Programming membrane fusion and subsequent apoptosis into mammalian cells.

By the delivery of specific natural or engineered proteins, mammalian cells can be programmed to perform increasingly sophisticated and useful functions. Here, we introduce a set of proteins that has potential value in cell-based therapies by programming a cell to target tumor cells. First, the delivery of VSV-G (vesicular stomatitis virus glycoprotein) allowed the cell to undergo membrane fusion with adjacent cells to form syncytia (i.e., a multinucleated cell) in conditions of low pH typically occurring at a tumor site. The formation of syncytia caused the clustering of nuclei along with an integration of the microtubule network and ER. Interestingly, the formation of syncytia between cells that are dynamically blebbing, a mode of migration preferred during tumor metastasis, resulted in the loss of these morphology changes. Lastly, the codelivery of VSV-G with L57R (an engineered photoactivated caspase-7) allowed cells to undergo low pH-dependent membrane fusion followed by blue light-dependent apoptosis. In cell-based therapies, the clearance of syncytia between tumor cells might further trigger an immune response against the tumor.

Parts-based assembly of synthetic transmembrane proteins in mammalian cells.

Transmembrane proteins span cellular membranes such as the plasma membrane and endoplasmic reticulum (ER) membrane to mediate inter- and intracellular interactions. An N-terminal signal peptide and transmembrane helices facilitate recruitment to the ER and integration into the membrane, respectively. Using a parts-based assembly approach in this study, we confirm that the minimum requirement to create a transmembrane protein is indeed only a transmembrane helix (TM). When transfected in mammalian cells, our fusion proteins in the schematic form X-TM-Y were localized to vesicles, the golgi apparatus, the nuclear envelope, or the endoplasmic reticulum, consistent with ER targeting. Further studies to determine orientation showed that X was facing the cytoplasm, and Y the lumen. Lastly, in our fusion proteins with an N-terminal TM, the TM effectively reversed the orientation of X and Y. This knowledge can be applied to the parts-based engineering of synthetic transmembrane proteins with varied functions and biological applications.

Engineered networks of synthetic and natural proteins to control cell migration.

Mammalian cells reprogrammed with engineered transgenes have the potential to be useful therapeutic platforms because they can support large genetic networks, can be taken from a host or patient, and perform useful functions such as migration and secretion. Successful engineering of mammalian cells will require the development of modules that can perform well-defined, reliable functions, such as directed cell migration toward a chemical or physical signal. One inherently modular cellular pathway is the Ca(2+) signaling pathway: protein modules that mobilize and respond to Ca(2+) are combined across cell types to create complexity. We have designed a chimera of Rac1, a GTPase that controls cell morphology and migration, and calmodulin (CaM), a Ca(2+)-responsive protein, to control cell migration. The Rac1-CaM chimera (named RACer) controlled lamellipodia growth in response to Ca(2+). RACer was combined with LOVS1K (a previously engineered light-sensitive Ca(2+)-mobilizing module) and cytokine receptors to create protein networks where blue light and growth factors regulated cell morphology and, thereby, cell migration. To show the generalizability of our design, we created a Cdc42-CaM chimera that controls filopodia growth in response to Ca(2+). The insights that have been gained into Ca(2+) signaling and cell migration will allow future work to combine engineered protein systems to enable reprogrammed cell sensing of relevant therapeutic targets in vivo.

Effects of rapamycin-induced oligomerization of parvalbumin, Stim1 and Orai1 in puncta formation.

Elevations of cytosolic Ca2+ from the endoplasmic reticulum (ER) regulate a diverse range of cellular processes. When these luminal stores become depleted, the transmembrane ER protein Stim1 oligomerizes and translocates within the ER membrane to puncta junctions to couple with Orai1 channels, activating store-operated calcium entry (SOCE). Stim1 oligomerization and puncta formation have generally been associated with its luminal domains, however, studies have implicated that the cytoplasmic domains may contribute to this oligomerization. Studies have also suggested that intermediate or regulating elements may be required to fine-tune puncta formation and activation of SOCE. Here we made fusion proteins of Stim1 and Orai1 with FRB and FKBP12 domains that associate in the presence of rapamycin. Rapamycin-induced coupling of Stim1 to Stim1, Orai1 to Orai1 and Stim1 to Orai1 was found to be insufficient for puncta formation. Rapamycin was then used to recruit the cytosolic Ca2+ buffer protein parvalbumin (Pav) to Stim1 in order to buffer the local cytosolic Ca2+ near the ER membrane. Interestingly, Pav buffering near the ER caused puncta formation that was indistinguishable from those caused by thapsigargin. Our results suggest that Stim1 oligomerization and puncta formation may be additionally regulated either by local Ca2+ levels near the ER membrane or by as yet unidentified Ca2+-dependent proteins interacting with the cytoplasmic domains of Stim1.

Design of a genetic differential amplifier.

A genetic differential amplifier is made using the control elements of genes. The output mRNA level is proportional to the difference between the concentrations of two input proteins. The active element is engineered from the right operator of bacteriophage lambda. Mutations are introduced to yield the correct gain characteristic and to provide a bias level; the latter allows for the representation of negative differences. Simulation is used to aid the design process. A test circuit has been constructed. Preliminary experimental results indicate excellent results for the inverting input and lower gain for the non-inverting input.