A synthetic transcription platform for programmable gene expression in mammalian cells.

Precise, scalable, and sustainable control of genetic and cellular activities in mammalian cells is key to developing precision therapeutics and smart biomanufacturing. Here we create a highly tunable, modular, versatile CRISPR-based synthetic transcription system for the programmable control of gene expression and cellular phenotypes in mammalian cells. Genetic circuits consisting of well-characterized libraries of guide RNAs, binding motifs of synthetic operators, transcriptional activators, and additional genetic regulatory elements express mammalian genes in a highly predictable and tunable manner. We demonstrate the programmable control of reporter genes episomally and chromosomally, with up to 25-fold more activity than seen with the EF1α promoter, in multiple cell types. We use these circuits to program the secretion of human monoclonal antibodies and to control T-cell effector function marked by interferon-γ production. Antibody titers and interferon-γ concentrations significantly correlate with synthetic promoter strengths, providing a platform for programming gene expression and cellular function in diverse applications.

Systemic biodistribution and hepatocyte-specific gene editing with CRISPR/Cas9 using hyaluronic acid-based nanoparticles.

The goal of this study was to evaluate hepatocyte-specific gene editing, via systemic administration of hyaluronic acid (HA)-based nanoparticles in naïve CD-1 mice. Using HA-poly(ethylene imine) (HA-PEI) and HA-PEI-mannose nanoparticles with differential mannose density (1X and 2X), we have evaluated systemic biodistribution and hepatocyte-specific delivery using IVIS imaging and flow cytometry. Additionally, we have investigated hepatocyte-specific delivery and transfection of CRISPR/Cas9 gene editing plasmid and eGFP gene payload to integrate at the Rosa26 locus. IVIS imaging showed uptake of HA-PEI nanoparticles primarily by the liver, and with addition of mannose at different concentrations, the nanoparticles showed increased uptake in both the liver and spleen. HA-PEI-mannose nanoparticles showed 55-65% uptake by hepatocytes, along with uptake by resident macrophage regardless of the mannose concentration. One of two gRNA targets showed 15% genome editing and obtained similar results for all three nanoparticle formulations. Cells positive for our gene payload were greatest with HA-PEI-mannose-1X nanoparticles where 16.2% of cells were GFP positive. The results were encouraging as proof of concept for the development of a non-viral biodegradable and biocompatible polymeric delivery system for gene editing specifically targeting hepatocytes upon systemic administration.

Fc-GDF15 glyco-engineering and receptor binding affinity optimization for body weight regulation.

GDF15 is a distant TGF-ß family member that induces anorexia and weight loss. Due to its function, GDF15 has attracted attention as a potential therapeutic for the treatment of obesity and its associated metabolic diseases. However, the pharmacokinetic and physicochemical properties of GDF15 present several challenges for its development as a therapeutic, including a short half-life, high aggregation propensity, and protease susceptibility in serum. Here, we report the design, characterization and optimization of GDF15 in an Fc-fusion protein format with improved therapeutic properties. Using a structure-based engineering approach, we combined knob-into-hole Fc technology and N-linked glycosylation site mutagenesis for half-life extension, improved solubility and protease resistance. In addition, we identified a set of mutations at the receptor binding site of GDF15 that show increased GFRAL binding affinity and led to significant half-life extension. We also identified a single point mutation that increases p-ERK signaling activity and results in improved weight loss efficacy in vivo. Taken together, our findings allowed us to develop GDF15 in a new therapeutic format that demonstrates better efficacy and potential for improved manufacturability.

Discovery and optimization of a novel anti-GUCY2c x CD3 bispecific antibody for the treatment of solid tumors.

We report here the discovery and optimization of a novel T cell retargeting anti-GUCY2C x anti-CD3ε bispecific antibody for the treatment of solid tumors. Using a combination of hybridoma, phage display and rational design protein engineering, we have developed a fully humanized and manufacturable CD3 bispecific antibody that demonstrates favorable pharmacokinetic properties and potent in vivo efficacy. Anti-GUCY2C and anti-CD3ε antibodies derived from mouse hybridomas were first humanized into well-behaved human variable region frameworks with full retention of binding and T-cell mediated cytotoxic activity. To address potential manufacturability concerns, multiple approaches were taken in parallel to optimize and de-risk the two antibody variable regions. These approaches included structure-guided rational mutagenesis and phage display-based optimization, focusing on improving stability, reducing polyreactivity and self-association potential, removing chemical liabilities and proteolytic cleavage sites, and de-risking immunogenicity. Employing rapid library construction methods as well as automated phage display and high-throughput protein production workflows enabled efficient generation of an optimized bispecific antibody with desirable manufacturability properties, high stability, and low nonspecific binding. Proteolytic cleavage and deamidation in complementarity-determining regions were also successfully addressed. Collectively, these improvements translated to a molecule with potent single-agent in vivo efficacy in a tumor cell line adoptive transfer model and a cynomolgus monkey pharmacokinetic profile (half-life>4.5 days) suitable for clinical development. Clinical evaluation of PF-07062119 is ongoing.

Engineering protein-protein devices for multilayered regulation of mRNA translation using orthogonal proteases in mammalian cells.

The development of RNA-encoded regulatory circuits relying on RNA-binding proteins (RBPs) has enhanced the applicability and prospects of post-transcriptional synthetic network for reprogramming cellular functions. However, the construction of RNA-encoded multilayer networks is still limited by the availability of composable and orthogonal regulatory devices. Here, we report on control of mRNA translation with newly engineered RBPs regulated by viral proteases in mammalian cells. By combining post-transcriptional and post-translational control, we expand the operational landscape of RNA-encoded genetic circuits with a set of regulatory devices including: i) RBP-protease, ii) protease-RBP, iii) protease-protease, iv) protein sensor protease-RBP, and v) miRNA-protease/RBP interactions. The rational design of protease-regulated proteins provides a diverse toolbox for synthetic circuit regulation that enhances multi-input information processing-actuation of cellular responses. Our approach enables design of artificial circuits that can reprogram cellular function with potential benefits as research tools and for future in vivo therapeutics and biotechnological applications.

Engineering modular intracellular protein sensor-actuator devices.

Understanding and reshaping cellular behaviors with synthetic gene networks requires the ability to sense and respond to changes in the intracellular environment. Intracellular proteins are involved in almost all cellular processes, and thus can provide important information about changes in cellular conditions such as infections, mutations, or disease states. Here we report the design of a modular platform for intrabody-based protein sensing-actuation devices with transcriptional output triggered by detection of intracellular proteins in mammalian cells. We demonstrate reporter activation response (fluorescence, apoptotic gene) to proteins involved in hepatitis C virus (HCV) infection, human immunodeficiency virus (HIV) infection, and Huntington's disease, and show sensor-based interference with HIV-1 downregulation of HLA-I in infected T cells. Our method provides a means to link varying cellular conditions with robust control of cellular behavior for scientific and therapeutic applications.

A multi-landing pad DNA integration platform for mammalian cell engineering.

Engineering mammalian cell lines that stably express many transgenes requires the precise insertion of large amounts of heterologous DNA into well-characterized genomic loci, but current methods are limited. To facilitate reliable large-scale engineering of CHO cells, we identified 21 novel genomic sites that supported stable long-term expression of transgenes, and then constructed cell lines containing one, two or three 'landing pad' recombination sites at selected loci. By using a highly efficient BxB1 recombinase along with different selection markers at each site, we directed recombinase-mediated insertion of heterologous DNA to selected sites, including targeting all three with a single transfection. We used this method to controllably integrate up to nine copies of a monoclonal antibody, representing about 100 kb of heterologous DNA in 21 transcriptional units. Because the integration was targeted to pre-validated loci, recombinant protein expression remained stable for weeks and additional copies of the antibody cassette in the integrated payload resulted in a linear increase in antibody expression. Overall, this multi-copy site-specific integration platform allows for controllable and reproducible insertion of large amounts of DNA into stable genomic sites, which has broad applications for mammalian synthetic biology, recombinant protein production and biomanufacturing.

Mechanical Strain Promotes Oligodendrocyte Differentiation by Global Changes of Gene Expression.

Differentiation of oligodendrocyte progenitor cells (OPC) to oligodendrocytes and subsequent axon myelination are critical steps in vertebrate central nervous system (CNS) development and regeneration. Growing evidence supports the significance of mechanical factors in oligodendrocyte biology. Here, we explore the effect of mechanical strains within physiological range on OPC proliferation and differentiation, and strain-associated changes in chromatin structure, epigenetics, and gene expression. Sustained tensile strain of 10-15% inhibited OPC proliferation and promoted differentiation into oligodendrocytes. This response to strain required specific interactions of OPCs with extracellular matrix ligands. Applied strain induced changes in nuclear shape, chromatin organization, and resulted in enhanced histone deacetylation, consistent with increased oligodendrocyte differentiation. This response was concurrent with increased mRNA levels of the epigenetic modifier histone deacetylase Hdac11. Inhibition of HDAC proteins eliminated the strain-mediated increase of OPC differentiation, demonstrating a role of HDACs in mechanotransduction of strain to chromatin. RNA sequencing revealed global changes in gene expression associated with strain. Specifically, expression of multiple genes associated with oligodendrocyte differentiation and axon-oligodendrocyte interactions was increased, including cell surface ligands (Ncam, ephrins), cyto- and nucleo-skeleton genes (Fyn, actinins, myosin, nesprin, Sun1), transcription factors (Sox10, Zfp191, Nkx2.2), and myelin genes (Cnp, Plp, Mag). These findings show how mechanical strain can be transmitted to the nucleus to promote oligodendrocyte differentiation, and identify the global landscape of signaling pathways involved in mechanotransduction. These data provide a source of potential new therapeutic avenues to enhance OPC differentiation in vivo.

A novel Bxb1 integrase RMCE system for high fidelity site-specific integration of mAb expression cassette in CHO Cells.

As CHO cell line development for biotherapeutic production becomes more sophisticated through the availability of the CHO genome sequence, the ability to accurately and reproducibly engineer the host cell genome has become increasingly important. Multiple well characterized systems for site-specific integration will enable more complex cell line engineering to generate cell lines with desirable attributes. We built and characterized a novel recombinase mediated cassette exchange (RMCE) system using Bxb1 integrase and compared it to the commonly used Flp/FRT RMCE system. We first integrated a DNA construct flanked by either Bxb1 attachment sites or FRT sequences (referred to as a landing pad) into the Fer1L4 genomic locus of CHO-S cells using CRISPR/Cas9 mediated homologous recombination. We characterized the resulting clones harboring either the Bxb1 or Flp/FRT landing pad using whole genome resequencing to compare their genomes with the parental host cell line. We determined that each landing pad was specifically integrated into the Fer1L4 locus in the selected clones and observed no major structural changes in the genome or variations in copy number as a result of CRISPR/Cas9 modification. We subsequently tested the ability of the Bxb1 and Flp/FRT landing pad clones to perform proper RMCE with donor vectors containing identical mAb expression cassettes flanked by either Bxb1 attachment sites or FRT sites. We demonstrated that both RMCE systems were able to generate stable pools in a similar time frame with comparable mAb expression. Through genetic characterization of up to 24 clones derived from either system, we determined that the BxB1 RMCE system yielded higher fidelity RMCE events than the Flp/FRT system as evidenced by a higher percentage of clones with expected integration of the mAb cassette into the landing pad in the respective cell lines. We conclude that Bxb1 RMCE is an excellent alternative to Flp/FRT RMCE and valuable addition to our toolbox enabling the engineering of more sophisticated cell lines for biotherapeutic production. Biotechnol. Bioeng. 2017;114: 1837-1846. © 2017 Wiley Periodicals, Inc.

Mammalian synthetic circuits with RNA binding proteins for RNA-only delivery.

Synthetic regulatory circuits encoded in RNA rather than DNA could provide a means to control cell behavior while avoiding potentially harmful genomic integration in therapeutic applications. We create post-transcriptional circuits using RNA-binding proteins, which can be wired in a plug-and-play fashion to create networks of higher complexity. We show that the circuits function in mammalian cells when encoded in modified mRNA or self-replicating RNA.

A platform for rapid prototyping of synthetic gene networks in mammalian cells.

Mammalian synthetic biology may provide novel therapeutic strategies, help decipher new paths for drug discovery and facilitate synthesis of valuable molecules. Yet, our capacity to genetically program cells is currently hampered by the lack of efficient approaches to streamline the design, construction and screening of synthetic gene networks. To address this problem, here we present a framework for modular and combinatorial assembly of functional (multi)gene expression vectors and their efficient and specific targeted integration into a well-defined chromosomal context in mammalian cells. We demonstrate the potential of this framework by assembling and integrating different functional mammalian regulatory networks including the largest gene circuit built and chromosomally integrated to date (6 transcription units, 27kb) encoding an inducible memory device. Using a library of 18 different circuits as a proof of concept, we also demonstrate that our method enables one-pot/single-flask chromosomal integration and screening of circuit libraries. This rapid and powerful prototyping platform is well suited for comparative studies of genetic regulatory elements, genes and multi-gene circuits as well as facile development of libraries of isogenic engineered cell lines.

Systematic transfer of prokaryotic sensors and circuits to mammalian cells.

Prokaryotic regulatory proteins respond to diverse signals and represent a rich resource for building synthetic sensors and circuits. The TetR family contains >10(5) members that use a simple mechanism to respond to stimuli and bind distinct DNA operators. We present a platform that enables the transfer of these regulators to mammalian cells, which is demonstrated using human embryonic kidney (HEK293) and Chinese hamster ovary (CHO) cells. The repressors are modified to include nuclear localization signals (NLS) and responsive promoters are built by incorporating multiple operators. Activators are also constructed by modifying the protein to include a VP16 domain. Together, this approach yields 15 new regulators that demonstrate 19- to 551-fold induction and retain both the low levels of crosstalk in DNA binding specificity observed between the parent regulators in Escherichia coli, as well as their dynamic range of activity. By taking advantage of the DAPG small molecule sensing mediated by the PhlF repressor, we introduce a new inducible system with 50-fold induction and a threshold of 0.9 µM DAPG, which is comparable to the classic Dox-induced TetR system. A set of NOT gates is constructed from the new repressors and their response function quantified. Finally, the Dox- and DAPG- inducible systems and two new activators are used to build a synthetic enhancer (fuzzy AND gate), requiring the coordination of 5 transcription factors organized into two layers. This work introduces a generic approach for the development of mammalian genetic sensors and circuits to populate a toolbox that can be applied to diverse applications from biomanufacturing to living therapeutics.

Rapid, modular and reliable construction of complex mammalian gene circuits.

We developed a framework for quick and reliable construction of complex gene circuits for genetically engineering mammalian cells. Our hierarchical framework is based on a novel nucleotide addressing system for defining the position of each part in an overall circuit. With this framework, we demonstrate construction of synthetic gene circuits of up to 64 kb in size comprising 11 transcription units and 33 basic parts. We show robust gene expression control of multiple transcription units by small molecule inducers in human cells with transient transfection and stable chromosomal integration of these circuits. This framework enables development of complex gene circuits for engineering mammalian cells with unprecedented speed, reliability and scalability and should have broad applicability in a variety of areas including mammalian cell fermentation, cell fate reprogramming and cell-based assays.

Multi-input RNAi-based logic circuit for identification of specific cancer cells.

Engineered biological systems that integrate multi-input sensing, sophisticated information processing, and precisely regulated actuation in living cells could be useful in a variety of applications. For example, anticancer therapies could be engineered to detect and respond to complex cellular conditions in individual cells with high specificity. Here, we show a scalable transcriptional/posttranscriptional synthetic regulatory circuit--a cell-type "classifier"--that senses expression levels of a customizable set of endogenous microRNAs and triggers a cellular response only if the expression levels match a predetermined profile of interest. We demonstrate that a HeLa cancer cell classifier selectively identifies HeLa cells and triggers apoptosis without affecting non-HeLa cell types. This approach also provides a general platform for programmed responses to other complex cell states.

Learning Protein Folding Energy Functions.

A critical open problem in ab initio protein folding is protein energy function design, which pertains to defining the energy of protein conformations in a way that makes folding most efficient and reliable. In this paper, we address this issue as a weight optimization problem and utilize a machine learning approach, learning-to-rank, to solve this problem. We investigate the ranking-via-classification approach, especially the RankingSVM method and compare it with the state-of-the-art approach to the problem using the MINUIT optimization package. To maintain the physicality of the results, we impose non-negativity constraints on the weights. For this we develop two efficient non-negative support vector machine (NNSVM) methods, derived from L2-norm SVM and L1-norm SVMs, respectively. We demonstrate an energy function which maintains the correct ordering with respect to structure dissimilarity to the native state more often, is more efficient and reliable for learning on large protein sets, and is qualitatively superior to the current state-of-the-art energy function.

Protein model refinement using an optimized physics-based all-atom force field.

One of the greatest challenges in protein structure prediction is the refinement of low-resolution predicted models to high-resolution structures that are close to the native state. Although contemporary structure prediction methods can assemble the correct topology for a large fraction of protein domains, such approximate models are often not of the resolution required for many important applications, including studies of reaction mechanisms and virtual ligand screening. Thus, the development of a method that could bring those structures closer to the native state is of great importance. We recently optimized the relative weights of the components of the Amber ff03 potential on a large set of decoy structures to create a funnel-shaped energy landscape with the native structure at the global minimum. Such an energy function might be able to drive proteins toward their native structure. In this work, for a test set of 47 proteins, with 100 decoy structures per protein that have a range of structural similarities to the native state, we demonstrate that our optimized potential can drive protein models closer to their native structure. Comparing the lowest-energy structure from each trajectory with the starting decoy, structural improvement is seen for 70% of the models on average. The ability to do such systematic structural refinements by using a physics-based all-atom potential represents a promising approach to high-resolution structure prediction.

Development of a physics-based force field for the scoring and refinement of protein models.

The minimal requirements of a physics-based potential that can refine protein structures are the existence of a correlation between the energy with native similarity and the scoring of the native structure as the lowest in energy. To develop such a force field, the relative weights of the Amber ff03 all-atom potential supplemented by an explicit hydrogen-bond potential were adjusted by global optimization of energetic and structural criteria for a large set of protein decoys generated for a set of 58 nonhomologous proteins. The average correlation coefficient of the energy with TM-score significantly improved from 0.25 for the original ff03 potential to 0.65 for the optimized force field. The fraction of proteins for which the native structure had lowest energy increased from 0.22 to 0.90. Moreover, use of an explicit hydrogen-bond potential improves scoring performance of the force field. Promising preliminary results were obtained in applying the optimized potentials to refine protein decoys using only an energy criterion to choose the best decoy among sampled structures. For a set of seven proteins, 63% of the decoys improve, 18% get worse, and 19% are not changed.

Analysis of TASSER-based CASP7 protein structure prediction results.

An improved TASSER (Threading/ASSEmbly/Refinement) methodology is applied to predict the tertiary structure for all CASP7 targets. TASSER employs template identification by threading, followed by tertiary structure assembly by rearranging continuous template fragments, where conformational space is searched via Parallel Hyperbolic Monte Carlo sampling with an optimized force-field that includes knowledge-based statistical potentials and restraints derived from threading templates. The final models are selected by clustering structures from the low temperature replicas. Improvements in TASSER over CASP6 involve use of better templates from 3D-jury applied to three threading programs, PROSPECTOR_3, SP(3), and SPARKS, and a fragment comparison method for better model ranking. For targets with no reliable templates, a variant of TASSER (chunk-TASSER) is also applied with potentials and restraints extracted from ab initio folded supersecondary chunks of the target to build full-length models. For all 124 CASP targets/domains, the average root-mean-square-deviation (RMSD) from native and alignment coverage of the best initial threading models from 3D-jury are 6.2 A and 93%, respectively. Following TASSER reassembly, the average RMSD of the best model in the template aligned region decreases to 4.9 A and the average TM-score increases from 0.617 for the template to 0.678 for the best full-length model. Based on target difficulty, the average TM-scores of the final model to native are 0.904, 0.671, and 0.307 for high-accuracy template-based modeling, template-based modeling, and free modeling targets/domains, respectively. For the more difficult targets, TASSER with modest human intervention performed better in comparison to its server counterpart, MetaTASSER, which used a limited time simulation.

Can a physics-based, all-atom potential find a protein's native structure among misfolded structures? I. Large scale AMBER benchmarking.

Recent work has shown that physics-based, all-atom energy functions (AMBER, CHARMM, OPLS-AA) and local minimization, when used in scoring, are able to discriminate among native and decoy structures. Yet, there have been only few instances reported of the successful use of physics based potentials in the actual refinement of protein models from a starting conformation to one that ends in structures, which are closer to the native state. An energy function that has a global minimum energy in the protein's native state and a good correlation between energy and native-likeness should be able to drive model structures closer to their native structure during a conformational search. Here, the possible reasons for the discrepancy between the scoring and refinement results for the case of AMBER potential are examined. When the conformational search via molecular dynamics is driven by the AMBER potential for a large set of 150 nonhomologous proteins and their associated decoys, often the native minimum does not appear to be the lowest free energy state. Ways of correcting the potential function in order to make it more suitable for protein model refinement are proposed.