Microbial warfare brought us CRISPR. What big breakthroughs could be next?

All around the world — in the oceans, the soil, your body — an invisible battle is raging. Earth’s vast population of roughly 10³⁰ bacteria faces an unending onslaught from an even larger army of viruses, known as bacteriophages. The bacteria have a variety of defences at their disposal: they chop up viral components, deny the invaders key ingredients for replication and even shut down their own biological systems to halt infections, sacrificing themselves to protect nearby kin. The viruses, in turn, evolve counter-defence mechanisms, resulting in an ever-escalating arms race.

Although microbiologists are just beginning to understand the extent of this eternal contest, microbial immune mechanisms have already inspired technologies that have revolutionized biology. The discovery of restriction enzymes, bacterial proteins that slice DNA at specific sites, sparked the field of molecular biology in the 1970s, enabling everything from the development of genetically modified organisms to DNA forensics. CRISPR–Cas, a bacterial defence system that recognizes and slashes specific sequences in viral genomes, gave scientists the power to delete or edit genes with remarkable precision. Developed in the early 2010s, CRISPR-based gene editing has attracted billion-dollar investments and earned its key discoverers the 2020 Nobel Prize in Chemistry.

CRISPR–Cas and other breakthroughs have spurred an explosion of interest and discoveries in microbial defence systems. Thanks to advances in computational biology and genomic sequencing, scientists have identified a slew of immune mechanisms that bacteria and the other prokaryotic form of life on Earth, archaea, use in their ongoing existential struggle with viruses (see ‘The many sides of microbial immunity’).

“I would venture to say that bacteria and archaea use everything that you can imagine for the purpose of defence — and then, some that you cannot,” says Eugene Koonin, an evolutionary biologist at the National Library of Medicine in Bethesda, Maryland.

The discoveries have already led to improvements in gene editing and other laboratory processes. And these defences — some of which have relatives in the human immune system — are likely to inspire new treatments, particularly in the area of phage therapies, which kill pathogenic bacteria but are harmless to humans, and antibiotics. And although no single technology seems poised to dethrone CRISPR–Cas as one of the most transformative biological discoveries of the century, the excitement is palpable.

“To do better than CRISPR is a very tall order,” says Koonin, but “there are some of these systems that have very significant applications”.

A strong defence

Although bacteria and archaea and their phages have been locked in competition for billions of years, microbiologists were oblivious to the panoply of defence systems until the past decade or so. In part, that’s because there was no easy way to search for them, says Rotem Sorek, a microbiologist at the Weizmann Institute of Science in Rehovot, Israel.

That changed in 2011, when Koonin, evolutionary biologist Kira Makarova and their colleagues showed that immunity genes in the microbial genome tend to cluster together in ‘defence islands’¹. This discovery meant that scientists could use known defence genes to find likely suspects nearby.

Biggest-ever AI biology model writes DNA on demand

Candidates were soon pouring in. In the years that followed, Sorek’s group and others reported hundreds of potential defence genes^2,3. Most of the research so far has focused on bacteria, for which scientists have ample tools to tinker with genes.

To verify a defence system, scientists can select a candidate, insert the genes into a standard strain of laboratory bacteria and test whether they allow the microbes to resist phages that they couldn’t fend off before. The discoveries are coming fast, says Artem Isaev, a microbial immunologist at the Center for Molecular and Cellular Biology in Moscow. “Right now, we are discussing in our journal club five papers a week, and we don’t have time to discuss everything.”

Studies have yielded surprising immune mechanisms, such as the one biochemist Philip Kranzusch encountered while he was a postdoctoral researcher in Jennifer Doudna’s lab at the University of California, Berkeley. Kranzusch was not a microbiologist: he trained in human virology. In the early 2010s, he was interested in a human immune protein called cGAS. It detects foreign DNA and produces a signalling molecule that kick-starts part of the immune system known as the interferon response.

Bacteria contain enzymes that produce a similar signalling molecule, so Kranzusch wondered whether studying those enzymes would provide insights into cGAS function. Remarkably, the bacterial proteins had a very similar structure to cGAS. “I remember running into Jennifer’s office,” Kranzusch recalls. “The same machine, found in human cells, is found in bacteria as well.”

‘Dark proteins’ hiding in our cells could hold clues to cancer and other diseases

Kranzusch, at his own lab at Harvard Medical School in Boston, Massachusetts, continued to investigate the system. Sorek was also studying the bacterial cGAS counterpart, called CBASS. Their groups showed that CBASS is a bona fide microbial defence system that even uses the same signalling molecule receptor, called STING, as the human immune system^4,5.

Kranzusch and Sorek also discovered bacterial counterparts of eukaryotic immune proteins called gasdermins, which create pores in the cell membrane to kill infected cells and block viral replication⁶. And microbiologist Aude Bernheim, then a postdoc in Sorek’s lab, and her colleagues described prokaryotic versions of eukaryotic viperins: these proteins produce molecules that halt the transcription of viral genes into RNAs⁷. Sorek’s team also described a microbial immune mechanism called Thoeris that has close parallels with plant defences².

“From an evolutionary perspective, this is highly unexpected,” says Bernheim, who is now at the Pasteur Institute in Paris. Scientists had presumed that immune systems would evolve rapidly in the face of ongoing viral assaults, leading to vast divergence between prokaryotic and eukaryotic defences. Instead, at least in some cases, a system that evolved in the precursor of prokaryotes and eukaryotes has persisted over billions of years of evolution, creating parallel biology in microbes and humans or plants.

Building tools

Now, biotechnologists are adapting these ancient innovations into lab or clinical tools.

At a basic level, they can protect valuable microbial cultures, says Owen Tuck, a graduate student in Doudna’s lab who is working on a DNA-chomping defence system called Hachiman. For instance, if a bioreactor became infected with phages, scientists could activate defences in the bacteria to vanquish the invaders, Tuck says.

Researchers are also creating bespoke tools. Consider the Argonaute system, an immune response initially identified in eukaryotes — first in plants, then in animals. It uses small RNA guides to target other RNAs, such as viral RNAs, for destruction. In microbes, Argonaute systems work in more variable ways, says Daan Swarts, a biochemist at Wageningen University & Research in the Netherlands. Argonaute sensors might notice loose DNA ends, he speculates, or an excess of circular DNA molecules. In response to infection, Argonaute-associated mechanisms might then damage pathogen genes or deplete the cell of crucial metabolites such as NAD⁺ and NADP⁺, which are involved in energy generation and other processes.

Researchers have used microbial Argonautes to edit bacterial genomes and cut precise DNA sequences in test tubes. In experiments seeking specific, rare sequences, scientists have also used Argonautes to eliminate common but undesired sequences so that the rare ones are easier to detect. And Swarts and his team have designed diagnostics to identify sequences of interest. To do so, they adapted the NAD⁺/NADP⁺-depleting system, called SPARTA, to create a test that changes colour or fluoresces once the target sequence is captured⁸. “Basically, any type of sequence can be detected with these Argonautes,” says Swarts. “We’ve not run into any limitations yet.”

Many of these tasks can be performed by CRISPR–Cas, but Argonautes offer some different features. Both systems require a guide RNA to find target sequences, and the Cas enzyme also needs an extra bit of sequence, called a PAM, that helps Cas to bind to and unwind the DNA double helix. Argonaute works without a PAM and the guide RNA sequence can be shorter, making the system simpler to manufacture. Furthermore, Swarts notes, it is easier to submit patent applications for Argonautes than to obtain patents for ideas in the saturated CRISPR–Cas field.

Feng Zhang, a molecular biologist at the Broad Institute of MIT and Harvard in Cambridge, Massachusetts, has discovered⁹ another efficient CRISPR-like system. He and his colleagues sought molecules with genetic sequences and protein structures similar to CRISPR–Cas and found a DNA-binding system they called TIGR–Tas (for tandem interspaced guide RNA–TIGR-associated protein). Like CRISPR, it’s a two-part system that detects specific DNA sequences and can be programmed to cut at desired locations. But TIGR–Tas requires no PAM, and the components are physically smaller than those for CRISPR, which could be useful in applications such as gene therapy, in which only a limited amount of molecular cargo can be delivered to tissues.

Another bacterial defence system primed for innovation involves genetic elements called retrons. These were discovered in the 1980s, when scientists noticed hundreds of copies of short, single-stranded DNAs in samples of the soil bacterium Myxococcus xanthus. They determined that these single-stranded DNAs were produced by a reverse transcriptase enzyme — which creates DNA from an RNA template — and this type of system became known as a retron. Seth Shipman, a bioengineer at the Gladstone Institutes and the University of California, San Francisco, came across retrons in the scientific literature in the early 2010s as he sought a way to manufacture specific DNA sequences in cells. “They were just the perfect tool,” he says.

It wasn’t until 2020 that Sorek and others reported that retrons were actually involved in microbial immunity^3,10. They fall into a common category of defensive methods known as toxin–antitoxin, in which an antiviral toxin — which can vary from system to system — remains caged by a corresponding antitoxin until an invader is sensed.