Activity-based nanosensors that implement AND-gate logic for programmable detection of antitumor immunity

Introduction: Advances in synthetic biology have historically focused on the genetic circuit paradigm to assemble sense-and-respond biocircuits that operate under transcriptional regulation, with far ranging applications in biomedicine to enhance cell therapies, drug delivery, molecular imaging, and biosensors. While gene circuits will remain at the forefront of development, cell- and gene-free biocircuits are gaining increased attention based on their potential to lower barriers for clinical translation while retaining the ability to perform logical functions. For example, RNA-based switches can regulate protein translation in response to cell-type-specific biomarkers while protein circuits can be engineered as bandpass filters for protein secretion. Moreover, recent work has described nanomaterials comprised entirely of synthetic components that process information from cell-secreted proteases for functions like tumor imaging, analog-to-digital conversion, and digital drug delivery. Here we describe a cell- and gene-free design of AND-gated nanosensors that require a specific pair of proteases before activation and the release of a reporter that can be detected locally in tissue or distally in urine. In preclinical mouse models, AND-gated nanosensors activated by the concomitant activity of granzyme B (GzmB) secreted by CD8 T cells and matrix metalloproteinases (MMPs) overexpressed by cancer cells detect tumors that are responsive to immune checkpoint blockade (ICB) therapy while minimizing signals from lungs during acute influenza infection, thereby leading to improved specificity for antitumor immunity.

Materials and

Methods: AND-gated nanosensors implement AND-gate logic via asymmetric bi-labile cyclic peptides (i.e., cyclic peptides bearing two different protease substrates), which actuate by releasing a reporter if and only if cleaved by a specific pair of proteases. To design AND-gated nanosensors, we synthesized bi-labile cyclic peptides with substrates selectively cleaved by GzmB and MMPs, conjugated a fluorophore and quencher pair flanking the substrates, and multivalently crosslinked these peptides to iron oxide nanoparticles to improve the catalytic efficiency of proteolysis and tissue retention in vivo. We assessed in vitro activation kinetics by incubating nanosensors with recombinant proteases and detecting cleaved peptides using fluorimetry. To assess signal during antiviral response, B6 mice were intranasally infected with 30 p.f.u. of PR8 influenza A virus (IAV). Eight days after infection, near-infrared fluorescent (NIRF) linear GzmB nanosensors or AND-gated nanosensors were intravenously injected, and mice were sacrificed three hours post-injection. Lungs were isolated and imaged by NIRF imaging. To assess signal during antitumor response, B6 mice were subcutaneously inoculated with 106 MC38 cancer cells. After tumors reached ~400 mm3, mice were treated with a combination ICB regimen of αPD1, αCTLA4, and IL-2, or control antibodies, administered every three days for three doses. After the final dose, NIRF AND-gated nanosensors were administered either intravenously or intratumorally, and mice were euthanized after three hours for analysis of sensor fluorescence in whole organs.

Results, Conclusions, and Discussions: We postulated that bi-labile cyclic peptides implement AND-gate logic by releasing a reporter if and only if both substrates are cleaved (Fig. 1a). To test this design concept, we incubated AND-gated nanosensors bearing substrates for GzmB and MMP9 with recombinant proteases and found that both proteases were required for production of a fluorescent reporter fragment, confirming AND-gate logic (Fig. 1b). Next, we tested whether AND-gated nanosensors implement AND-gate logic in vivo to improve detection specificity. GzmB activity is a key driver of antitumor T cell responses but is not specific to antitumor responses as T cells secrete GzmB during pathologies like viral infection and autoimmunity. However, we expected that AND-gated sensing increases detection specificity by minimizing signal from T cell activity without concomitant MMP activity (Fig. 1c). To test this, we first used a mouse model of acute influenza viral infection and intravenously administered linear GzmB nanosensors or AND-gated nanosensors. While linear GzmB nanosensors produced 3.0-fold higher fluorescence in the lungs of infected mice compared to naïve controls, AND-gated nanosensors did not significantly activate in infected lungs (Fig. 1d). We next assessed whether AND-gated nanosensors detect antitumor responses during ICB therapy. AND-gated nanosensors intravenously administered to mice bearing syngeneic MC38 tumors produced elevated fluorescence in ICB-responding tumors, without activation in non-tumor organs (Fig. 1e). Finally, given that in vivo assays based on fluorescence are not routinely used in the clinic, we evaluated an approach based on intratumoral injection of nanosensors via quantification of reporters in urine. Urinary biomarkers accurately differentiated mice with ICB-responsive tumors from both untreated and treatment-resistant B2m-/- tumors (Fig. 1f). Collectively, our results demonstrate AND-gated nanosensors implement synthetic logic without genetic circuitry to increase on-tumor precision. While genetic circuits continue to drive major advances in cell therapies and diagnostics, their translation often involves the need to reduce immunogenicity and requires approaches like “off-the-shelf” therapies to expand accessibility to larger patient populations. Our cell- and gene-free nanomaterial design could represent a lower barrier for translation.

Acknowledgements (Optional):