To many working biologists, DNA-modifying enzymes are tools, not unlike a pair of molecule-sized scissors. To Dr. Barry Stoddard, a professor in the Basic Sciences division at Fred Hutch, these enzymes are fascinating molecular machines in their own right and a never-ending source of wonder. “For over fifteen years now, my group has been studying bacterial restriction-modification (RM) systems, in an effort to both understand the biochemical underpinnings of how these diverse enzymes work and evolve, and to harness their functions for more practical purposes.”
So, what are RM systems, exactly? If you’ve been around a molecular biology lab, chances are you’re familiar with restriction endonucleases—enzymes which recognize and cleave DNA sequences at specific positions determined by a short recognition sequence. “While many people are familiar with these enzymes as tools in the molecular biology lab, they actually evolved as a bacterial weapon to defend against bacteriophages (viruses which infect bacteria). Acting as a bacterial innate immune system, restriction endonucleases function by recognizing and cleaving phage DNA,” notes Stoddard.
To get even further into the weeds, the restriction endonucleases that biologists know and love are actually just one class of RM system—one constellation in a veritable universe of DNA-shredding anti-phage weapons deployed by bacteria. “Most familiar restriction enzymes are in a family called Type II RM systems—and as far as anti-phage weapons go, they’re quite rudimentary.” continues Stoddard. “Many of these systems work by requiring two separate enzymes (an endonuclease and a methyltransferase) to recognize and act at the same DNA sequence. As a result, such systems are 'locked into' action at a single unchanging target, making it easy for phage to avoid and overcome their action.”
To make it more difficult for phage to overcome their defenses, bacteria have evolved more sophisticated RM systems, which can be grouped into three other broad families known as Type I, III, and IV. Among the innovations found in these enzymes are the ability to bind more diverse target sequences and the combination of DNA binding, methylation, and cutting activities into single protein chains and assemblages that collectively interact to form the functional RM system. “Really, there is just a ton of functional diversity in bacterial RM systems, most of which we still don’t understand, that provides fertile ground for new biochemical discoveries,” says Stoddard.
In a recent study published in Nucleic Acids Research, a team led by Stoddard and former staff scientist Dr. Betty Shen reports the structural and functional characterization of a specific Type II enzyme called BsaXI—an odd child among RM systems. “BsaXI caught our attention because it has an odd mix of characteristics found in other families of RM systems,” notes Stoddard. “Similar to Type I and III enzymes, BsaXI has a DNA recognition (S’) domain separate from a fused restriction-methylation (‘RM’) domain, and it actually cuts DNA to either side of its recognition sequence (unlike the restriction enzymes that most people are familiar with). But similar to Type II systems, BsaXI doesn’t bind and move along DNA to find its target site—rather, it ‘hops around’ until it finds the recognition site and methylates or cuts depending on the context.”
So, how does BsaXI work? A structural biologist through-and-through, Stoddard firmly believes that seeing is believing: if you want to understand how a protein works, you need to understand its three-dimensional structure. This kicked off a six-year quest by Shen to solve the X-ray crystal structure of BsaXI in action—a quest which was repeatedly met with failure. “Betty was steadfast in trying to crystallize BsaXI, but no matter what she tried, the protein wouldn’t form crystals of sufficient quality,” Stoddard said.
Unable to solve a structure, the team put the project aside to focus on other things—until the late 2010s, when a glimmer of hope arrived in the form of a new technology called cryo-electron microscopy (cryo-EM). Unlike X-ray diffraction, which requires protein crystals to work, cryo-EM solves structures by rapidly freezing proteins in thin layers of ice and imaging them with high-powered electron microscopes. “In her previous position as faculty at UChicago, Betty had actually been involved with early efforts in developing cryo-EM, so when this technology became available, she was itching to use cryo-EM to try a solve a structure of BsaXI,” says Stoddard.
New technology in hand, the team was at long last able to solve the protein structure of BsaXI in both DNA-bound and unbound states—and what they found was striking. Individual subunits of DNA-free BsaXI adopted a conformation that Stoddard compares to a ‘door handle,’ with the S subunit in the center and two RM subunits at either end. Within each of these RM subunits, Shen and Stoddard noted an atypical structural feature consisting of two long alpha helices; they termed this structure a ‘paddle,’ and were surprised to find very few examples of this structural feature in the Protein Data Bank (PDB, a repository of all known protein structures).
Examining the BsaXI structure in DNA-bound form, the team noted a large-scale shift in the protein to accommodate a more ‘closed’ conformation. “If you can imagine the center S subunit binding the DNA recognition site, that leaves the two RM subunits on either side to ‘grab’ the DNA helix from either the top or bottom, with the paddle and endonuclease domains acting as sort of a gripper,” explains Stoddard. Subsequent mutagenesis experiments revealed several amino acids in both the paddle and endonuclease domains that are required for DNA cutting. Closer inspection of the DNA-bound structure revealed additional functional clues: methylation happens closer to the middle of the complex, one DNA strand at a time, and methylation on just a single DNA strand is sufficient to prevent DNA cutting on either side of the recognition site.
An uncommon ‘paddle’ fold, a distinct subunit arrangement, and an intriguing methylation mechanism are just a few of the insights gleaned from this study of BsaXI structure—overall, these results add to our growing functional understanding of the diverse world of bacterial RM systems. So, are BsaXI’s secrets finally revealed? “Not quite!” says Stoddard. “An astute observer will notice that—in order to cut the DNA on either side of the recognition site—BsaXI needs to make four cuts (both DNA strands on either side). However, each RM domain can only cut a single strand, and our structure only has two such RM domain—one on either side of the recognition site. So, the numbers don’t work out. This suggests that to effectively cut the DNA, BsaXI needs to form a higher order assemblage of two ‘door handles,’ which would also require two distinct recognition sites. An enticing hypothesis is that this arrangement might also contribute to how BsaXI ‘decides’ whether to methylate or cut the DNA—to distinguish host from non-host.”
Thus, the quest to understand bacterial RM systems continues, but what Stoddard and his lab find at each step keeps them coming back for more. Speaking about his lab’s work more broadly, Stoddard notes, “I’ve just been incredibly lucky to have had amazing lab members and collaborators, and to have had the funding and institutional support to pursue this work over the years.”
The spotlighted work was funded by the National Institutes of Health and New England Biolabs.
Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Member Dr. Barry Stoddard contributed to this research.
Shen, B. W., Heiter, D., Yang, W., Xu, S., & Stoddard, B. L. (2025). Cryo-EM structures of DNA-free and DNA-bound BsaXI: Architecture of a Type IIB restriction–modification enzyme. Nucleic Acids Research, 53(7), gkaf291.
Science Spotlight writer David Sokolov is a graduate student in the Sullivan Lab at the Fred Hutch. He studies how cancer cells modify their metabolism to facilitate rapid proliferation and accommodate tumor-driving mitochondrial defects. He's originally from the east coast and has bachelors' and masters' degrees from West Virginia University. Outside of the lab, you'll find him enjoying the outdoors, playing music, or raising composting worms in his front yard.
For the Media