetic code is quite ruthless: a single error during repair can completely change the protein encoded by a gene, thereby making it inefficient. Protein synthesis can also be stopped in that manner. By modifying the genome, scientists are able to study and understand what happens to a cell when a gene or protein is defective, thereby understanding its function.
Also, 98% of the human genome does not code for proteins. Hence, scientists do not really know what most of DNA is used for. But they suppose that a big part of it is carrying important functions. CRISPR/Cas9 provides the hope that we will soon find out what these functions are. Most studies concerning human diseases and/or drug screening initially involve animal models before any clinical trials are made. These models are usually chosen for the similarities they share with humans in terms of physiology, but also for genetic reasons. However, genetically modifying animal models can be quite tricky and time-consuming. CRISPR/Cas9 lets scientists engineer more species in more complex ways. The increase in the kind of animal model used could seem derisory, but it could give another dimension to therapy research in the next few years. Indeed, having a higher number of animal models improves the possibility of using the most appropriate one in a given experiment.
Needless to say: the real revolution is in the lab. CRISPR/Cas9 offers one main advantage, and that is specificity. Targeting specific DNA sequences in the genome and editing it is just one way of using it. In a broader sense, this tool can be “hacked” in order to send particular proteins to precise DNA targets to toggle genes on or off, with the ultimate goal of engineering an entire biological pattern. One example of this “CRISPR hack” is the mutation of the Cas9 enzyme so that it still binds DNA at the specific site recognized by the guide RNA — but it doesn’t cut it. Instead, the hacked enzyme literally turns a gene on or off, without altering the sequence.
Moreover, there is a different repair mechanism that can fix a cut by using a DNA template. If scientists supply the template, they are able to edit the genome at any wanted site in the genome and replace it with any sequence of their liking (usually a functional gene or a dysfunctional one to study a disease).
Epigenetics
Epigenetics involves a wide range of external chemical compounds that affect DNA, but the DNA packaging proteins called histones. The latter control access to DNA, thereby allowing or disabling gene expression. Epigenetic phenomena are highly dependent on the environment and the molecules involved usually act accordingly whenever there is a shift in paradigm.
The significant role of epigenetics in the function of cancer-related genes has been highlighted with previous gene editing tools such as ZFP, but CRISPR-Cas9 could democratize the field of epigenetics even more.
Agriculture
We often hear about genetically modified organisms in a pejorative sense. But by disabling specific genes in crops, researchers could easily make really resistant strains of wheat and rice, for instance. Unfortunately, in some jurisdictions as the European Union, the introduction of foreign DNA into a plant’s genome could lead to the classification of such plants as GMO, making their acceptance by regulatory bodies difficult. With CRISPR-Cas9, this issue could be avoided because no foreign DNA would be introduced in crops.
Ecosystems
Gene manipulation cannot only play a role in issues such as health and food production, but also in the conservation of ecosystems. Indeed, this technology can rapidly and efficiently spread engineered genes in target populations.
In fact, altering the genome of an entire species could be an efficient way to end the spread of parasites by reinforcing the immunity of their host. This can potentially stop the propagation of deadly diseases affecting crops and animals. Making mosquitoes unable to carry Plasmodium (the parasite which causes malaria), or mice resistant to Borrelia (the bacteria causing Lyme disease), are incredible goals that CRISPR could make possible.
Moreover, CRISPR could be used to stop the spread of invasive species. By reducing the number of eggs produced by a mosquito, or decreasing the fertility of Red King crabs, an entire ecosystem balance could be reinstated in some cases.
This technique could be a fast and cheap way to get rid of threatening populations. However, we should be careful to avoid an unexpected chain of events: gene drive technology could potentially cause the extinction of an entire species.
Transport
The ability to create complicated biological circuits leads to the possibility to also convert, for instance, a cell’s metabolic pattern into a biofuel factory. This is certainly one of the main goals of synthetic biology. The use of CRISPR-Cas9 has actually allowed the use of a particular yeast strain to produce lipids and polymers that could be used as new precursors for biofuels. The latter would be less expensive and less toxic than anything on the market.
VII. History: From Intestinal Microbes to Revolutionary Genome Editing Tool
Discovery of CRISPR
In December 1987, during the early days of the genomic era, Yoshizumi Ishino and his team from the Research Institute of Microbial Diseases, located in Osaka, Japan, published the DNA sequence of the iap gene taken from the common intestinal bacterium Escherichia coli. In a similar way, thousands of labs around the world had begun to map the genes of species ranging from yeast to humans. In order to better understand how the iap gene functioned, the Japanese scientists also sequenced some of the DNA that surrounded it. The part of DNA surrounding a gene is usually non-coding (does not encode protein sequences), but is known to help gene function. When the data was examined, the team was surprised to find cellular structures that none of them recognized. These cellular structures were comprised of clustered repeats: particular sequences of base pairs that were repeated in the genome. Not knowing what to do of this strange phenomenon, they nonetheless took note of it, and wrote in the final sentence of their report, published in the Journal of Bacteriology, that the “biological significance of these sequences is not known.”
Around the same time, in 1993, Francisco Mojica, a microbiologist at the University of Alicante, had observed the same clustered repeats in Haloferax mediterranei, an archeal microbe with extreme salt tolerance that had been isolated from Santa Pola’s mashes. These repeated sequences, comprised of 30 bases, were roughly palindromic and separated by spacers of about 36 bases. This phenomenon had never been observed in microbes at that point. The 28-year-old Mojica was captivated and devoted the next decade of his career to unraveling the mystery.
In 2000, he called these repeats short regularly spaced repeats (SRSR) after he discovered them in some other 20 different microbes, including Mycobacterium tuberculosis and the plague bacteria Yersinia pestisis. In 2002, SRSR was renamed CRISPR after a suggestion by Mojica. Still, no function could be ascribed to the CRISPR system. Hypotheses put forward involved gene regulation, replicon partitioning, DNA repair, and other roles.
CRISPR: an Adaptive Immune System
Mojica finally struck gold in 2003. Indeed, after painstakingly extracting each spacer from his word processor and inputting each of them into the BLAST program for comparisons, he realized that CRISPR loci must encode the instructions for an adaptive immune system that protected microbes against specific infections. He came to that conclusion because he first noticed that one of the spacers matched the sequence of a P1 phage (a virus) that infected many E.coli strains. In addition, that particular strain carrying the spacer was known to be resistant to P1 infection. So, after going though 4,500 spacers in one week, Mojica found that two-thirds of the 88 spacers sharing similarity to known sequences matched viruses or conjugative plasmids related to the microbe carrying the spacer. In raptures, Mojica went on to celebrate with his colleagues but little did he know about the 18 months of hardships lying ahead: his paper was refused by Nature in November 2003, on grounds that the idea was already known, and then by the Proceedings of the National Academy of Sciences in 2004 because the paper was deemed to lack sufficient “novelty and importance.” Likewise, the paper was refused by Molecular Microbiology and Nucleic Acid Research. Finally, after a year of review and revision, the paper first appeared in the Journal of Molecular Evolution on February 1, 2005.
At about the same time, CRISPR was being studied at a rather unexpected venue: a unit of the French Ministry of Defense, a few kilometers south of Paris. It all started in the late 1990s when intelligence reports raised concerns about Saddam Hussein’s regime potentially developing biological weapons. Hence, the Ministry of Defense asked Gilles Vergnaud, a human geneticist who had trained at the Institut Pasteur, to shift his team’s effort to forensic microbiology in order to develop methods that could trace the source of pathogens based on subtle genetic differences. During his research — and with the help of his colleague, Christine Pourcel — Vergnaud discovered that the spacers located at the “front” end of the CRISPR locus in Y. pestis (obtained from a plague outbreak in Vietnam in 1964–1966) corresponded to a prophage present in the strain’s genome. Vergnaud and Pourcel thus concluded that the CRISPR locus serves in a defense mechanism, or as they put it, “CRISPRs may represent a memory of ‘past genetic aggressions.’” Like Mojica, Vergnaud had difficulties getting his paper published as it was rejected by the Proceedings of the National Academy of Sciences, Journal of Bacteriology, Nucleic Acids Research, and Genome Research. Finally, the paper was published in Microbiology on March 1, 2005.
Alexander Bolotin, a microbiologist who worked at the French National Institute for Agricultural Research, also published a paper in Microbiology in September 2005. In his paper, Bolotin was the first to speculate about how CRISPR conferred immunity, but his guess would prove to be wrong.
Experimental Evidence of CRISPR as an Adaptive Immune System
The pace of research quickened. In 2005, scientists working for the Danish firm Danisco (which would then be acquired by DuPont in 2011) tried to experimentally determine whether CRISPR conferred immunity to Streptococcus thermophilus, a bacteria used to make dairy products such as yoghurt and cheese. Indeed, understanding how certain strains of S. thermophilus protect themselves was of both scientific and economic importance. Philippe Horvath had first heard of CRISPR in 2002 at a Dutch conference on lactic-acid bacteria and had since then used it to genotype different strains of bacteria. He had set up his molecular biology lab in Dangé-Saint-Romain in western France, and asked the help of his colleagues, including Rodolphe Barrangou and Sylvain Moineau, to test his hypothesis. To do this, they infected S. thermophilus with two viruses (or phages). Most of the bacteria died, but those which survived had one property in common: they all contained CRISPR molecules to defend them. Indeed, the resistant strains of S. thermophilus had acquired phage-derived sequences at their CRISPR loci and the insertion of multiple spacers correlated with increased resistance. In addition, the scientists studied the role of two of the cas genes, cas7 and cas9, and discovered that cas9 was necessary for phage resistance. Indeed, the Cas9 protein was an active component of the bacterial immune system and presumably cut the nucleic acids of the virus. Finally, they noticed that immunity depended on a precise DNA sequence match between spacer and target (nucleic acids of the virus) as resistant phages carried single-base changes in their genome. Their paper was published in Science in 2007.
Programming CRISPR
John van der Oost, a researcher at Wageningen University in Amsterdam, first heard of CRISPR systems in 2005, when one of his collaborators, Eugene Koonin — an expert in microbial evolution and computational biology at the National Center for Biotechnology Information at the National Institutes of Health, introduced the subject to him. van der Oost and his team studied the CRISPR system of one strain of E. coli by inserting it in another strain of E. coli that lacked its own endogenous system. In doing so, they were able to biochemically characterize five Cas proteins. They found out that these proteins were necessary to cleave CRISPR RNAs (crRNAs). Then, they created the first artificial CRISPR arrays to target genes in a particular virus in order to demonstrate that the crRNAs sequences are responsible for CRISPR-based resistance. Ultimately, the results showed that it was possible to directly program CRISPR-based immunity in bacteria and also hinted that the target of CRISPR was the DNA of the virus, not its RNA.
CRISPR Targets DNA
In 2005, Luciano Marraffini, then a PhD student at University of Chicago, had seen the importance of CRISPR as an adaptive immune system. He later joined Erik Sontheimer’s lab at Northwestern University to pursue his post-doctoral work on CRISPR. In a clever experiment using a modified gene in a plasmid and the CRISPR system, the two scientists were the first to directly demonstrate that the target of CRISPR was indeed DNA. Marrafini and Sontheimer also suggested that CRISPR was essentially a programmable restriction enzyme, i.e. a molecule that could cut DNA at specific regions. They were the first to recognize that CRISPR could be use for genome editing, and they therefore filed a patent application in 2008 including the use of CRISPR to cut or correct DNA in eukaryotic cells. However, due to a lack of sufficient experimental demonstration, the patent application was eventually abandoned.
crRNAs Guide Cas9 to Cleave Double-Stranded DNA
In order to better understand how CRISPR cuts DNA, Sylvain Moineau continued his collaboration with Danisco. The role of the Cas9 nuclease as the cleaving agent was confirmed in his first experiments. By analyzing the target plasmids of CRISPR, Moineau and his team discovered a particular sequence, composed of 3 nucleotides and located upstream of the proto-spacer motif (PAM) sequence. Further analyses confirmed that viral DNA is cut in precisely the same position relative to the PAM sequence. The specific sequences where Cas9 cuts the DNA were then found out to be encoded by the crRNAs.
Discovery of trancrRNA
Emmanuelle Charpentier trained as a microbiologist at the Pasteur Institute and set up her second lab in Umea, Sweden, in 2008. In her attempts to identify new microbial RNAs, Charpentier sought the help of Jörg Vogel, whom she had met at the 2007 meeting of RNA Society in Madison, Wisconsin. Vogel had set up his first lab at the Max Planck Institute for Infection Biology in Berlin and later moved to Würzburg to lead a research center on infection diseases. While studying immunity in Steptococcus pyogenes, the two scientists discovered a novel small RNA that was transcribed from a sequence immediately adjacent to the CRISPR locus. This small RNA was later called trans-activating CRISPR RNA (tracrRNA). The structure of tracrRNA suggested that it hybridized with crRNA to form a product that would then mature by various processes (including cleavage by RNase III). In 2011, a study confirmed this notion, thereby showing that trcrRNA was essential for processing crRNAs, and thus for CRISPR function. In addition, trcrRNA was also shown to be essential for the Cas9 nuclease complex to cut DNA.
Reconstitution of CRISPR in a Distant Organism
While reading the 2007 Horvath-Barrangou-Moineau paper, Virginijus Siksnys — an expert in restriction enzymes at the Institute of Applied Enzymology in Vilnius, Lithuania — decided that he would try to reconstitute the CRISPR system in vitro. With the help of his collaborators, he tried to reconstitute the functional CRISPR system of S. thermophilus in a very distant microbe, E. coli. To their great joy, the team found that transferring the entire CRISPR locus was enough to confer immunity to E. coli. A critical milestone had been reached in the field: the CRISPR-Cas9 system was now characterized. The only components necessary and sufficient for a functional system were the Cas9 nuclease, crRNA and tracrRNA.
Studying CRISPR In Vitro
Thanks to bioinformatics, genetics and molecular biology, the CRISPR-Cas9 system was well understood. However, precise biochemical experiments were needed to dig deeper and to confirm the results in vitro. Building on their earlier results, Siksnys and his colleagues studied the Cas9-crRNA complex of S. thermophilus in a test tube. Their results matched those obtained in vivo by Moineau and his colleagues: the CRISPR system cuts DNA precisely 3 nucleotides from the PAM sequence. Most importantly, they also demonstrated that custom-designed spacers in the CRISPR array could enable them to target specific regions in vitro and thereby cut those regions which were chosen beforehand. Finally, Siksnys and his colleagues also confirmed that tracrRNA and crRNA were essential for Cas9 to cleave DNA in vitro, and that crRNA could be shortened to just 20 nucleotides and still maintain its function properly. They reported all their work in a U.S. patent application which they filed in March 2012. On April 6, 2012, Siksnys submitted his paper on CRISPR to Cell, but it was rejected six days later without external review. Siksnys decided to shorten his paper and sent it to the Proceedings of the National Academy of Sciences on May 21, 2012. On September 4, 2012, the paper was published online.
Around the same time, Charpentier had begun to biochemically characterize CRISPR with a colleague in Vienna. In March 2011, she gave a lecture on tracrRNA at the American Society for Microbiology meeting in Puerto Rico. There, she met Jennifer Doudna, a famous structural biologist and RNA expert at the University of California, Berkeley. Doudna had set up her first lab at Yale in 1994 but then moved to Berkeley in 2002 where she solved the structures of type 1 CRISPR systems of different microbes, including E. coli, by using crystallography and cryo-electron microscopy. Charpentier and Doudna decided to work together and they ultimately obtained the same results as Siksnys: Cas9 could cut purified DNA in vitro; crRNA and tracrRNA were indeed necessary for Cas9 to function; and Cas9 could be programmed with custom-designed crRNAs. However, Charpentier and Doudna also demonstrated a novel mechanism: crRNA and tracrRNA could be fused in vitro and function as a single RNA. This RNA was called a single-guide RNA (sgRNA) and this concept would be used widely in genome editing thereafter (especially after some modifications so that it could work efficiently in vivo). Charpentier and Doudna sent their paper to Science on June 8, 2012, and it appeared online on June 28 after being reviewed.
Clearly, both Siksnys and the Charpentier-Doudna pair recognized the biotechnological potential of the CRISPR system. The former declared that “these findings pave the way for engineering of universal programmable RNA-guided DNA endonucleases” while the latter mentioned “the potential to exploit the system for RNA-programmable genome editing.”
Genome Editing in Mammalian Cells
In September 2012, experts were still skeptical about the use of CRISPR as a potential tool to edit the mammalian genome, let alone humans’. The reasons were numerous: unlike microbes, mammalian cells have very big internal environments and their genomes are 1,000-fold larger, reside in nuclei, and are embedded in an elaborate chromatin structure. For all these reasons, the mammalian genome is rather inaccessible and requires a lot of fiddling in order to modify it. Previous attempts to transfer microbial systems had failed and efforts to use nucleic acids to target genomic loci had been problematic.
Feng Zhang, a young scientist at MIT’s Department of Brain and Cognitive Sciences and the Broad Institute, was the one who found a solution. While he was pursuing his Ph.D. in chemistry at Stanford, he worked with Karl Deisseroth — a neurobiologist and psychiatrist — and Edward Boyden to develop a revolutionary technique: optogenetics. This technique uses light to stimulate neurons which have been engineered to contain a specific light-dependent protein obtained from microbes. In other words, it allows scientists to activate modified neurons at will by using light. Today, optogenetics is widely used in the field of neurobiology. Still, Zhang was on the lookout for new techniques that he could add to his molecular toolbox.
In February 2011, Zhang went to a talk on CRISPR given by Michael Gilmore, a Harvard microbiologist. The next day, he flew to a scientific meeting in Miami but he could not get out of his room. Captivated by Gilmore’s talk, he decided to spend his entire time reading all the literature on CRISPR. When he got back to Boston, Zhang tried to create a version of S. thermophilus Cas9 to be used in human cells. He succeeded modestly by expressing Cas9 and an engineered CRISPR RNA that could target plasmids placed in human embryonic kidney cells.
Over the course of a year, Zhang tried to optimize the system. In doing so, he figured that using S. pyogenes Cas9 was more efficient than using S. thermophilus Cas9 because the former was more evenly distributed within the nucleus whereas the latter had a tendency to clump in the nucleolus — a specific part of the nucleus. He also tested different forms of tracrRNA to identify the one that would be more stable in human cells. In addition, he discovered that crRNA could be processed in mammalian cells despite lacking a microbial protein — RNase III. Zhang had a robust system consisting of three components — Cas9 (either from S. pyogenes or S. thermophilus), trancrRNA and a CRISPR array — by mid-2012.
In addition, Zhang also showed that the CRISPR system could be used to efficiently and accurately cause mutations in the human and mouse genomes by causing deletions. Furthermore, programming the CRISPR arrays with spacers matching multiple genes enabled simultaneous editing of these genes. Having read Doudna’s and Charpentier’s paper, Zhang also tried a two-component system, consisting of the sgRNA previously described in vitro studies, and Cas9. The sgRNA, which is obtained from a fusion of shortened RNAs, did not work well in vivo as it only cut a minority of loci with low efficiency. However, Zhang found that using a full-length fusion of the sgRNA solved the problem. On October 5, 2012, Zhang submitted a paper to Science reporting mammalian genome editing. The paper was published on January 3, 2013 and would become the most cited paper in the field. As a consequence, the reagents used were distributed by Addgene, a non-profit organization, in response to the 25,000 requests that emerged over the next 3 years.
At the same time, George Church, a senior Harvard professor who had collaborated with Zhang, also submitted a paper on genome editing in human cells. His paper appeared in Science on the same date as Zhang’s paper. Like Zhang, Church showed that full-length fusions of sgRNA worked better than the short fusions in vivo. Other short papers, notably by Keith Young and Jin-Soon Kim, were also submitted in late 2012 and appeared online in late January 2013.
CRISPR Goes Viral
In early 2013, Google searches for “CRISPR” escalated quickly and have not declined since. Within a year only, various studies showed that genome editing with the CRISPR system was possible in many organisms — including yeast, fruit fly, monkey, zebrafish, and fruitfly. Scientists around the world began to investigate CRISPR in a broader sense, thereby improving and extending the technology for genome editing. Besides the scientific interest, the potential use of CRISPR for applications in human therapeutics and agriculture also triggered a commercial interest in many. Biotech start-ups were born, respected newspapers like the New York Times took notice, and international ethic summits shifted their attention to this revolutionary tool.
VIII. Patents and commercialization
Like with any discovery, fights over licensing rights and patents are common. In the case of CRISPR, two main parties are embroiled in such a dispute: on the one hand, there are Feng Zhang, the Broad Institute, and M.I.T., while on the other hand, we find Jennifer Doudna, Emmanuelle Charpentier and the University of California. In the United States, patents are generally awarded to the first people to file — in this case Doudna and Charpentier. However, Zhang and the Broad Institute argued that Doudna’s patent application did not include any proof that CRISPR would work in complex organisms in order to treat and prevent diseases. As a consequence, Zhang was awarded the patent, but the University of California has requested an official reassessment. A ruling has yet to be issued. However, both Doudna and Zhang described the suit as “a distraction.” Indeed, the two scientists pledged to release all intellectual property to researchers without charge, and they have consistently done so. Thus, they are somewhat inclined to a movement known as open science. In addition, when researchers at the Broad Institute — along with many other similar institutes — create a new guide for CRISPR, they typically donate a copy to Addgene. As stated previously, Addgene stocks thousands of ready-made guides used for genome editing with CRISPR, and as a nonprofit repository, the organization also distributes those guides to any researchers seeking them. Other companies also manufacture parts required to modify DNA, and ordering such parts is almost as easy as ordering shoes online. One example is Integrated DNA Technologies, a company that easily delivers RNA guides to researchers. However, such companies do not provide the CRISPR system as a whole for genome editing.
Both Doudna and Zhang are involved in new companies that want to develop CRISPR technology as therapies, and many pharmaceutical firms and other profit-seeking enterprises are also on the lookout. For instance, venture-capital firms are fiercely competing with each other to invest millions in CRISPR research, and anyone who holds a patent would win the right to impose licensing fees, thereby making a fortune.
Doudna co-founded Caribou Biosciences in October 2011, and the company — located in Berkeley, California — is now a leader in its field, using the CRISPR-Cas technology to focus on research, industry, therapeutics, and agriculture. In November 2013, Editas Medicine was born, with therapeutics as its main focus. Its co-founders included Doudna and Zhang, but since the latter was awarded the patent, Doudna left the company. As a consequence, without her intellectual property rights, Editas is struggling to have the strongest technology on the market. At the same time, CRISPR Therapeutics, also focusing on therapeutics, was created in Basel, Switzerland. One of its cofounders includes Charpentier. Finally, Intelliza Therapeutics, which has obtained a license from Doudna (from her pending patent) was founded in November 2014 and also focuses on therapeutics. Both Editas and Intellia are located in Cambridge, Massachusetts.
Today, none of these companies has begun clinical trials, but they are expected to do so in less that three years. The race to the top has now started, and whoever develops the best CRISPR system will probably have a monopoly on the market thanks to licensing fees. It is no surprise, then, that the number of patent applications filed for CRISPR has skyrocketed. In the end, it might be more of a legal battle than a technological one.
IX. The future of genome editing and CRISPR
Genome editing is nothing new in biomedical sciences: scientists have been trying to change human DNA for decades, mainly to treat genetic diseases. The use of modified viruses for genetic engineering has long been the principal method used. But such therapies have never been widely accepted because of the huge costs associated as well as the poor results obtained — mainly due to off-target mutations. The consequences of such mutations can be fatal. For instance, there is the well-documented case of child developing leukemia in the early 2000’s after he received gene therapy. Sadly, the child subsequently died. However, with CRISPR-Cas9, a new era of genetic engineering is now underway but the fears associated to it have not disappeared.
“Gattaca” is a popular film from the 90’s which depicts a dystopian future in which genetic engineering, and its worst consequence — eugenics — are commonplace: segregation between natural-born babies and genetically modified ones is the norm. Today, it seems unrealistic that such a plot could become reality, thanks in part to ethical committees which help prevent such scenarios. For instance, the International Summit on Human Gene Editing was held at the U.S. National Academy of Sciences in December 2015. It included a select group of biologists, physicians and bioethicists who discussed the potential impacts of CRISPR-Cas9.
Recently, critics have been very vocal about a group of scientists at Sun Yat-sen University, located in China. Their study was called irresponsible and it was suggested that the scientists had violated an established code of conduct. Indeed, the group used CRISPR on eighty-six human embryos in the hopes of modifying the gene responsible for betathalassemia, a rare blood disorder. The Chinese scientists were not trying to create genetically modified humans, but merely tried to understand the disease. Also, the embryos that they used were triploid zygotes, meaning that no human could develop from them. Most experts suggested that the study was an ethical and important one, but the media reported it in a rather bad way. The Times, for instance, headlined the story as “CHINESE SCIENTISTS EDIT GENES OF HUMAN EMBRYOS, RAISING CONCERNS.”
In fact, as long as this tool provokes too many off-target mutations, it’s easy to consider that this technology is not reliable, and so that editing human embryos is not ethical.
Given the rapid progress made with CRISPR, most researchers assert that the question regarding the specificity and efficiency of this tool is not really “if it works” but rather “when it will work”.
Actually, these improvements will soon allow the possibility of safe human gene editing, thereby raising ethical questions. In fact, what we should be really concerned about is the modification of germ cells. By definition, germ cells can be passed down to offsprings, essentially introducing those changes into the human population. For instance, it would be really easy to remove or replace a morbid gene from spermatozoids or eggs with CRISPR. In this way, treating dreadful genetic diseases would be a reality.
But the major concern arises when parents wish to replace undesired characteristics instead of real diseases. Getting a smarter, taller, blue-eyed child is the kind of potential requests that could be asked, and the possibility of an unregulated laboratory proposing such services is indeed a worrisome thought. Moreover, if one is imaginative enough, one can easily see the other possibility: engineering humans so that they obtain abilities outside of our current range. Creating a new eye-color, a superior race, or a super soldier is the kind of science fiction scenario that could potentially become a reality.
It is no surprise, then, that Jennifer Doudna is calling for a temporary research moratorium on the use of CRISPR on germ-line cells until it is better understood. Researchers in the United States are already addressing the necessity for regulating human germ-line editing. At the end of 2015, the National Institutes of Health (NIH) still refuses to fund research proposals for CRISPR/Cas germ-line editing therapies.
Nevertheless, it has to be emphasized that this policy is not applicable to projects that are funded privately. The private use of CRISPR is what worries most scientists. The ease of use and low cost of CRISPR means that it could be used by a private company, a private lab or even a lone scientist who would have no ethical considerations whatsoever.
It is obvious that there is a need to carefully think about any new powerful technology. Open discussions with the involvement of the general public are essential. However, we should be careful to not overreact, as it has been the case previously. Recombinant DNA technology, gene therapy, in vitro fertilization, and organ replacements (implanting animal parts in humans) also caused disproportionate reactions. Yet, they were gradually accepted when their benefits were proven.
Obviously, the use of CRISPR for human therapies needs to be strongly regulated: ethical legislature will have to rapidly, but cautiously, set the exact limits of what its applications should be.
X. Conclusion
The story of CRISPR highlights the importance of scientific research — sometimes expanding over decades — and the unpredictable impacts that can emerge from it. Most notably, it demonstrates that scientific breakthroughs, especially in the life sciences, are rarely eureka moments. In fact, CRISPR/Cas9 could not have been engineered without the help of curious and passionate scientists, located in various parts of the world, and most often at the early stage of their careers. Throughout the accounts given by all these scientists, it is undeniable that they would not have stumbled upon CRISPR were it not for some kind of serendipity. In addition, the role of funding agencies, whether in the form of governments (like the French military) or industrial companies, should not be understated.
CRISPR/Cas9 is undeniably the best genome-editing tool that exists today. It is far from being perfect (constant improvements are being made every day to increase its efficiency), but it has disrupted the field of biomedical research in two ways: it is inexpensive and can be used with unprecedented ease (both in terms of time and level of skills required). Furthermore, its applications can be extended to other fields, including agriculture, transport, and food production.
Understandably, this revolutionary tool has raised concerns over ethical issues because of its potential to alter the human genome. However, fears of eugenics are far from being rational, as we simply do not have the capacity to modify humans nowadays — there are both technological and regulatory constraints involved. The risks always depend on the potential results. If CRISPR enables us to understand genetic diseases, then we will be a step closer to curing them. Likewise, other medical problems like HIV/AIDS and other infections could potentially be eradicated thanks to this new tool. All in all, then, CRISPR offers a path for a better future, as long as we remain cautious and use this technology responsibly.
References
Lander, E.S. (2016, January 14). The Heroes of CRISPR. Cell 164 (1–2): 18–28.
Ledford, H. (2015, June 3). CRISPR, the disruptor. Nature News. Nature (ed.). Retrieved from http://www.nature.com/news/crispr-the-disruptor-1.17673
Loria, K. (2015, June 11). Elon Musk doesn’t want to get into genetic engineering because he doesn’t know how to avoid ‘the Hitler problem.’ Business Insider UK. Retrieved from http://uk.businessinsider.com/elon-musk-doesnt-want-to-get-into-genetic-engineering-because-he-doesnt-know-how-to-avoid-the-hitler-problem-2015-6?r=US&IR=T
Regalado, A. (2014, December 4). Who Owns the Biggest Biotech Discovery of the Century? MIT Technology Review. Retrieved from https://www.technologyreview.com/s/532796/who-owns-the-biggest-biotech-discovery-of-the-century/
Specter, M. (2015, November 16). The Gene Hackers: A powerful new technology enables us to manipulate our DNA more easily than ever before. The New Yorker. Retrieved from http://www.newyorker.com/magazine/2015/11/16/the-gene-hackers
Travis, J. (2015, December 11). GENETIC ENGINEERING. Germline editing dominates DNA summit. Science 350, 1299–1300.
Ledford, H. (2015, June 3). CRISPR, the disruptor. Nature News. Retrieved fromhttp://www.nature.com/news/crispr-the-disruptor-1.17673
Raven, P. (2014) Biologie 3ème édition, Paris, De Boeck.
Reis, A. (2015) CRISPR/Cas9 and Targeted Genome Editing: A New Era in Molecular Biology. NEB expression. Retrieved from
Nemudryi, A. (2014) TALEN and CRISPR/Cas Genome Editing Systems: Tools of Discovery. The National Center for Biotechnology Information. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4207558/
Figure 1 & 2 : “Chapter 1: How Genes work.” www.publications.nigms.nih.gov. National Institute of General Medical Sciences. Retrieved 13–05–2016.
Figure 3 : “Gene Expression”. www.biosocialmethods.isr.umich.edu. Retrieved 13–05–2016.
Figure 3 : “Chapter 1: How Genes work.” www.publications.nigms.nih.gov. National Institute of General Medical Sciences. Retrieved 13–05–2016.
Figure 4 : “CRISPR Cas9 Genome Editing”. www.origene.com. OriGene. Retrieved 13–05–2016.
Figure 5: “CRISPR-Cas9: the new golden child of genetic engineering”. http://www.rcsismj.com. RCSI student medical journal. Retrieved 13–05–2016.
Supplementary Figure: Lander, E.S. (2016, January 14). The Heroes of CRISPR. Cell 164 (1–2): 20.
Appendix
Supplementary Figure: The Twenty-Year Story of CRISPR Unfolded across Twelve Cities in Nine Countries
Glossary
Base in DNA: a unit of the DNA. There are 4 bases: adenine, guanine, cytosine and thymine.
Base pair: one of the pairs of chemical bases joined by hydrogen bonds that connect the complementary strands of a DNA molecule or of an RNA molecule that has two strands; the base pairs are adenine with thymine and guanine with cytosine in DNA and adenine with uracil and guanine with cytosine in RNA.
Conjugative plasmid: a plasmid that can affect its own intercellular transfer by means of conjugation; this transfer is accomplished by a bacterium being rendered a donor, usually with specialized pili.
DNA: deoxyribonucleic acid, an extremely long, double-stranded nucleic acid molecule arranged as a double helix that is the main constituent of the chromosome and that carries the genes as segments along its strands: found chiefly in the chromatin of cells and in many viruses.
Loci/Locus : the chromosomal position of a gene as determined by its linear order relative to the other genes on that chromosome.
P1 Phage: P1 is a temperate bacteriophage that infects Escherichia coli and some other bacteria.
Phage: virus that infects and replicates within a bacterium.
Plasmid: a plasmid is a small DNA molecule within a cell that is physically separated from a chromosomal DNA and can replicate independently.
Replicon: a DNA molecule or RNA molecule, or a region of DNA or RNA, that replicates from a single origin of replication.
Replicon partitioning: mechanism that assures the stable transmission of replicons during cell division.
Spacer: region of non-coding DNA between genes.
DNA binding domain: an independently folded protein domain that contains at least one motif that recognizes double- or single-stranded DNA.
Gene Drive: genetic elements that can spread even if they reduce the fitness of individual organisms
