Drug Delivery
Paul J. Rice-Boucher (he/him/his)
Doctoral Candidate
Washington University in St. Louis
Saint Louis, Missouri, United States
Hangil Park, PhD
Executive Director
Walking Fish Therapeutics, Inc, United States
Rosa Romano, PhD
Associate Director, Cell Therapy
Walking Fish Therapeutics, Inc, United States
Mark Selby, PhD
Vice President, Immunology
Walking Fish Therapeutics, Inc, United States
Jacqueline M. Tremblay
Senior Research Assistant
Tufts University, United States
Charles Shoemaker, PhD
Professor
Tufts University, United States
David T. Curiel, MD, PhD
Professor
Washington University in St. Louis, United States
Zhi Hong Lu, PhD
Assistant Professor
Washington University in St. Louis, United States
Genetic medicine requires the safe and efficient delivery of nucleic acids to the target cell of interest. Despite decades of research in various viral and non-viral vectors, targeting to specific cells and specific receptors remains a significant challenge. To address this issue, our group and others have utilized Adenovirus vectors (Ads), one of the oldest and best studied viruses. A variety of techniques have been utilized to retarget the virus, including adaptor proteins, genetic engineering of the virus protein responsible for cell binding, and others. In this study we built upon these efforts by utilizing SpyCatcher-SpyTag family molecular glues. These peptide-protein partners spontaneously form covalent bonds under gentle conditions, allowing for new opportunities in protein engineering1. We developed an Adenovirus vector with the “DogTag” peptide genetically incorporated into the virus capsid, and developed fusion proteins consisting of the “DogCatcher” glue partner linked with a single-domain or single chain variable fragment antibody (sdAb and scFv)2. Tethering of these antibodies to the virus surface allowed retargeting of the vector towards the cell receptor bound by the antibody. We thus demonstrated this technology could be used to achieve cell specific targeting of primary murine lymphocytes, allowing a single pre-configured vector to be redirected using simple reagents produced separately. We anticipate these vectors will have utility in several fields, including gene delivery, vaccine development, and oncolytics.
We utilized a first-generation vector based on Human Adenovirus C5 (Ad5) expressing eGFP for these studies3. DogTag was inserted into the fiber protein (which the virus utilizes to bind to cells) at the HI loop, similarly to previous work4. The modified fiber was synthesized by IDT and assembled into the backbone using New England Biolabs HiFi DNA Assembly. The resulting vector, termed Ad5FDgT, was upscaled in HEK293 cells as previously described by our group and others.
To generate antibody-DogCatcher fusions, antibody DNA sequences were obtained from our own internal partners or public repositories. Antibodies were fused with DogCatcher at their C-terminus and produced in E. Coli strain BL21(DE3) (for sdAb constructs) or HEK293T cells (for scFv constructs) and purified using NiNTA columns. Antibodies were conjugated to the vector by co-incubation at room-temperature for 2-hours.
Murine splenocytes were isolated from C57BL/6J mice as described in recent studies5. Both B and T lymphocytes were isolated using commercial magnetic kits (B lymphocytes; Miltenyi, 130-090-862, T lymphocytes; Miltenyi 130-095-130). B lymphocytes were cultured as previously described using 50 µg/mL LPS as the activation agent for approximately 30 hours at 37C prior to infection5. T lymphocytes were cultured in RPMI 1640 with 250U/mL IL2 and anti-CD3/CD28 microbeads at a 2:1 bead:cell ratio as activating agents. Cultured cells were infected with 1000 MOI virus and scored for eGFP expression using flow cytometry approximately 40 hours later.
We successfully incorporated the DogTag peptide into the Ad fiber HI loop and upscaled the vector to similar titers to the control virus. We found that this vector remained highly infectious both in vitro and in vivo, indicating DogTag insertion did not affect the virus capsid stability or structure. For antibodies, we first generated a fusion based on an sdAb previously utilized in our lab, F8, which targets murine CD40. We incubated this fusion with our vector and found via SDS-PAGE that the fiber protein was successfully conjugated by the antibody, resulting in a nearly 100% complete reaction with the fiber. Infection of CD40+ primary B lymphocytes resulted in a strong enhancement of gene transfer compared to the non-conjugated control. To build on this concept, we next expressed the murine CD8 targeting YTS169 as an scFv-DogCatcher fusion. We found that the vector was also completely conjugated by the antibody, and that gene transfer to CD8+ T lymphocytes was strongly enhanced. Critically, infection of T lymphocytes with the F8 conjugated vector did not result in gene transfer enhancement, and neither did infection of B lymphocytes with the YTS169 vector (Figure 2).
Our study demonstrates that a modified Ad vector could be targeted towards cell type restricted receptors using previously characterized antibodies, potentially resulting in truly cell-specific gene delivery. Several features make this technology an advancement over previous Ad targeting systems. The upscale of our DogTag modified vector is simple and rapid, comparable to unmodified Ad5 vectors, and does not require a specialized cell line. Similarly, generation of the antibody-DogCatcher fusions is relatively simple, especially for sdAb fusions. In theory, this technology could allow targeting of the virus with any antibody which can be converted into an sdAb or scFv, potentially allowing vector targeting with antibodies that already have an established biological profile. This may yield opportunities in fields which require gene delivery that were not accessible in the past. In further work, we aim to validate these reagents in vivo and build upon our vector design by modifying other capsid proteins demonstrated to be involved in in vivo gene transfer.
1. 1. Keeble, A. H. et al. Approaching infinite affinity through engineering of peptide-protein interaction. Proc Natl Acad Sci U S A 116, 26523–26533 (2019).
2. 2. Keeble, A. H. et al. DogCatcher allows loop-friendly protein-protein ligation. Cell Chem Biol 29, 339-350.e10 (2022).
3. 3. Haldeman, J. M. et al. Creation of versatile cloning platforms for transgene expression and dCas9-based epigenome editing. Nucleic Acids Res 47, (2019).
4. 4. Krasnykh, V. et al. Characterization of an adenovirus vector containing a heterologous peptide epitope in the HI loop of the fiber knob. J Virol 72, 1844–1852 (1998).
5. 5. Huang, D. et al. Vaccine elicitation of HIV broadly neutralizing antibodies from engineered B cells. Nature Communications 2020 11:1 11, 1–10 (2020).