“When you see something very peculiar, you have no alternative but to research it. I thought this was a nice thing to keep working on,” he says as we continue our stroll along the salterns. He had a hunch the mystery repeats might be tied to salt adaptation, perhaps by changing their conformation and thus gene activity by sensing changes in the cell’s osmotic pressure. “At the time, [DNA] supercoiling was the answer to everything in the regulation of gene expression!” It was a nice hypothesis, but wrong.II
Working after hours in the university library, Mojica eventually unearthed a Japanese report from 1987. Atsuo Nakata and Yoshizumi IshinoIII at Osaka University described a similar repeating motif in the genome of E. coli. While sequencing a gene of interest, the Japanese group had noticed an unusual nearby sequence with a distinctive repeat pattern, like crop circles carved into the DNA terrain. This region consisted of a series (the cluster) of short repeated stretches of palindromic sequence (reading the same forward and backward); each identical repeat, twenty-nine letters in length, was separated by a thirty-two-base stretch of unique sequence (the interspaced DNA). But because nothing similar had been seen before, and there were no clues as to their biological function, the researchers let it slide. The team dutifully wrote up and published their observations, which attracted scant attention at the time.6
Two years later,7 Mojica described a short sequence repeated hundreds of times in tandem, spanning more than 1,000 bases. Between each pair of repeats was a unique DNA sequence of unknown function. Mojica’s boss suggested calling these repeats—which were also observed in another extremophile, a volcano-loving Archaea—TREPs, for tandem repeats. Surely there was a reason why prokaryotes devote up to 2 percent of their precious compact DNA to these strange repeats? Microorganisms “cannot allow themselves luxuries,” Mojica thought, “they must have an important function.”8
Other scientists also stumbled upon these repeats. German microbiologist Bernd Masepohl puzzled over a stretch of thirteen DNA repeats, found in a cyanobacterium, which he called LTRR, for “long tandemly repeated repetitive.” But in focusing on the repeated DNA elements, Masepohl paid little attention to the unique sequences in between.9 Another team also came close to solving the mystery of the DNA repeats. In 2002, Eugene Koonin, a Russian expat computational biologist at the National Center for Biotechnology Information at the NIH, and his colleague Kira Makarova, described a series of bacterial genes they suspected to be part of a DNA repair system.10 What they didn’t realize was that these genes were sitting adjacent to the CRISPR array and—as we shall soon see—play an essential role in the function of CRISPR and gene editing.
After a few years working in Oxford, Mojica returned to Alicante in 1997 to set up his own group. With little funding, Mojica tried to do some very cheap experiments, “even though I had no idea about bioinformatics.” The nagging question was the origin of the spacer DNA, the sequencers interspersed between the repeats. “The easiest thing is to look at the databases and expect that something comes out, but we didn’t get anything—until 2003.” By now, the DNA databases were bursting with bacterial and archaea genomes, many of which carried versions of these repeats.
In 2000, Mojica renamed his obsession SRSRs (short regularly spaced repeats). That didn’t last long. Later he exchanged emails with Ruud Jansen in the Netherlands, who was studying a family of genes adjacent to the mystery repeats. Jansen felt a new name was needed, so Mojica suggested “CRISPR.” On November 21, 2001, Jansen emailed his enthusiastic approvaclass="underline"
Dear Francis,
What a great acronym is CRISPR. I feel that every letter that was removed in the alternatives made it less crispy so I prefer the snappy CRISPR over SRSR and SPIDR.11
CRISPR finally had a name, as did a group of unusual genes that seemed to piggyback with the CRISPR elements. Jansen reasonably, if unimaginatively, dubbed these “CRISPR-associated” genes, or Cas. For now, Mojica was still focused on the curious CRISPR spacers.
The breakthrough finally came one picture-postcard afternoon in August 2003. Mojica was vacationing with his wife close to home, near the salterns. Feeling the heat, Mojica made an excuse to pop back to his air-conditioned lab, where he could run a few more computer searches. This was routine, almost like playing a video game: Mojica would copy one of the mystery spacer sequences and paste it into a computer program called BLAST that would search for matches in the massive DNA database, GenBank. Mojica had run this program hundreds of times to no avail. His colleagues called his quest a waste of time. But on this day, to his astonishment, the computer flagged a match. This particular spacer from E. coli matched a stretch of viral DNA called P1 that crucially infects the same bacterium. Over the next few weeks, Mojica catalogued dozens of other examples of matches between CRISPR spacers and various viruses.
In October, Mojica submitted the biggest research paper of his life to the top journal—Nature. “I remember the title of the submission was: ‘Prokaryotic repeats are involved in an immunity system.’ To convince the editor and referees, we wrote that the existence of an acquired immune system in prokaryotes will have tremendous repercussions in biology and clinical sciences.” And the result? “It wasn’t even reviewed!”12
Perhaps something was lost in translation, but Nature’s editors didn’t consider this to be a conceptual advance “of sufficient general interest” that merited publication in its prestigious pages. Mojica appealed, arguing this was the first description of a bacterial immune system with a memory function. Nature indicated it was willing to reconsider if he could describe the mechanism underlying this immunity. But Mojica’s group couldn’t generate any experimental proof for the hypothesis, just the smoking gun of the sequences. One reason, it turned out, is that CRISPR is repressed in the most popular lab workhorse, E. coli.13 It was as if Mojica had incriminating physical evidence but no security camera footage.
Mojica licked his wounds and resubmitted to another journal… and another. Three more journals, including the Proceedings of the National Academy of Sciences (PNAS), all passed on CRISPR. Each delay increased the chances he might get scooped. Finally, in October 2004, Mojica submitted his manuscript to a lesser known journal specializing in evolution. It took a full six months before he finally heard some encouragement from the editor. Three months later, the paper was accepted. “I remember those two years like a nightmare,” he told me. “When you have something so big in your hands and you send it to very good journals—and all of them agreed it was not interesting enough to be published—you think, is it me who is crazy or something else?”14
Most scientific papers are team efforts, the fruit of months if not years of planning, reviewing, and repeating experiments, a continual exchange of ideas between student and mentor. If one member receives the spotlight, other members, rightly or wrongly, can feel aggrieved.