CRISPR timeline from 1987 to 2026 in 34 dated events: who found the repeats, the 2012 Cas9 papers, the Broad v. Berkeley patent fight, the gene-edited babies and the first approved therapies, with a map and sources. Every event names the people and the lab or court behind it.
Yoshizumi Ishino and colleagues at Osaka University were sequencing the iap gene of Escherichia coli when they found a string of short, identical DNA sequences separated by stretches of unrelated DNA. They described the pattern in the Journal of Bacteriology in December 1987. They could not say what it did. The repeats sat in the literature for years as a curiosity, and nobody yet had a name for them.
Francisco Mojica, a doctoral student at the University of Alicante, was studying Haloferax mediterranei, a microbe from the salt ponds of Santa Pola, when he found regularly spaced repeats in its genome. In 1993 he published the first characterization of a CRISPR locus. He kept working on the repeats with little funding for the next 12 years. By 2000 his team had found the same repeat structure in many unrelated bacteria and archaea, which suggested it did a job.
Ruud Jansen and colleagues at Utrecht University published a paper in Molecular Microbiology in March 2002 that used the acronym CRISPR, for clustered regularly interspaced short palindromic repeats. Jansen and Mojica had agreed on the name in correspondence beforehand. The same paper reported a set of genes that always sat next to the repeats. Jansen called them cas genes, for CRISPR associated. Every later CRISPR tool, including Cas9, takes its name from this paper.
Alexander Bolotin and colleagues at the French National Institute for Agricultural Research (INRA) studied Streptococcus thermophilus, a bacterium used to make yogurt and cheese. Its CRISPR locus contained a large gene for a protein with nuclease activity, the protein later named Cas9. Bolotin also noticed that the viral targets of the spacers shared a short sequence next to them. That sequence, later called the PAM, is the signal Cas9 still needs before it will cut. The paper ran in Microbiology in 2005.
Mojica compared the stretches of DNA between the repeats, the spacers, against public sequence databases and found that many matched viruses and plasmids. He proposed that CRISPR was an immune memory: bacteria stored pieces of past invaders to recognise them again. Several journals turned the paper down before the Journal of Molecular Evolution published it in February 2005. A French group led by Christine Pourcel reported similar findings in Microbiology the same year.
Rodolphe Barrangou, Philippe Horvath and colleagues at Danisco, a Danish food ingredients company, exposed Streptococcus thermophilus starter cultures to phages, the viruses that ruin dairy fermentations. Surviving bacteria had added new spacers matching the phage. Removing the spacers removed the resistance. The work, published in Science on 23 March 2007, was the first experimental proof of Mojica’s idea. It also showed that Cas9 was likely the only protein needed for interference. Danisco used the result to build phage resistant cultures for yogurt and cheese.
John van der Oost’s group at Wageningen University showed in E. coli that each spacer is transcribed and cut into a small CRISPR RNA, or crRNA. The crRNA leads a complex of Cas proteins to the matching invader sequence. The paper appeared in Science on 15 August 2008, with co-authors at Sheffield and the US National Institutes of Health. It established that CRISPR worked through RNA guides, the feature that would later make it programmable.
Luciano Marraffini and Erik Sontheimer at Northwestern University showed in Staphylococcus epidermidis that CRISPR blocked plasmids by attacking their DNA, not their RNA. The paper was published in Science in December 2008. In it they wrote that a system able to cut a chosen DNA sequence could be useful outside bacteria. It was the first published suggestion that CRISPR might become a genome editing tool. They did not have the protein that could do it.
Sylvain Moineau’s team at Université Laval in Quebec City, working with Danisco, showed in Nature in November 2010 that Cas9 makes a double-stranded break in invading DNA. The cut falls at a fixed position, three nucleotides upstream of the PAM. The paper also showed that Cas9 was the only protein needed for the cleavage. A precise cut at a predictable place is what a genome editor needs, and by late 2010 the parts were known.
Emmanuelle Charpentier, working between Umeå University in Sweden and the University of Vienna, sequenced the small RNAs of Streptococcus pyogenes. Her team found a second RNA, tracrRNA, that pairs with the crRNA and is needed both to make it and to steer Cas9. The paper ran in Nature in March 2011. Charpentier met Jennifer Doudna of UC Berkeley at a conference in Puerto Rico that month, and the two agreed to work out how Cas9 cut DNA.
On 25 May 2012 the University of California, the University of Vienna and Charpentier filed a provisional US patent application covering CRISPR-Cas9 and a single guide RNA. The application described the system in test tubes and claimed its use in any cell. The lack of an example in human or animal cells became the center of a patent dispute that ran for more than a decade. The universities and Charpentier are often called CVC in the court papers.
Martin Jinek, Krzysztof Chylinski, Charpentier, Doudna and colleagues showed that Cas9 can be pointed at any DNA sequence by changing its guide RNA. They also fused crRNA and tracrRNA into one synthetic single guide RNA, which made the tool simpler to use. Science published the paper online on 28 June 2012. All the experiments used purified proteins and DNA in a test tube. The paper said the system could be used for gene targeting and editing.
Virginijus Siksnys at Vilnius University had moved the S. thermophilus CRISPR system into E. coli in 2011. By spring 2012 his team had shown biochemically that Cas9 can be reprogrammed by changing its crRNA, and that two parts of the protein each cut one DNA strand. He submitted the paper before the Berkeley group, but Cell returned it without review, according to Eric Lander’s 2016 account. PNAS published it online on 4 September 2012, after the Berkeley paper.
On 12 December 2012 the Broad Institute of MIT and Harvard filed a provisional application covering CRISPR-Cas9 in eukaryotic cells, the cells of plants, animals and people. It named Feng Zhang as inventor. Broad later paid for accelerated examination of the resulting application. That decision put the Broad patent ahead of the earlier University of California filing in the patent office queue.
Science published online on 3 January 2013 two papers showing CRISPR-Cas9 editing inside human cells. Feng Zhang’s lab at the Broad Institute and MIT’s McGovern Institute engineered two Cas9 versions and edited human and mouse cells, including several genes at once. George Church’s lab at Harvard Medical School reported similar results in the same issue. Within weeks other labs edited zebrafish and yeast. Genome editing had moved from a test tube to living animal cells in about six months.
In November 2013 two companies formed to turn CRISPR into medicine. Editas Medicine started in Cambridge, Massachusetts, with Zhang, Church and, briefly, Doudna among its founders. CRISPR Therapeutics started in Switzerland with Charpentier as a co-founder. Doudna had co-founded Caribou Biosciences in Berkeley in 2011, and Intellia Therapeutics followed in 2014. The rival patent holders were now also rival shareholders.
On 15 April 2014 the US Patent and Trademark Office issued patent 8,697,359 to the Broad Institute for CRISPR-Cas9 editing in eukaryotic cells, with Feng Zhang as inventor. The University of California application, filed seven months earlier, was still under examination. The fast track had worked. Berkeley asked the patent office to declare an interference, a proceeding to decide who invented first.
After a meeting in Napa, California in January 2015, Doudna, David Baltimore, Paul Berg and other scientists published a call in Science on 19 March 2015. They asked researchers to refrain from heritable editing of the human genome while the risks and ethics were discussed. Doudna had organised the meeting. The statement echoed the 1975 Asilomar conference on recombinant DNA, which Baltimore and Berg had also attended.
Junjiu Huang’s team at Sun Yat-sen University in Guangzhou reported in the journal Protein & Cell in April 2015 that they had used CRISPR-Cas9 on human embryos that could not develop into babies. They tried to edit the gene for beta thalassemia, a blood disorder. Many edits missed or cut the wrong places. The same month the US National Institutes of Health said it would not fund gene editing of human embryos.
The US National Academies, the UK Royal Society and the Chinese Academy of Sciences held the International Summit on Human Gene Editing in Washington from 1 to 3 December 2015. The organising committee said it would be irresponsible to edit embryos for pregnancy until safety was resolved and society broadly agreed. That month Science named CRISPR its Breakthrough of the Year.
In January 2016 the Patent Trial and Appeal Board declared interference 106,048 between the University of California group and the Broad Institute. It asked whether the two sets of claims covered the same invention. On 15 February 2017 the board ruled that they did not. Broad’s claims to editing in eukaryotic cells were, in its view, not obvious from the Berkeley test tube work, so both sides kept their patents.
On 10 September 2018 the US Court of Appeals for the Federal Circuit affirmed the board’s 2017 decision that there was no interference. Broad kept its eukaryotic patents. The ruling did not decide who invented CRISPR editing first. It decided only that the two sets of claims could both stand, which left that question for a second proceeding.
On 25 and 26 November 2018 MIT Technology Review and the Associated Press reported that He Jiankui, a researcher at the Southern University of Science and Technology in Shenzhen, had edited human embryos to disable the CCR5 gene, which HIV uses to enter cells. He said twin girls had been born. Two days later he presented the work at the second human genome editing summit in Hong Kong. Scientists there condemned it. Chinese authorities opened an investigation that week.
On 24 June 2019 the Patent Trial and Appeal Board declared interference 106,115. This time the question was priority: who first invented CRISPR-Cas9 editing in eukaryotic cells. The University of California argued it had conceived the invention by 9 March 2012. Broad argued that conception required knowing the system would work in such cells. The proceeding ran for two years and eight months.
In July 2019 a woman from Mississippi with sickle cell disease became the first patient treated in a US trial of CTX001, a therapy from Vertex and CRISPR Therapeutics. Doctors collected her blood stem cells, edited them outside the body to switch fetal hemoglobin back on, and infused them after chemotherapy. The treatment was given at a hospital in Nashville, Tennessee. CTX001 was later named exa-cel and sold as Casgevy.
On 30 December 2019 a court in Shenzhen sentenced He Jiankui to three years in prison for illegal medical practice and fined him 3 million yuan, about $430,000. Two colleagues, Zhang Renli and Qin Jinzhou, received shorter sentences and fines. The court said three gene-edited babies had been born, the twins and a third child. The hearing was closed to the public.
On 7 October 2020 the Royal Swedish Academy of Sciences awarded the Nobel Prize in Chemistry to Emmanuelle Charpentier, then at the Max Planck Unit for the Science of Pathogens in Berlin, and Jennifer Doudna of UC Berkeley, for developing a method for genome editing. They were the first two women to share a science Nobel. The prize cited the 2012 paper. Siksnys, Zhang and Mojica were not included.
On 26 June 2021 Intellia and Regeneron reported in the New England Journal of Medicine that NTLA-2001, a single infusion of CRISPR-Cas9 carried to the liver, cut levels of the disease-causing protein TTR in six patients with hereditary transthyretin amyloidosis. The mean drop at day 28 was 52% at 0.1 mg/kg and 87% at 0.3 mg/kg, with one patient reaching 96%. The trial ran in the United Kingdom and New Zealand.
On 28 February 2022 the Patent Trial and Appeal Board ruled that the Broad Institute had priority for CRISPR-Cas9 editing in eukaryotic cells. It found Broad had reduced the invention to practice by 5 October 2012. The board did not accept that the Berkeley group’s March 2012 idea was a complete conception, pointing to their own doubts and failed experiments in the months after. The University of California appealed.
On 16 November 2023 the UK Medicines and Healthcare products Regulatory Agency authorised Casgevy for sickle cell disease and transfusion dependent beta thalassemia in patients aged 12 and over. It was the first approval of a CRISPR-based medicine anywhere. Casgevy edits the patient’s own blood stem cells to restart production of fetal hemoglobin. The European Medicines Agency granted a conditional authorisation in February 2024.
On 8 December 2023 the US Food and Drug Administration approved Casgevy, from Vertex and CRISPR Therapeutics, for sickle cell disease in patients 12 and older. The same day it approved Lyfgenia, a non-CRISPR gene therapy from bluebird bio, for the same disease. Vertex set a US list price of $2.2 million for the one-time treatment. Eleven years had passed since the 2012 Science paper.
Doctors at Children’s Hospital of Philadelphia gave a baby with severe CPS1 deficiency, a disorder that lets toxic ammonia build up in the blood, a base editing therapy designed for his single mutation. He received three infusions from February to April 2025, the first at seven months old. Rebecca Ahrens-Nicklas and Kiran Musunuru led the team. A year later CHOP reported he could tolerate more dietary protein and needed less medication.
On 12 May 2025 the Federal Circuit vacated the 2022 decision for Broad and remanded the case. The court held that the board had confused conception with reduction to practice by requiring the Berkeley inventors to know their system would work in eukaryotic cells. It upheld the finding that the first University of California provisional application lacked written description. Related claims by Sigma-Aldrich and ToolGen stayed on hold pending the remand.
On 23 February 2026 the FDA, joined by Ahrens-Nicklas and Musunuru, announced a plausible mechanism framework for individualized therapies. Variant-specific versions of the same gene editor would be treated as one drug. The agency had agreed to a trial protocol that could involve as few as five patients with at least three genetic variants. Musunuru put the cost of the first treatment near that of a liver transplant, about $1 million.