(110e) Harnessing Protein-Protein Interactions for Programmable Control of Protein Secretion in Mammalian Cells

Motivation: One of the primary goals in the field of Synthetic Biology is to create programmable cellular therapeutics for improving various biomedical applications, such as cancer immunotherapy and cell transplantation. Creating programmable behaviours using protein-based circuits have advantages such as fast operation, compact delivery, and robust context-independent performance compared to traditional transcriptional circuits. Previously, we described a generalizable platform called RELEASE (Retained Endoplasmic Cleavable Secretion) which enabled intercellular signalling through the removal of ER-retention motifs compatible with pre-existing protein-based circuits (Fig. 1A). However, to program complex expression patterns such as logic operations or quantitative processing, multiple orthogonal proteases are required, which may preclude their use in viral vectors with limited packaging capacities. Some protein receptors control their surface expression through interactions with additional endogenous scaffolding proteins known as 14-3-3 proteins. The binding of 14-3-3 proteins sterically block interactions of the protein with the retrograde transport machinery, causing the protein to be expressed on the surface. Since RELEASE uses ER retention motifs that rely on the native retrograde transport machinery, we hypothesized that we could harness protein-protein interactions such as those with 14-3-3 proteins to control the secretion and surface expression of any tagged protein of interest.

Methods: To harness these protein-protein interactions with 14-3-3 proteins to create programmable outputs, we fused a small peptide binder of 14-3-3 proteins to the C-terminus of the RELEASE construct, effectively blocking its retention capabilities. By placing a protease cut site upstream of the 14-3-3 recruiting peptide, we could manipulate interactions with 14-3-3 proteins and implement NOT Gate logic control of protein secretion (Fig. 1B). This new design (Compact RELEASE) can function as both the processing and output module of protein circuits and achieves functional completeness of every possible logic operation.

Results: The Compact RELEASE suite overcomes the need to use additional proteases to confer logic iteration or quantitative processing through manipulating protein-protein interactions with native 14-3-3 scaffold proteins. For example, Compact RELEASE variants can be designed to have conditional activation of protein secretion in response to multiple inputs (i.e. NIMPLY gate – Fig. 1C). Through manipulating the cleavage efficiencies of different protease cut sites, Compact RELEASE also can be used for quantitative activation of protein secretion. One such example is bandpass activation, where protein secretion only occurs within a specific concentration range of input (Fig. 1D). Furthermore, by layering different RELEASE variants (i.e. standard, NOT and bandpass), we can enable concentration-dependent activation of multiple proteins for sequential activation of protein secretion, without requiring additional proteases to process the inputs. Pilot studies are underway to subcutaneously deliver engineered cells to control the local concentration of secreted proteins using small-molecule activators.

Conclusion: The Compact RELEASE suite enables functional complete programmable control of all Boolean logic gates and quantitative control of protein secretion. This platform expands the programmable capabilities of synthetic protein circuits in a compact manner, enabling their delivery using pre-existing viral vector methods such as adeno-associated viruses or adenoviruses.

Figure Caption: A) Schematic of RELEASE for controlling protein secretion in response to protease activity. B) Recruitment of 14-3-3 proteins via small peptides antagonizes the activity of the ER retention motif, resulting in constitutive protein secretion. Removal of the peptide via protease activation restores ER retention activity, implementing NOT gate logic. C) Conditional activation of protein secretion with multiple outputs. D) By creating a variant with two protease cut sites of different cutting efficiencies (K_cat), we can implement bandpass activation of protein secretion. Protein secretion (output) will only occur when the concentration of the input is within a specific range.