Scientists from MIT and the Broad Institute have leveraged an existing bacterial system to create a novel strategy for protein delivery that is effective in both human and animal cells. This technique may be used to target certain cell types for the delivery of a wide range of proteins, including those employed in gene editing. This approach shows promise as a reliable vehicle for the transport of therapeutic agents in the fields of gene therapy and cancer treatment. The findings were published online on March 29, 2023, in Nature with the headline "Programmable protein delivery with a bacterial contractile injection system".

Led by Feng Zhang, a member of the Broad Institute Core Institute and a researcher at MIT's McGovern Institute for Brain Research, the team used a tiny syringe-like injection structure produced by bacteria that naturally binds to insect cells and injects a protein payload into them. They used the artificial intelligence tool AlphaFold to design these syringe structures to deliver a range of useful proteins to human cells and cells in living mice.

Joseph Kreitz, first author of the paper, said, "This is a very beautiful example of how protein engineering can alter the biological activity of a natural system. I think it confirms that protein engineering is a useful tool for bioengineering and developing new therapeutic systems."

Feng Zhang added, "Delivery of therapeutic molecules is a major bottleneck, and we will need a deep selection platform to get these powerful new therapies into the right cells in the body. By learning how nature transports proteins, we are able to develop a new platform that will help address this gap."

Symbiotic bacteria use approximately 100 nm long syringe-like molecular machines to inject proteins into host cells to help adjust the biological environment around them and enhance their survival. These molecular machines, called the extracellular contractile injection system (eCIS), consist of a rigid tube inside a sheath that contracts, driving a tip at the end of the rigid tube across the cell membrane. This forces the protein cargo inside the rigid tube into the host cell.

On the outside of one end of the eCIS are tail fibers that recognize and bind to specific receptors on the cell surface. Previous studies have shown that eCIS can naturally target insect cells and mouse cells, but Kreitz believes that by redesigning these tail fibers to bind to different receptors, it may be possible to deliver the protein to human cells.

Using AlphaFold, which predicts protein structures based on amino acid sequences, these authors redesigned the tail fibers of eCIS produced by Photorhabdus to bind to human cells. By redesigning the other part of this complex, they induced eCIS to deliver the protein of their choice, in some cases with very high delivery efficiency.

The authors produced eCIS that targeted cancer cells expressing the EGF receptor and showed that it killed almost 100 percent of these cells, but did not affect cells without this receptor. Although the efficiency depends in part on the receptors targeted by such systems, Kreitz said the findings show the promise of the system when thoughtfully engineered.

The authors also used eCIS to deliver the protein to the brains of living mice, where it elicited no detectable immune response, suggesting that eCIS might someday be used to safely deliver gene therapy to humans.

According to Kreitz, the eCIS system is flexible enough to be used to transport a wide variety of proteins, including base editor proteins (which make single-base modifications to DNA), proteins that are harmful to cancer cells, and Cas9, a major DNA cutting enzyme utilized in many gene editing systems.