Scientists led by a team at Stanford University have developed a compact, efficient CRISPR-Cas system, called CasMINI, which is about half the size of existing CRISPR-Cas systems, and which could have broad utility for gene therapy applications as well as cell engineering. The researchers confirmed in experiments that CasMINI could, just like its larger counterparts, delete, activate, and edit target gene sequences. But the smaller relative size of CasMINI means that it should be easier to deliver into human cells and the human body, making it a potential tool for treating a wide range of disorders, including eye disease, organ degeneration, and genetic diseases.
Whereas the commonly used CRISPR systems, such as those based on the Cas9 and Cas12a CRISPR-associated (Cas) proteins are made of about 1,000 to 1,500 amino acids, the new CasMINI system has just 529.
“This is a critical step forward for CRISPR genome-engineering applications,” suggested senior study author Stanley Qi, PhD, an assistant professor of chemical and systems biology in the Stanford School of Medicine and a Stanford ChEM-H institute scholar. “The work presents the smallest CRISPR to date, according to our knowledge, as a genome-editing technology. If people sometimes think of Cas9 as molecular scissors, here we created a Swiss knife containing multiple functions. It is not a big one, but a miniature one that is highly portable for easy use.”
Qi and colleagues reported on their development in Molecular Cell, in a paper titled, “Engineered miniature CRISPR-Cas system for mammalian genome regulation and editing.”
The development of CRISPR-Cas systems for human cells has “revolutionized genome engineering,” the authors wrote. But while these systems offer opportunities for the development of gene therapies for a variety of genetic diseases, their large sizes often restrict delivery into cells, and which can limit their potential clinical applications. For example, adeno-associated virus (AAV) vectors that are widely applied for in vivo delivery have a payload packaging capacity limit of less than 4.7 kb, and many Cas fusion proteins are beyond this. “The large size of CRISPR-Cas effectors and their fusion proteins has posed a challenge for efficient cell engineering and in vivo delivery,” the investigators stated. “ … there is a great need to engineer highly efficient, compact Cas systems to facilitate the next generation of genome engineering applications.”
One potential solution is Cas12f, also known as Cas14, which is less than half the size of currently used CRISPR systems such as Cas9 or Cas12a. But until now, it wasn’t clear whether this compact protein could be used in mammalian cells. “Recent years have identified thousands of CRISPRs, which are known as bacteria’s immunity defense system,” explained Qi. “More than 99.9% of discovered CRISPRs, however, cannot work in human cells, limiting their use as genome-editing technologies.”
For their newly reported work, Qi and colleagues applied RNA and protein engineering to the Cas12f system to generate an efficient miniature Cas system for mammalian genome engineering. By optimizing the single-guide RNA design and performing multiple rounds of iterative protein engineering and screening, the researchers generated a class of Cas12f variants named CasMINI.
The researchers decided to start with the CRISPR protein Cas12f because it contains only about 400 to 700 amino acids. “The size of the engineered CasMINI molecule is 529 amino acids, which is 62% and 57% smaller than the commonly used SpCas9 (1,368 amino acids) and LbCas12a (1,228amino acids), respectively,” the scientists noted. However, like other CRISPR proteins, Cas12f naturally originates from Archaea—single-celled organisms—which means it is not well-suited to mammalian cells, let alone human cells or bodies. Only a few CRISPR proteins are known to work in mammalian cells without modification. Unfortunately, CAS12f is not one of them. This makes it an enticing challenge for bioengineers like Qi. “We thought, ‘Okay, millions of years of evolution have not been able to turn this CRISPR system into something that functions in the human body. Can we change that in just one or two years?’” said Qi. “To my knowledge, we have, for the first time, turned a nonworking CRISPR into a working one.”
Xiaoshu Xu, a postdoctoral scholar in the Qi lab and lead author of the newly reported research, saw no activity of the natural Cas12f in human cells. Xu and Qi hypothesized that the issue was that human genome DNA is more complicated and less accessible than microbial DNA, making it hard for Cas12f to find its target in cells. By looking at the computationally predicted structure of the Cas12f system, she carefully chose about 40 mutations in the protein that could potentially bypass this limitation and established a pipeline for testing many protein variants at a time. A working variant would, in theory, turn a human cell green by activating green fluorescent protein (GFP) in its genome.
“At first, this system did not work at all for a year,” Xu said. “But after iterations of bioengineering, we saw some engineered proteins start to turn on, like magic. It made us really appreciate the power of synthetic biology and bioengineering.”
While the first successful results were only modest the team was encouraged because even modest results meant the system worked. Over many additional iterations, Xu was able to further improve the protein’s performance. “We started with seeing only two cells showing a green signal, and now after engineering, almost every cell is green under the microscope,” Xu said. But by optimizing the single-guide RNA design and performing multiple rounds of iterative protein engineering and screening, the researchers generated a class of Cas12f variants named CasMINI.
In addition to their protein engineering efforts, the researchers also engineered the RNA that guides the Cas protein to its target DNA. Modifications to both components were crucial to making the CasMINI system work in human cells. Encouragingly, the engineered Cas12f protein variants, combined with engineered single-guide RNAs, exhibited efficient gene-regulation and gene-editing activity. “By optimizing the single guide RNA (sgRNA) design and performing multiple rounds of iterative protein engineering and screening, we generate a class of Cas12f variants (i.e., CasMINI) which, when fused to a transcriptional activator, can efficiently activate reporter and endogenous gene expression,” the authors reported. They tested CasMINI’s ability to delete and edit genes in lab-based human cells, including genes related to HIV infection, anti-tumor immune response, and anemia. It worked on almost every gene they tested, with robust responses in several.
The tests demonstrated that CasMINI can drive high levels of gene activation, comparable to those associated with Cas12a, and allows for robust base editing and gene editing. Moreover, the system was found to be highly specific, and to produce no detectable off-target effects. “This dCasMINI-mediated gene activation has significant improvement over the wild-type dCas12f system, has comparable activation ability with the dCas12a system, and is specific in mammalian cells without detectable off targets,” the authors noted. “… CasMINI provides a useful tool for broad genome engineering applications that require compact Cas fusion proteins for delivery and cellular function.”
“Here we turn a nonworking CRISPR in mammalian cells, via rational RNA engineering and protein engineering, into a highly efficient working one,” Qi said. “There were previous efforts from others to improve the performance of working CRISPRs. But our work is the first to make a nonworking one working. This highlights the power of bioengineering to achieve something evolution has not yet done.”
The size of the engineered CasMINI molecule is only 529 amino acids. This small size makes it suitable for a wide range of therapeutic applications. The CasMINI fusion proteins are also well suited for AAV packaging. “We analyzed fusions of CasMINI to widely used repressors, activators, and gene editing domains and observed that all of them were below the AAV packaging limit (<4.7 kb),” the investigators stated. In addition, CasMINI mRNA can be easily packaged into lipid nanoparticles or other RNA-delivery modalities, potentially enhancing its entry into cells. “We also hypothesize that its small size and non-human pathogen source make it likely less immunogenic compared with large protein payloads,” the team added. “ … we envision that these synthetic, compact Cas effectors developed in this study will be broadly useful for gene therapy and cell engineering applications.”
More work is needed to further optimize the efficiency of CasMINI for base editing and gene editing and to test the performance of the system in vivo with different delivery modalities. The researchers plan to test the system for in vivo gene-therapy applications. “The availability of a miniature CasMINI enables new applications, ranging from in vitro applications such as engineering better tumor-killing lymphocytes or reprogramming stem cells to in vivo gene therapy to treat genetic diseases in the eye, muscle, or liver,” Qi pointed out. “It is on our wish list that it will become a therapy to treat genetic diseases, to cure cancer, and to reverse organ degeneration.”
The researchers are already setting up collaborations to pursue gene therapies. They are also interested in how they could contribute to advances in RNA technologies—such as those used to develop the mRNA COVID-19 vaccines—where size can also be a limiting factor.
“This ability to engineer these systems has been desired in the field since the early days of CRISPR, and I feel like we did our part to move toward that reality,” said Qi. “And this engineering approach can be so broadly helpful. That’s what excites me—opening the door on new possibilities.”
The authors pointed out that while previous studies have used protein engineering to generate enhanced Cas12a or Cas12b variants, the new work shows that it is possible to engineer efficient Cas12f effectors starting from an initial system that has no detectable activity in mammalian cells. “The RNA and protein engineering approach used in this work may be applicable to engineer more Cas12f/Cas14 effectors from other bacterial or archaeal species,” They stated, “These results likely suggest that many systems in the Cas12 family could be optimized for better efficiency via protein and guide RNA engineering.”