Understanding and engineering of complex biological systems are key challenges to the scientific community for addressing health problems, especially diseases. Studies of biological systems can vary in scale from molecules, cells, to entire organisms. Cells, as the basic building blocks of life, compose various biological entities of higher order and precisely control both structures and functions. Therefore, it is of vital importance to understand how cell functions and behaviors are regulated in order to engineer and manipulate cell behaviors. Recently, advances in molecular imaging technologies and synthetic biology approaches have enabled us to better detect and manipulate cell signals and behaviors, and provide a broad set of tools to influence and revolutionize basic research, medicine and therapy. In my dissertation, a general high-throughput platform has been established to systematically optimize biosensors and a new Src biosensor with high sensitivity has been developed accordingly. Combining fluorescence/Förster resonance energy transfer microscopy and laser scissors technology, the signaling transmission between neighboring cells and the underlying mechanism could thus be revealed, which could provide a cell model to understand intercellular communications and wound healing process. Furthermore, I have successfully developed, characterized and demonstrated specific molecular machineries to program cell behaviors by rewiring molecular signaling pathways, specifically immune responses in immune cells against cancer cells. These new tools and understanding may open a new avenue towards cancer therapy and lead to potential therapeutic strategies.

Protein engineering through directed evolution has been extensively used to improve and modify the structure and function of proteins for wide range of applications. In this dissertation, I utilized the conventional directed evolution through yeast surface display library to engineer a single molecular module, i.e., monobody to have high binding affinity to R-Phycoerythrin. This engineered monobody was then applied to 1) a FRET biosensor with improved spatial resolution to monitor biological process of cancer cell invasion, and 2) a universal chimeric antigen receptor (CAR) for cancer immunotherapy. To further extend the power of directed evolution, I combined it with other technologies, including mammalian cell library, functional screening by FACS-based FRET, high-throughput DNA sequencing, and sequence-function analysis, to systematically optimize the sensitivity, specificity, and dynamic range of the FRET biosensors for monitoring kinase activities with high spatiotemporal resolution in living cells. These optimized biosensors can be used to study the cell signaling of CAR T cells upon engagement with cancer cells to improve the efficiency of CAR T cell immunotherapy. Overall, this dissertation illustrates the potential of directed evolution as a tool to engineer the molecular modules/sensors for biomedical diagnosis and immunotherapy.

In recent years, CAR-T cell therapy has revolutionized cancer immunotherapy, but there are still major obstacles preventing its broader applications in treating solid tumors. Challenges including on-target off-tumor toxicities increased the risks associated with CAR-T cell therapy and currently proposed strategies still lacked precise spatiotemporal control for CAR-T cell activation. To address this issue, we identified a calcium ion channel TRPV1 that is activated by heat and integrated it into an AND-gate inducible genetic circuits in HEK293T cells, to convert the TRPV1 thermal-activation into transcriptional activities. We first demonstrated heat-inducible luciferase reporter gene expression in engineered HEK293T cell line. We further engineered HEK293T cells to present tumor specific antigen CD19 upon heat activation and study the killing by anti-CD19 CAR-T cells. We showed that engineered HEK293T cells presented CD19 upon thermal-activation and were recognized and attacked by anti-CD19 CAR-T cells, demonstrating the feasibility of the system in immunotherapy. In the future, with a well-established focused ultrasound system, we hope to induce temperature changes in the confined volume of tissue in vivo to address the on-target off-tumor toxicities associated with current CAR-T cell therapies.

Programmed cell death protein 1 (PD-1) and signal regulatory protein α (SIRPα) are receptors that transduce inhibitory signals to negatively regulate effector functions of immune cells, which were both utilized by cancers to evade and inhibit the immune response. Through protein engineering, integrated sensing and activating proteins (iSNAPs) were developed to replace the cytoplasmic domains of PD-1 and SIRPα receptors. With Fyn as the activating module, engineered Fyn-iSNAP replaced the cytoplasmic tail of PD-1, which was intended to reprogram PD-1/PD-L1 inhibitory signaling into activating Fyn signaling upon ligand binding. However, characterization of Fyn-iSNAP in HeLa cells demonstrated that this engineered protein failed to show desired functions. In the second chapter, in order to rewire SIRPα/CD47 inhibitory pathway, Shp2-iSNAPs and Syk-iSNAPs were developed and used to replace the original SIRPα cytoplasmic domain in our previous studies (data not published yet). In order to increase gene delivery efficiency, the sizes of the engineered SIRPα-iSNAPs were reduced. The expression of the truncated proteins in mouse macrophages led to enhanced phagocytosis of opsonized red blood cells (RBCs). Also, human promyelocytic leukemia cell derived macrophages (PLDMs) were engineered for the first time using the SIRPα-Shp2-iSNAPs and its truncated version, which demonstrated enhanced engulfment against opsonized RBCs as well as tumor cells. Therefore, these engineered proteins with activating module potentially offer a new approach for future cancer immunotherapy.

On the molecular level, the immune response begins in T-cells where lymphocyte-specific protein tyrosine kinase p56­Lck (Lck), a Src family kinase (SFK), is one of the first molecules involved in early T-cell activation at T-cell receptors (TCRs). A biosensor utilizing fluorescence resonance energy transfer (FRET) was designed to better understand and monitor Lck kinase activity during TCR activation by undergoing FRET change when an active Lck phosphorylates the biosensor. According to the in vitro data, the Lck FRET biosensor can also report proto-oncogene tyrosine-protein kinase p59­Fyn (Fyn) activity faster than to Lck activity, implying that the Fyn kinase has a higher activity level than the Lck kinase. However, mammalian HeLa cell data showed that the biosensor is more specific to Lck in a physiological setting where kinase-dead Lck(K273R) and Lck(Y394F) can induce a higher FRET change than Lck(WT). In J.CaM1.6 cells, Lck-deficit T-cells, kinase-dead Lck(K273R) did not elicit a FRET change, but kinase-dead Lck(Y394F) continued to elicit a higher biosensor response than Lck(WT). This suggests that Lck may have an adaptor function to recruit other Src family kinases (SFKs) for continuous phosphorylation on the biosensor.

Chemotherapy is the dominant treatment approach to many cancers. For decades, Gemcitabine (GEM) has been used as the first-line therapy for pancreatic ductal adenocarcinoma (PDAC), one of the most fatal solid tumor, but its competence is disappointingly constrained byintrinsic or adaptive resistance. The mechanisms beneath such resistance have been intensively studied with great efforts and its correlations with many genes and pathways have been found. However, our knowledge has not been unified to provide abundant information for major advances, since the whole landscape in which drug adaptation and gene expressions are associated is not yet clear. Moreover, little had we known about the initiation of adaptation, which makes it more difficult to define better therapies and achieve better clinical outcomes.

Recently, epigenetic alterations have become potential prognostic biomarkers and offered vast options for PDAC detection. Previously, we had observed different histone 3 lysine 9 trimethylation (H3K9me3) patterns among GEM sensitive and resistant PDAC cells during the GEM treatment, suggesting the role of histone modification in GEM resistance initiation and exhibition. In this paper, we would provide our analysis workflow to identify the candidate genes that might be modulated by H3K9me3 upon GEM treatment, and predict how they dynamically affect GEM sensitivity. Our results would be considered as a general framework to construct the regulatory network from epigenetic drug response to transcriptomic plasticity, and discover new opportunities for therapeutic development.

While chimeric antigen receptor (CAR) therapy has emerged as a promising method for cancer therapy by engineering autologous T cells to redirect them to the specific tumor-associate antigen and kill the tumor cells, these engineered T cells also have “on-target, off-tumor” toxicity, which can harm normal tissues and can be life-threatening. The non-specific activity inspires us to control CAR expression with high spatiotemporal precisions. Therefore, we present the design of heat inducible CAR, which can upregulate CAR expression after heat stimulation. The thesis introduces the study of heat inducible anti-prostate specific membrane antigen (PSMA) CAR T cells, which has been proved to be able to sense heat and produce CARs for the targeting of prostate cancer cells and triggering activation of T cells. The study can serve as a foundation for broader application in solid tumor therapy.

Histone proteins in chromatin undergo various modifications that have profound impacts on many cellular processes, including cell cycle control, cancer, senescence, X-inactivation, cell fate decisions, and stem cell differentiation. These histone marks do not occur isolated, but often occur mutually exclusive or concurrently. Despite the extensive research ongoing, a large part of the regulation of histone marks remained elusive due to the lack of powerful and efficient methods. Here, we present a method of constructing an epigenetic fluorescence energy resonance transfer (FRET) biosensor by tuning variables including the linker length and orientations between the fluorescent proteins in the case of a H3K27me3 FRET biosensor. It reveals that shortening the linker connecting the fluorescent protein pairs and binding partners could indeed increase the FRET change between the bound and unbound states in the biosensor. This key concept can be generally applied to optimize various different FRET biosensors, especially histone epigenetic FRET biosensors with binding domains that have relatively low binding affinities.

Despite its incredible potential in treating cancer, chimeric antigen receptor (CAR) T cell therapy carries the deadly risk of cytokine release syndrome. For CAR T cell therapy to be an effective therapeutic, it must be controlled in a spatiotemporal manner to prevent unwanted CAR activation. To address this, a cre-lox recombination system called the “TamPA-Cre” system was developed that requires both the presence of 4OHT, and blue light for recombination. Genetic cassettes containing CARs, which only get transcribed after recombination, were then inserted into T cells, so both blue light and 4OHT could induce the activation of CARs. First, plasmids for the TamPA-Cre system were developed by molecular cloning, and then transfected into mCherry-EGFP reporter HEK293T cells. Initial testing demonstrated that both blue light and 4OHT were required for recombination, confirming the AND logic. Second, the TamPA-Cre system was optimized by manipulating: the light stimulation pattern, the time to begin light stimulation after 4OHT addition, the molar ratio between the CreN and CreC halves, and the affinity of nMag. Next, CD19CAR genetic constructs were validated through coculture with CD19+ cells. Lastly, a Jurkat cell line containing the TamPA-Cre and CD19CAR construct showed the expression of CD69 only when these cells were: cocultured with CD19+ cells, given 4OHT, and stimulated with blue light. Overall, the TamPA-Cre system provides a platform for the spatiotemporal control of CAR T cells using AND logic.

With the emergence of various cell therapies, including CAR T-cell therapy and TCR T- cell therapy, it has become more important than ever to elucidate the underlying mechanism of T-cell activation with a higher temporal and spatial resolution for better predication and interpretation of laboratory and clinical results. Lymphocyte-specific kinase (Lck) is one of the molecules involved in T-cell proximal signaling right after TCR engagement with MHC on antigen-presenting cells (APC). Many studies have been done to elucidate its crystal structure and kinetics. However, a majority of these observations were based on Western blots and immunostaining, which could only demonstrate the endpoint effects of certain perturbations.

Seeing is believing. Here we present a FRET-based biosensor with an optimized Lck- specific substrate to visualize Lck dynamics in living cells with high spatial and temporal resolution. With this biosensor, we also observed the unique activation dynamics of LckY394F mutant, which was reported to have only minimal or no kinase activities, under CD3/CD28 coreceptor stimulation. It was noted that LckY394F was able to lower the basal activation level of Lck in resting T-cells while capable of transducing robust signals downstream when activated. We further presented preliminary results that LckYF-incorporated CAR constructs had superior in vitro killing capacity under physiologically relevant E:T ratios and designed future experiments to fully characterize and optimize this new design. We believe that this mutant Lck has huge potentials in CAR T-cell therapy as well as in off-the-shelf therapeutic product design.