Nature's Needle: Harnessing Bacterial Molecular Syringes

Abstract

GEN BiotechnologyVol. 2, No. 3 Views & NewsFree AccessNature's Needle: Harnessing Bacterial Molecular SyringesChristopher M. Baehr and Ross C. WilsonChristopher M. Baehrhttps://orcid.org/0000-0003-2618-1332Innovative Genomics Institute, University of California, Berkeley, California, USA.California Institute for Quantitative Biosciences, University of California, Berkeley, California, USA.Molecular & Cell Biology Department, University of California, Berkeley, California, USA.Search for more papers by this author and Ross C. Wilson*Address correspondence to: Ross C. Wilson, Innovative Genomics Institute, University of California, 2151 Berkeley Way, Room 212E, Berkeley, CA 94704, USA, E-mail Address: rosswilson@berkeley.eduhttps://orcid.org/0000-0002-0644-5540Innovative Genomics Institute, University of California, Berkeley, California, USA.California Institute for Quantitative Biosciences, University of California, Berkeley, California, USA.Molecular & Cell Biology Department, University of California, Berkeley, California, USA.Search for more papers by this authorPublished Online:19 Jun 2023https://doi.org/10.1089/genbio.2023.29101.cmbAboutSectionsPDF/EPUB Permissions & CitationsPermissionsDownload CitationsTrack CitationsAdd to favorites Back To Publication ShareShare onFacebookTwitterLinked InRedditEmail A new report from Zhang and colleagues introduces an exciting new vehicle with potential for delivering molecular therapeutics into cells based on contractile injection systems.A vast array of protein-based therapeutics are currently under development for applications in cancer therapy and genetic interventions such as gene therapy or genome editing. However, these potential macromolecular therapeutics face the common challenge of intracellular delivery: the need for cargo to bypass the cell membrane and achieve cytosolic delivery. In contrast to small-molecule drugs, which can diffuse through or permeate cell membranes, this physical barrier is a critical hurdle that must be overcome for macromolecular cargo, such as CRISPR-based genome editing tools, to have effect. Although other groups have successfully harnessed viral vectors or synthetic lipid nanoparticles—both of which are showing considerable promise in the clinic—a new study from Zhang and colleagues (led by grad student Jacob Kreitz) at the Broad Institute reveals the tantalizing prospect of developing a distinct delivery approach derived from bacterial “molecular syringes.”1Extracellular contractile injection systems (eCISs) are sophisticated macromolecular complexes used by bacteria for delivery of protein factors across cell membranes. The structure of eCISs closely resembles that of a headless bacteriophage, and they share similar structural proteins, including a tail, sheath, and baseplate.2 When the tail domain meets the target cell, it triggers the contraction of the sheath. The inner tube and tail spike are then propelled through the target cell membrane, delivering the eCIS payload directly into the cytoplasm of the target cell. It is this striking and effective activity that earns eCISs the “molecular syringes” moniker.Several bacterial species utilize eCISs to interact with the membranes of other cells, frequently delivering protein cargo into eukaryotic cells. These eCIS payloads can have a variety of functions, including DNA fragmentation, modulation of the host cytoskeleton, and the induction of metamorphosis. The Photorhabdus virulence cassette (PVC) is an eCIS, which is able to deliver payload to the cytosol of insect cells (Fig. 1), triggering rapid actin condensation and cell death.3 Indeed, Kreitz et al. demonstrate that wild-type PVCs expressed in Escherichia coli confer a lethal effect to Sf9 insect cells with high efficiency.FIG. 1. Molecular syringe structure enables injection of molecular cargo into target cells.(A) Schematic representing the components of the PVC. (B) PVC mechanism of action: free particle diffusion to reach the cell surface, target cell recognition, sheath contraction, and payload release (left to right). PVC, Photorhabdus virulence cassette.It had been previously demonstrated that fusion of target proteins with an N-terminal signal peptide allows for eCIS packaging of a broad range of proteins, including mRFP, nanoluciferase, and TcsT, a ribosome-inactivating protein with proposed antitumor activity.4 Kreitz et al. adopted this strategy to package gene-editing proteins of interest into PVCs, including Cre recombinases, CRISPR-Cas9, and zinc finger nucleases. Cre-loaded PVCs were shown to deliver active Cre recombinase into Sf9 cells harboring a loxP-green fluorescent protein (GFP) cassette, resulting in robust GFP signal. Despite this promising activity, native PVCs are limited in their capacity for cell-specific targeting by their tail domain. For ideal therapeutic delivery, PVCs would efficiently target mammalian cells, with the ability to discriminate between cell types.To accomplish delivery to specific mammalian cell types, the team leveraged AlphaFold to develop a truncated PVC-13 tail protein fused to either an adenovirus binding domain (Ad5-knob) or an anti-epidermal growth factor receptor (EGFR) domain. These targeting domains enabled the targeting of mammalian cells and allowed for the delivery of Cre recombinase payload into EGFR-expressing A549 cells, harboring a similar GFP-on locus. Further modulation of the tail domain demonstrated that a range of recognition epitopes (e.g., FLAG, SunTag, and MoonTag) could be inserted, selectively facilitating delivery of cargo protein into human embryonic kidney 293FT cells bearing a range of complementary antibody “receptors.” This demonstrated the versatility of the PVC tail domain of PVC, suggesting its compatibility with diverse targeting domains.Special DeliveryAlthough many methods exist for intracellular delivery of therapeutic effectors in vitro, relatively few methods exist that can deliver candidate cargo effectively in vivo. The brain represents an opportune solid tissue for direct administration as it is surgically accessible and its immune privilege mitigates the risk of host responses.Kreitz et al. dosed several loxP-tdTomato mice (which bear a Cre-inducible red fluorescent protein reporter) stereotactically in the brain parenchyma with the Ad5-knob modified PVCs loaded with Cre recombinase. The engineered PVC eCISs demonstrated some ability to distribute in the brain and deliver the Cre payload, resulting in TdTomato signal. The material was shown to have a targeting preference for neuronal cells (as opposed to microglial cells). Neuronal tropism is desirable as many conventional gene targets in the brain (such as the mutant huntingtin, responsible for Huntington's disease) primarily impact neurons.5Moreover, T cell and granulocyte infiltration was similar to the sham control injection, suggesting PVCs are not markedly immunogenic when administered in the brain. The team was able to isolate and detect, through transmission electron microscopy, PVCs from the brain up to 1 day after administration, but not 7 days after administration, suggesting the delivery formulation is readily cleared from the brain parenchyma.These early results are promising, and compare well with other reports of in-brain editing using Cre recombinase.6 Of course, Cre is conventionally viewed as placeholder cargo, which can subsequently be exchanged for a CRISPR ribonucleoprotein (RNP) enzyme or other proteins of interest. There are several advantages of transient protein or RNP delivery systems when compared with DNA and RNA delivery vectors. For CRISPR effector delivery, the highly transient RNP format has been observed to generate fewer off-target edits as compared with mRNA7 or DNA8 cargo formats.Overall, this molecular syringe platform offers a promising new macromolecular delivery tool, with the potential to deliver a wide range of protein cargos to a variety of cells. Targeting moieties can be exchanged with ease, suggesting a capacity for “plug & play” retargeting for different cell types and indications. Indeed a variety of applications are already apparent: other groups have demonstrated the ability of eCISs—loaded with antitumor protein—to hamper tumor growth after intertumoral injection.4 However, the capacity for these systems to biodistribute broadly, or locate to a tissue after IV administration, has not yet been demonstrated. Future iterations of eCIS could play a critical role in the genetic therapy landscape, particularly if they can be designed to distribute extensively and produce robust cellular impacts in vivo.