A macrophage reporter cell line for high-throughput screening of gene delivery vehicles.

Introduction: Macrophages are phenotypically diverse orchestraters of inflammation and resolution of tissue regeneration. Yet macrophage phenotype is dysregulated in cancer and delayed tissue regeneration. Therefore, macrophages are a prime candidate for cell therapies (e.g. chimeric antigen receptor macrophages, CAR-M) and targeted gene delivery. However, gene delivery to macrophages is a long-standing challenge and attempts to transduce with viral gene carriers or ionizable lipids and polymers can lead to unintended polarization of targeted macrophages. To address this challenge, we developed a high-throughput screening technique for screening gene delivery vehicles. In parallel, we adapted an RNA-guided gene activation approach for activating endogenous IL-10 in macrophages. Together, this effort aims to establish macrophages as a cell therapy platform with stimuli-sensitive expression.

Materials and

Methods: We prepared a lentiviral construct containing an NFkB responsive promoter containing 14 NFkB binding motifs and a minimal CMV promoter driving an mCherry fluorophore (red). The mouse macrophage cell line (RAW 264.7) was transduced and selected with puromycin. The responsiveness of the promoter was characterized by a dose response with lipopolysaccharide (LPS) and benchmarked with a TNFa ELISA. A range of viral and non-viral vehicles carrying green fluorescent protein (GFP) were tested to allow simultaneous analysis of gene delivery efficacy (green) and macrophage polarization (red). The vehicles include AAV serotypes AAV1, AAV2, AAV5, AAV8, AAV9, Adenovirus Ad5/35, and commercial non-viral carriers. Finally, dCas9-VP64 and gRNAs targeting IL-10 were transduced to determine upregulation of endogenous IL-10.

Results, Conclusions, and Discussions: Fig. A shows mCherry fluorescence after stimulation with LPS. A dose response of LPS showed a highly predictable profile with TNFa as determined by ELISA (Fig. B). We noted a baseline expression of mCherry and TNFa, even in unstimulated RAW cells. We documented an optimized protocol for using these cells for high-throughput screening including kinetics and cell confluency. To demonstrate proof-of-principle, we screened 8 delivery vehicles spanning viral and non-viral vectors Fig C quantifies the four quadrants indicating stimulated macrophages (mCherry+) and successful gene delivery (GFP+). We found that Ad5/35 achieved the highest GFP+/mCherry- cell population. Ad5/35 is currently used in the production of CAR-M cells owing to the known affinity for CD46. AAV1 was the best performing AAV and could be used as a scaffold for future high-throughput screening. A commercial liposome and polymeric particle showed high delivery efficiency but were highly stimulatory. Fig D shows representative plots. In addition, when stimulated with LPS with simultaneous AAV1 delivery, we found a significant increase in gene delivery efficiency. This indicates that the inflammatory state of the macrophage can change viral tropism. We also tested a newly described myotropic AAV variant (MyoAAV). This AAV has a modified surface loop containing an RGD motif which increases muscle tropism ~10 fold over AAV9. However, MyoAAV demonstrated no delivery to the reporter line. We also tested combinations of AAV to see if synergistic delivery would increase delivery. Finally, we tested dCas9-VP64 with gRNAs targeting IL10 (Fig. E). We found that all tested formulations increased IL-10 expression, including the scrambled control owing to the stimulation caused by transduction. However, gRNA2 showed a significant increase in IL10 relative to controls. With the cell line established, libraries of viral and non-viral vehicles can be screened in mouse cell lines and primary macrophages for simultaneous gene delivery and macrophage programming to improve cancer therapy and tissue regeneration.

Acknowledgements (Optional): This work was supported by an NIH/NIBIB R00 #EB023979, ASGCT Career Development Award, University of Arkansas Chancellor’s Innovation Grant, and The Arkansas Bioscience Institute. CEN was supported by the 21st Century Chair in Biomedical Engineering. AI was supported by a distinguished doctoral fellowship. VK was supported by an NSF-sponsored summer REU.