CRISPR/Cas: History and Perspectives

Discovery of the CRISPR/Cas system revolutionized biology and biomedicine in the 21st century. Here we discuss the milestones in the development of CRISPR/Cas genome editing technology, from the history of discovery to current developments, including medical applications. Technical and ethical problems associated with the use of CRISPR/Cas for editing human embryonic genomes are also discussed.

INTRODUCTION Discovery of the system of clustered regularly interspaced short palindromic repeats (CRISPR) in prokaryotic genomes and the Cas proteins (CRISPR associated proteins) is one of the most groundbreaking events in modern biology. CRISPR are DNA regions in the prokaryote genome consisting of identical short repeats (30-40 bp) separated by unique spacer sequences of the same length; CRISPR-associated Cas proteins are located in the vicinity of these regions (Hille and Charpentier, 2016). Short palindromic repeats are quite common: CRISPR regions are found in the genomes of 50% of all known bacteria and 90% of archaea (Grissa et al., 2007;Hille et al., 2018); this may indicate their extreme importance for prokaryotes. In 2020, Emmanuelle Charpentier and Jennifer Doudna were awarded the Nobel Prize in Chemistry for their work on development of the CRISPR/Cas system for genome editing.
CRISPR/Cas research has progressed from the discovery of unusual and inexplicable repeats found by researchers in the genomes of various bacteria and archaea to the description of CRISPR/Cas involvement in prokaryotic acquired immunity and the use of this knowledge for targeted eukaryotic genome editing and other aims. Using CRISPR/Cas-based tools, researchers made breakthroughs in clinical trials and biotechnology companies have launched gene therapy trials for a range of diseases. The technology continues to develop rapidly, with great promise for further work in biology, medicine, bioengineering, biochemistry and other sciences. This review describes the history of CRISPR/Cas discovery and the use of the CRISPR/Cas technique for fundamental and applied research.

DISCOVERY AND DEVELOPMENT
OF THE CRISPR/Cas METHOD FOR GENOME EDITING The first unusual repetitive sequences were described in 1987 in the genome of Escherichia coli (E. coli) by a group of Japanese scientists led by Yoshizumi Ishino, who discovered that the 3' end of the gene Iap, whose products are responsible for the isoenzymatic conversion of alkaline phosphatase, contained "five highly homologous sequences of 29 nucleotides arranged as direct repeats with 32 nucleotides as spacing" (Ishino et al., 1987). They found no biological explanation for the presence and function of these repeats, and this work did not receive much attention from colleagues either: until 2007, Ishino's publication was cited 1-2 times a year.
The unusual repetitive sequences in prokaryotic genomes interested the Spanish researcher Francisco Mojica, who discovered them in the genome of the archaea Haloferax mediterranei in 1993 (Mojica et al., 1993); at that time, the scientist was only 30 years old. In 1995, Mojica and co-authors described in detail these "tandem repeats" as they called them, in the genomes of Haloferax mediterranei and H. volcanii: a 30 bp sequence with dyad symmetry repeated in tandem with insertions of unique sequences of 33-39 bp and extending over long stretches, 1.4 kb in the H. mediterranei chromosome and ~3 kb in the H. volcanii chromosome (Mojica et al., 1995). To understand the role of this DNA region, the Mojica's group decided to introduce an extra copy of this DNA region into H. volcanii by transformation with a recombinant plasmid containing a 1.1 bp fragment of tandem repeats. This resulted in a significant decrease of cell viability and unequal distribution of the genome among daughter cells (Mojica et al., 1995). The first hypothesis about the biological role of tandem repeats in prokaryotic genomes proposed their participation in the chromosome segregation during mitosis. Around the same time, similar repeats were described in the genomes of Mycobacterium tuberculosis (Groenen et al., 1993), Streptococcus (Hoe et al., 1999), cyanobacteria Anabaena sp. (Masepohlet al., 1996), Shigella dysenteriae, Salmonella typhimurium (Nakata et al., 1989) and other bacterial species. It has been proposed that these repeats might be involved in chromosomal rearrangements, recombination, or were landing sites for regulatory proteins (Nakata et al., 1989;Groenen et al., 1993), but these assumptions were not tested experimentally.
After defending his PhD thesis in 1995, Francisco Mojica worked as a postdoc at Oxford University and then, driven by his interest in cryptic repeats, he returned to Spain where he tried to set up his own research group to study "tandem repeats." At the time, he could not obtain research grants and was severely restricted in funding his work and building the infrastructure of his own laboratory (Mojica and Rodriguez-Valera, 2016). Despite the difficulties, the scientist continued his research. Mojica refocused on E. coli as a model organism, but reproduction of the experiments he had conducted earlier on H. volcanii did not yield the expected results: no clear phenotype of E. coli genome segregation defects was observed when an additional copy of the "tandem repeat" was introduced into its genome. Mojica's second hypothesis was that tandem repeats served as reference points for DNA binding to cellular structures (e.g. cell membrane proteins or soluble proteins). However, no repeat-binding proteins were found in E. coli cell extracts. The third assumption was that the repeats could affect the three-dimensional structure of DNA, but this was not confirmed: analysis of plasmid DNA showed that the introduction of the repeat arrays had no effect on its topology (Mojica and Rodriguez-Valera, 2016).
Francisco Mojica did not give up on his ambition to find the functions of the mystery repeats. Gradual advances in sequencing technology made it easier to find such structures in the genomes of other organisms and Mojica's colleague, César Díez-Villaseñor, created software to search for repetitive regions in prokaryote genomes. By 2000, Mojica's lab systematized the data on genomic repeats in 9 species of archaea and 10 species of bacteria and gave them a name: Short Regularly Spaced Repeats (SRSR; an acronym that also indicates alternation of spacers and repeats, because it can also be decoded as: Spacer-Repeat-Spacer-Repeat) (Mojica et al., 2000). This highly important work for science was published in a Micro-Correspondence two-page format (a consolation to modern scientists who are asked to shorten their papers by journal editors).
The group of Dutch microbiologist Ruud Jansen, who described tandem repeats in Mycobacterium tuberculosis and other prokaryotic species, named them SPIDR (SPacers Interspersed Direct Repeats) (Jansen et al., 2002a). To avoid further confusion in the rapidly growing subject, Mojica and Jansen jointly decided to replace the names: direct repeats, tandem repeats, SRSR, SPIDR and other variations of the names, with a simple, "crispy" (as R. Jansen himself aptly put it) and the name we know today, CRISPR (Fig. 1). In 2002, Jansen and colleagues also identified protein-coding genes located in the vicinity of the repeat loci which they named as CRISPR-associated (Cas) proteins (Jansen et al., 2002b).
Although the role of genomic repeats remained a mystery to scientists at that time, the widespread distribution of repeat sequences in various prokaryote species indicated their undoubted importance and fundamental cellular role. Having discovered Cas proteins, Jansen suggested that they were involved in DNA metabolism or regulation of gene expression in a genomic region functionally related to the CRISPR locus, as they had a structure similar to DNA helicases or exonucleases (Jansen et al., 2002b).
However, the final key to unraveling the functions of the CRISPR/Cas system was the discovery of the origin of unique intermediate spacers. The role of these sequences has long been a mystery. Francisco Mojica initially did not give them much importance, he wrote: "the name itself (spacer) hints at an irrelevant role in the repeats just separating the palindromes" (Mojica and Rodriguez-Valera, 2016). In the early 2000s, his research group continued to work on CRISPR repeats in E. coli. They routinely amplified CRISPR loci using PCR, sequenced them and compared them with sequences in publicly available nucleotide databases. And then one day the scientists were lucky: one of the queries yielded a matching sequence, they found that the sequence of the spacer is homologous to a piece of the genome of the E. coli bacteriophage. Gradually, Mojica accumulated data on other spacers that had similarities to sequences in bacteriophages or in conjugative plasmids (Mojica et al., 2005). It turned out that bacteriophages whose sequences were found in CRISPR spacers were unable to infect a spacer host cell but infected closely related strains lacking this spacer (Mojica et al., 2005). An eureka moment occurred and Mojica was the first to make the correct assumption that the role of the CRISPR system is to acquire immunity against for-eign DNA and that the locus itself is a "compartment for storing DNA chunks of invaders" (Mojica et al., 2005;Mojica and Rodriguez-Valera, 2016). It is worth mentioning again the difficulties of recognition that the authors of this revolutionary discovery encountered when trying to publish their findings (Lander, 2016). We cite these facts not in order to disappoint the reader in the objectivity of modern scientific publishing houses, although this would not be unreasonable in this case, but rather to console those scientists to whom this story gives hope of subsequent recognition of their work regardless of the initially negative reaction of the scientific community to their discovery. Realizing the importance of his scientific conclusion, Mojica submitted his article to Nature. In November 2003, the editors of Nature rejected the paper without even sending it to the reviewers: for unknown reasons, the editor stated that the idea for the paper was already known. In January 2004, the Proceedings of the National Academy of Sciences (PNAS) also decided that the paper lacked the 'novelty and importance' to send it for review. Molecular Microbiology and Nucleic Acids Research were the next journals to reject Mojica's manuscript. Desperate and worried that he may be outrun, Mojica sent the article to the Journal of Molecular Evolution where after a year (!) of reviews the article was finally published on February 1, 2005. Mojica recalled this period as follows: "Imagine you have something you know it's big in your hands and there's the possibility that another article that takes the originality of your work is published. I remember sending mail every month to the editor, saying 'please tell me if you're going to publish or not so we can submit to another journal'. I was in absolute despair" (Fernández, 2019).
In March and August 2005, two independent research groups from France described similar findings in Yersinia pestis and Streptococcus (Bolotin et al., 2005;Pourcel et al., 2005). The study of such an unusual model organism, Y. pestis, the causative agent of plague, was explained by the fact that the research group, headed by Gilles Vergnaud, was commissioned by the French Ministry of Defense and was developing methods to trace the source of potential biological weapons. Their unique collection of Y. pestis was obtained during the 1964-1966 Vietnam plague outbreak (Lander, 2016). Similarly to Mojica, Vergnaud faced the reluctance of journals to publish the patterns found: his paper was rejected by PNAS, Journal of Bacteriology, Nucleic Acids Research and Genome Research before finally being published in Microbiology on March 1, 2005.
The second French research group included our former compatriots, Alexander Bolotin and Alexei Sorokin, who studied streptococci in Paris. According to Sergei Kiselev, professor of the Vavilov Institute of General Genetics recalls, Bolotin "was approached by a large yogurt company at the time with a request to find out why they were no longer able to kill unwanted bacteria in the sourdough. Producers always used special viruses to suppress their life activity, but at some point bacteriophages stopped killing bacteria" (Vedeneeva, 2020). Bolotin was the first to describe the now famous Streptococcus nuclease Cas9 protein which was named Сas5 or Сsn1 at that time (Bolotin et al., 2005). In the course of their work, Bolotin and collaborators noticed that the spacers had a common sequence at the end: the protospacer adjacent motif (PAM) necessary for target recognition (Bolotin et al., 2005).
Interestingly, both the biological weapons and yogurt studies led to similar scientific conclusions. Both groups confirmed Mojica's hypothesis on the role of CRISPR in the formation of acquired immunity in bacteria. In 2006, a group of US researchers (from ex-USSR), Kira S. Makarova, Nick V. Grishin, Svetlana A. Shabalina, Yuri I. Wolf, Eugene V. Koonin analyzed all prokaryotic genomes available at that time and found several gene clusters corresponding to Cas proteins (Makarova et al., 2006). The researchers classified Cas into protein families and described their potential functional and structural features. Assuming that the CRISPR/Cas immune defense system operates on the principle of RNA interference, they analyzed the similarity of Cas proteins to proteins of the RNA interference system, but found no similarities. They nevertheless made a number of assumptions about the mechanism of CRISPR/Cas operation and how new spacers can be acquired.
Without direct experimental confirmation, these findings only supported the hypothesis but did not prove it unequivocally. However, the proof of the beautiful hypothesis of acquired immunity in prokaryotes did not wait long: in 2007 Philip Horvath's group in France demonstrated that after virus infection, bacteria integrated new spacers derived from phage genomic sequences; and removing or adding certain spacers modified prokaryotic cell phage resistance (Barrangou et al., 2007). They showed that Cas9 was a key protein required for the process by which the CRISPR system inactivated the invading phage (Barrangou et al., 2007). Once again, the discovery was facilitated by a request from the food industry: the authors of the article were at the time working for Danish food ingredient manufacturer Danisco (now DuPont) and were sequencing the genomes of bacteria used as starter cultures in the dairy industry to produce yogurt and cheese. They also sequenced bacteriophages that infect and destroy dairy cultures (Nair, 2017). Remarkably, since 2011 DuPont yogurt and cheese cultures have been 'vaccinated' against bacteriophages naturally using CRISPR (Nair, 2017). In 2008, John van der Oost's group, together with Kira S. Makarova and Eugene V. Kunin showed that it is the small RNA (CRISPR RNA, crRNA) transcribed from the CRISPR locus that binds to Cas proteins and guides towards the target DNA for immune protection (Brouns et al., 2008). The same year, Luciano Marraffini and Eric Sontheimer described the defense mechanism of bacteria against plasmid DNA (Marraffini and Sontheimer, 2008). They elegantly demonstrated that the target molecule of Cas proteins is DNA and not RNA. They were also the first to suggest that the CRISPR/Cas system could be used outside bacteria as a molecular tool: "From a practical point of view, the ability to direct specific targeted degradation of DNA containing any given target sequence of 24-48 nucleotides can have significant functional utility, especially if the system can function outside its native bacterial or archaeal context…. The main difference between a restriction-modification system and CRISPR interference is that the latter can be programmed with a suitable effector crRNA" (Marraffini and Sontheimer, 2008). Now the CRISPR topic has finally received the attention it deserved. New discoveries have appeared with snowballing speed: S. Moineau's group (Garneau et al., 2010) found that the Cas-crRNA complex cuts foreign DNA 3 nucleotides above the PAM; K.S. Makarova and colleagues developed a classification of CRISPR/Cas systems (Makarova et al., 2011); Francisco Mojica in 2010 described in detail how the E. coli CRISPR system is turned on and off and suggested a mechanism of the system (Mojica and Díez-Villaseñor, 2010). The final solution to the puzzle of the CRISPR/Cas9-driven natural immunity mechanism was found in 2011 by Emmanuelle Charpentier's group: in addition to sgRNA, there was a second small RNA, which the authors called transactivating CRISPR RNA (tracrRNA). The latter was discovered by nextgeneration sequencing, the benefit of tracrRNA being the third most abundant transcriptome after rRNA and tRNA. Researchers showed that tracrRNA formed a duplex with sgRNA and it is this duplex that directed Cas9 to its DNA targets (Deltcheva et al., 2011).
Once the mechanism of CRISPR/Cas was finally decrypted, the scientific research immediately switched to the practical application of this system, as scientists understood the enormous potential of the discovered mechanism. The study of CRISPR/Cas began to take place away from the host bacteria: the CRISPR/Cas was used as a targeted tool for targeted DNA cutting. In 2011, Lithuanian biochemist Virginijus Šikšnis, a graduate student in chemistry at Moscow State University, was able to transfer a functional CRISPR/Cas from S. thermophilus and express it in E. coli. This confirmed that the CRISPR/Cas system is capable of working autonomously and all its necessary elements (Cas9, crRNA and tracrRNA) were already known (Sapranauskas et al., 2011). In 2012, his group characterised the structure of Cas9 in detail, identified its nuclease domains, showed that crRNA can be shortened to 20 nucleotides, and that Cas9 targets could be changed by replacing crRNA (Gasiunas et al., 2012). The authors showed that the Cas9/crRNA/tracrRNA complex can cleave target DNA in vitro. Šikšnis submitted his paper to Cell on April 6, 2012, and the journal rejected the paper without peer review six days later, then he sent it to PNAS on May 21, where it was published on September 4 (Lander, 2016).
A technological breakthrough was provided by Emmanuelle Charpentier (E. Charpentier) and Jennifer A. Doudna, who reported that crRNA and tra-crRNA can be combined together to form a single synthetic guide RNA (sgRNA); this is currently the technology used by researchers who now need only two elements to program the targeted cutting of DNA: a Cas9 nuclease and a single guide RNA that guides the nuclease (Jinek et al., 2012). Like Šikšnis, Charpentier and Doudna demonstrated that Cas9 could cut purified DNA in vitro, and that the latter could be programmed by specially designed sgRNAs. patent application for the CRISPR/Cas9 gene editing system. Finally, in 2013, Feng Zhang, who had previously worked on TALEN (transcription activator-like effectors) programmable nucleases, was the first to successfully adapt CRISPR/Cas9 for genome editing in eukaryotic cells (Cong et al., 2013). He applied CRISPR/Cas9 to target editing of different genome loci in human and murine cells (Cong et al., 2013). Around the same time, similar results were published by George M. Church's laboratory (Mali et al., 2013). After publication, Feng Zhang also applied for a patent in his own name and was able to obtain one before Doudna.
NOBEL PRIZE CRISPR is considered nowadays the most important discovery in molecular biology since PCR, entailing the creation of the latest and most successful genetic engineering technology. Understanding the importance of this discovery, researchers and people interested in science have been wondering when the Nobel Prize for CRISPR/Cas would be awarded. Many feared that the technology was expected to make some sort of practical advance in order to finally be awarded the prize.
The 2020 Nobel Prize in Chemistry went to Emmanuelle Charpentier and Jennifer Doudna for developing the CRISPR/Cas9 gene editing technique. Firstly, this is of course a choice in favor of applied rather than basic science, as if the prize had been awarded for the discovery of CRISPR/Cas, it would certainly have been worth celebrating the merits of Francisco Mojica. Nevertheless, the wording of the Nobel Committee unequivocally speaks specifically of "method development." Secondly, a maximum of three people can be awarded the Nobel Prize for a single discovery, and awarding it to the two female researchers Charpentier and Doudna can be regarded as a kind of statement of the Nobel Committee. Before the award, part of the scientific community thought it would go to Charpentier, Doudna and Zhang (less frequently Šikšnis was named third), as those who had made major contributions to the application of CRISPR/Cas for genome editing in vitro and in vivo, in prokaryotic and eukaryotic cells.
Overall, in the era of collectivism in science, the justification of awarding Nobel Prizes to two or three people is increasingly questioned, as many discoveries are the achievement of a dozen scientists and their research teams.
CURRENT DEVELOPMENTS USING CRISPR/Cas One of the industrial sectors most interested in CRISPR/Cas is agriculture. Targeted editing of plant genomes is actively used to increase yields in fruit and grain crops, give them resistance to disease and changing weather conditions, and other desirable traits. For example, editing genes related to cytokine signaling has significantly increased yields in rice and wheat (Cong et al., 2013;Wang et al., 2014). Mutation of CLV (Rodríguez-Leal et al., 2017) and ENO (Yuste-Lisbona et al., 2020) genes responsible for meristem size has increased tomato yields. Using CRISPR/Cas technology, a variety of wheat with a reduced gluten content and 85% less immunogenicity has been created as it is important for people with celiac disease to control their dietary intake of gluten in wheat products. This result would have been almost impossible to achieve by classical crossbreeding methods, as gluten group proteins are scattered in about a hundred loci in the wheat genome (Sánchez-León et al., 2018).
Most genetically modified plants are still under development and are not used for mass cultivation, but there are many exceptions. By now, more than a hundred CRISPR/Cas-derived plant varieties have been officially approved. These include, for example, soybean with increased oleic acid content due to a mutation in the omega-6 fatty acid desaturase gene and wheat that is resistant to powdery mildew by editing the MLO gene locus that suppresses the plant's defense mechanisms against fungal infections (Wang et al., 2014).
Another promising area for the use of CRISPR/Cas in living organisms is editing the genomes of vectors of infectious and parasitic diseases to reduce their populations, or their ability to transmit a particular disease: for example, research is ongoing to induce sterility in malaria mosquitoes, which could lead to a significant reduction in malaria vector populations (North et al., 2020). This approach could help control disease transmission and protect ecosystems from species that are dangerous to humans.
Genetic engineering using CRISPR/Cas also offers unique medical applications; for example, the use of CRISPR/Cas technology solves transplant compatibility problems and other problems associated with the use of porcine organs in humans. These include the problem of porcine endogenous retroviruses pathological to humans. George Church's lab successfully applied CRISPR/Cas9 to inactivate all 62 copies of retroviruses in pig cells (Yang et al., 2015); this incidentally set a record for the number of modified loci in a single experiment.
Muhammad Mohiuddin form the University of Maryland, USA is working on immunocompatibility problems between pig organs and the human body in heart xenotransplant. His group has created pigs with a knockout of three genes responsible for carbohydrate antigen synthesis and human rejection of the pig heart (GGTA1KO, β4GalNT2KO, CMAHKO), a growth hormone receptor gene (GHR) knockout to inhibit pig heart growth in the human body, and the addition of six human genes: two anti-inflammatory (hCD47, hHO-1), two genes that promote normal blood clotting (hTBM, hEPCR), and two complement system regulator genes (hCD46, hDAF) (Goerlich et al., 2021). Heart transplant from genetically modified pigs has been tested in baboons and has shown good results (Goerlich et al., 2021). Thanks to the work of Mohiuddin's team, the high-profile achievement of this year was the world's first heart transplant of a genetically modified pig to a human on January 7, 2022 by M. Mohiuddin's team (Reardon, 2022;Jee, 2022).
In 2021, the results of ex vivo clinical trials using CRISPR/Cas9 in human cells were published. Haydar Frangoul's group performed haematopoietic stem cell editing to knockout the BCL11A enhancer in two patients with sickle cell anaemia and β-thalassaemia (monogenic diseases caused by mutations in the gene encoding the β-subunit of haemoglobin) (Frangoul et al., 2021). BCL11A is a repressor of haemoglobin γ-subunit (HbF) and its knockout results in increased HbF expression, which increases survival in patients with sickle cell anaemia and β-thalassaemia. Autologous haematopoietic stem cell transplantation of ex vivo corrected cells resulted in an increase in blood HbF concentration and relief of disease symptoms (Kaiser, 2020;Frangoul et al., 2021).
Currently, research on the CRISPR/Cas technology develops in several directions, including the development of effective ways to deliver CRISPR/Cas components into cells (as DNA, RNA or ribonucleoproteins) using biological (viruses, virus-like particles, cellular penetration peptides), chemical (liposomes, nanoparticles) and physical methods (electroporation, sonoporation, microinjection) (Taha et al., 2022); as well as the development of approaches aimed at improving the on-target and reducing the off-target mutagenic activity of Cas proteins (Nidhi et al., 2021). To minimize off-target editing, various software programs are being developed aimed at in silico prediction of off-target activity and selection of optimal and specific sgRNAs; chemically modified sgRNAs with higher specificity are used; improved variants of Cas are selected and constructed (Naeem et al., 2020).
A number of scientific groups in Russia are actively engaged in CRISPR/Cas research. Konstantin Severinov's laboratory from SkolTech is engaged in the prediction and functional characterisation of the new CRISPR-Cas system. Sergey Shmakov from this team has created a semi-automated search system that detects novel CRISPR/Cas systems (Shmakov et al., 2019). Using this software, members of Severinov's laboratory found and characterized the C2c2 protein (now Cas13a), unique in that it cuts RNA rather than DNA (Abudayyeh et al., 2016). Yana Fedorova from the same laboratory described a compact Cas9 orthologue with all the same functional features but reduced in size, making it easier to deliver to cells (Fedorova et al., 2020). Suren Zakiyan's Epigenetics of Development Laboratory from the Institute of Cytology and Genetics in Novosibirsk is successfully working on the creation of cellular models of various human neurodegenerative diseases (Medvedev et al., 2021), muscular dystrophies (Medvedev et al., 2021), including Huntington's disease (Morozova et al., 2018;Malankhanova et al., 2020), amyotrophic lateral sclerosis (Ustyantseva et al., 2019), and spinal muscular atrophy (Valetdinova et al., 2017) using CRISPR/Cas9 system. Maxim Karagyaur from the Laboratory of Biochemistry and Molecular Medicine at the Faculty of Fundamental Medicine, Moscow State University has been using the CRISPR/Cas technology to knockout genes of interest, to regulate gene expression at the epigenetic level, to model single-nucleotide polymorphisms to create cellular and tissue models to study tissue development and regeneration Rysenkova et al., 2018;Tyurin-Kuzmin et al., 2018;Dyikanov et al., 2019;Slobodkina et al., 2020;Rusanov et al., 2020). Other interesting works by Russian scientists include the development of a method for rapid sorting of cells with genetic modifications after CRISPR/Cas9 action based on a short peptide expression (Zotova et al., 2019) and the use of CRISPR/Cas9 to simultaneously create doublestranded breaks in different chromosomal locations to study the mechanisms of chromosomal translocations (Shmakova et al., 2019;Canoy and Vassetzky, 2021). Thus, the use of CRISPR/Cas genome editing technology in various fields and in different organisms, provides unique industrial and medical opportunities to improve the quality of life.

USING CRISPR/Cas TO EDIT GERMLINE DNA
Since the CRISPR/Cas system is a convenient tool for genome editing, almost immediately after the description of its use in eukaryotic cells, studies began on the possibility of its application to human embryos, both for the correction of pathological mutations and for fundamental studies of early human embryonic development. The effectiveness of this approach, offtarget mutagenicity, the editing efficacy (embryo mosaicism), and the possibility of subsequent embryo development have been studied (Ormond et al., 2017;Lea and Niakan, 2019). The sources of embryos in such works were unclaimed embryos from in vitro fertilisation procedures (Fogarty et al., 2017). It appears that knockout models in mice do not always accurately reflect the role of the genes under study in human embryonic development (Fogarty et al., 2017). Despite many attempts, accurate targeted editing of embryos (through DNA repair by homologous recombination mechanism) remains low, with deletions or insertions occurring after double-strand breaks in most cases, while mechanisms of early embryonic DNA repair remain poorly understood (Ma et al., 2017). For this reason, the use of CRISPR/Cas to correct pathological mutations in embryos is difficult. At RUSSIAN  the same time, the use of CRISPR/Cas for targeted mutagenesis and gene knockout also poses certain risks: apparently, double-stranded breaks introduced by Cas9 at a single locus result in deletions that can extend over several thousand bases, including in embryonic and progenitor cells (Kosicki et al., 2018). Despite significant progress in the field, there are a number of concerns about the clinical use of CRISPR/Cas in humans: questions about both the safety of the method and the ethics of its application are increasingly being raised. A precedent has been set in the case of He Jiankui, a researcher at Southern University of Science and Technology in Shenzhen (China), who used CRISPR/Cas to modify the genome of human embryos. Providing in vitro fertilisation services, He Jiankui's group suggested that a couple in which the husband was infected with HIV should make a genetic modification to the CCR5 gene in the embryos that would lead to the cells being resistant to HIV infection (Regalado, 2019). The natural CCR5Δ32 mutation is found in Europe and Western Asia, where its average frequency is around 10%, and homozygotes for the mutant allele do have resistance to HIV (Novembre et al., 2005;Lopalco, 2010). According to He Jiankui, who announced his experiment in November 2018, the genome editing was successful and resulted in the birth of healthy twin girls Lulu and Nana (Cyranoski and Ledford, 2018). It is worth noting that neither the plan nor the results of this work have been fully published or peer-reviewed by the scientific community. On November 29, 2018, Chinese authorities suspended He Jiankui's scientific activities and he was prosecuted for violating Chinese laws on human experimentation and for providing unlicensed medical care; in December 2019, the scientist was sentenced to three years in prison and a fine of three million yuan.
He Jiankui, in an attempt to publish the results of his high-profile experiment, submitted an article titled "Twin births after genome editing for HIV resistance" to Nature and JAMA, both journals rejected the article (Regalado, 2019). Interestingly, He Jiankui's coauthors included Michael Deem, a scientist at Rice University (Houston, USA). Despite claims that he did not consent to the publication of the data, an internal review was initiated against Deem, the results of which are classified, but as of 2021 he is no longer employed by Rice University.
He Jiankui's study has been heavily criticised by scientists for several reasons: (1) The claims made in the article were not supported by the data provided. Although the article stated that they were trying to replicate a frequent variant of the CCR5Δ32 mutation, in fact they were not: other mutations have been introduced into the CCR5 gene, whose role in providing HIV resistance had not even been studied in vitro. The never-published article did not provide evidence that the genetic manipulation did lead to HIV resistance, although this could have been tested before the embryos were implanted. Moreover, analysing the sequencing data cited in the article, one could see an extended region of multiple uninterpretable peaks in the sequencing data of the CCR5 gene (Regalado, 2019), suggesting that the embryos were mosaic, i.e. different cells had different mutations in this region, whose role was also unexplored.
(2) The parents of the twins may have been poorly informed about the nature of the experiment or may have agreed to the experiment under pressure. There are established protocols for in vitro fertilisation in HIV+ parents that reduce the risks of infection to the embryo or foetus to zero. In this regard, the genome modification procedure offered no medical benefit but introduced unnecessary risks that the couple may not have been aware of. Moreover, in China, HIVpositive people do not have access to fertility treatment and in vitro fertilisation, indicating that the couple may have been forced to experiment because it gave them the only chance of having a child.
(3) The claimed medical benefit of CCR5 deletion is questionable. Even if CRISPR is effective in making people resistant to HIV, it is unlikely to be widely used, especially in places where HIV epidemics are unfolding, such as southern Africa, due to its complexity, cost, need for constant monitoring and a variety of other reasons. It will probably take many decades of widespread use of genetic editing using CRISPR (assuming it is effective) to halt the HIV epidemic. Public health initiatives, education and widespread access to antiretrovirals are more logical and effective solutions to control the HIV epidemic.
(4) The side-effects of applying human genome editing are poorly understood and researchers led by He Jiankui have begun creating genetically modified live humans before they have fully grasped the implications of their edits. CRISPR/Cas technology is not 100% specific to a given gene and the insertion of a nuclease with sgRNA can lead to off-target mutations elsewhere in the genome. He Jiankui's team tested selected 3-5 cells from embryos at an early stage before implantation for off-target mutations and found a single nucleotide insertion in a non-coding region of the genome in one embryo. A key problem here, however, is that sequencing for mutations involves lysing the cells and isolating the DNA, meaning that the cells tested cannot be further used for fertilisation and may be different from the embryo from which they were taken. Conversely, embryos that gave rise to twins could not be fully screened for off-target mutations in each of the cells. For example, the use of CRISPR/Cas9 on sheep embryos together with preimplantation screening and selection of embryos with the desired mutation nevertheless leads to mosaicism in half of the fetuses (Vilarino et al., 2018). A recent study applying CRISPR/Cas9 to correct the EYS gene in human embryos found that on-target and off-target cutting of Cas9 can lead to complete or partial chromosomal losses (Zuccaro et al., 2020).
In conclusion, it must be reiterated that the genetic modification of human embryos should be regarded as a high-risk procedure and the mass introduction of CRISPR/Cas technology will require legislation to ensure that the technology is not used in violation of ethical rules.
In addition to editing of the human embryo genome, one can edit human germline cells (sperm and egg cells). The introduction of CRISPR/Cas9 into oocytes together with spermatogonial cells (as part of in vitro fertilization) has been proposed to improve editing efficiency and reduce the risk of embryo mosaicism (Ma et al., 2017), although there are rather few experimental developments in human germ cell editing. Genetic modification of spermatogonial stem cells has been tested in mice (Wu et al., 2015) and pigs (Webster et al., 2021).
Ethical and social guidelines regarding clinical editing of the human embryonic genome have been set out by the American National Academies of Sciences, Engineering and Medicine and the English Nuffield Council on Bioethics (National Academies of Sciences, 2017; Nuffield Council on Bioethics, 2018). In 2019, a call for a global moratorium on clinical use of germline DNA editing (in sperm, eggs or embryos) to create genetically modified children on technological, scientific, medical and ethical grounds was published by several leading scientists, including Emmanuel Charpentier, Eric Lander and Feng Zhang (Lander et al., 2019). In 2020, the Geneva Statement on the need for a course correction in the editing of the human inherited genome was issued (Andorno et al., 2020). Its authors insist that a global public consensus is required before any steps towards reproductive editing of the human genome can be initiated. The authors have included an explanation to the general public about the current misunderstandings surrounding genome editing, the mainstreaming of social issues, including equity, and the development of criteria for empowering the public to influence decision-making in this area.

CONCLUSIONS
Until 20 years ago, the CRISPR/Cas system was only known to a narrow circle of scientists working on the subject. The mysterious system of repeats and spacers was of interest only to a few microbiologists working with bacteria or archaea and did not seem like a major scientific breakthrough. Today, the technique has become the subject of intense scrutiny by the entire scientific community, including both specialists and historians of science, philosophers discussing ethical issues that have arisen, as well as the general public. CRISPR/Cas studies are actively receiving grant support, and articles are willingly published on it. It is worth remembering that only the undying interest of the discoverers (especially Francisco Mojica) has brought us to the level of knowledge where we are today.
Researchers in almost every biochemistry, molecular biology or cytology lab now use CRISPR/Cas for genome editing. The rapid development of the technique has brought society to a turning point: the possibility of editing human genomes has opened up before us. Nevertheless, the risks also exist. Is the technology completely safe for humans because of possible off-target effects? How can possible criminal and antihuman use of the technique be ruled out? The inability to test the side effects of CRISPR/Cas at the early stages of embryonic development raises an important moral and ethical issue: who will be responsible if a child is born with genetic abnormalities?
Other pressing questions arise: can CRISPR/Cas be used for non-lethal, treatable diseases? Can CRISPR/Cas be used at a population level? Will the introduction of this cutting-edge and expensive technology further divide the poor from the rich, giving the latter more advantages? Finally, could CRISPR/Cas be used en masse to "improve" human genetics?
It would seem that not so long ago, fantasy literature described how human phenotypic traits could be altered on demand, right down to the choice of eye or hair colour. Now it seems that this possibility is much less fantastic. But while eye and hair colour are polygenic traits that are difficult or even impossible to change using CRISPR/Cas, editing single genes seems like a solvable scientific task. The precedent of editing the CCR5 gene has caused deep reflection among an informed public, because in addition to the described resistance to HIV, the CCR5 mutation has been associated with improved memory and learning ability (Zhou et al., 2016). Will prospective parents want to improve the cognitive abilities of their unborn children through genome editing? All these important questions remain to be answered in the near future, and it is best to do so before embarking on human genome editing.

COMPLIANCE WITH ETHICAL STANDARDS
The authors declare that there is no conflicts of interest. This paper does not describe research performed by the authors involving humans or using animals as subjects.

OPEN ACCESS
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.