Profiling of drug resistance in Src kinase at scale uncovers a regulatory network coupling autoinhibition and catalytic domain dynamics.

Kinase inhibitors are effective cancer therapies, but resistance often limits clinical efficacy. Despite the cataloging of numerous resistance mutations, our understanding of kinase inhibitor resistance is still incomplete. Here, we comprehensively profiled the resistance of ∼3,500 Src tyrosine kinase mutants to four different ATP-competitive inhibitors. We found that ATP-competitive inhibitor resistance mutations are distributed throughout Src's catalytic domain. In addition to inhibitor contact residues, residues that participate in regulating Src's phosphotransferase activity were prone to the development of resistance. Unexpectedly, we found that a resistance-prone cluster of residues located on the top face of the N-terminal lobe of Src's catalytic domain contributes to autoinhibition by reducing catalytic domain dynamics, and mutations in this cluster led to resistance by lowering inhibitor affinity and promoting kinase hyperactivation. Together, our studies demonstrate how drug resistance profiling can be used to define potential resistance pathways and uncover new mechanisms of kinase regulation.

Molecular determinants of Hsp90 dependence of Src kinase revealed by deep mutational scanning.

Hsp90 is a molecular chaperone involved in the refolding and activation of numerous protein substrates referred to as clients. While the molecular determinants of Hsp90 client specificity are poorly understood and limited to a handful of client proteins, strong clients are thought to be destabilized and conformationally extended. Here, we measured the phosphotransferase activity of 3929 variants of the tyrosine kinase Src in both the presence and absence of an Hsp90 inhibitor. We identified 84 previously unknown functionally dependent client variants. Unexpectedly, many destabilized or extended variants were not functionally dependent on Hsp90. Instead, functionally dependent client variants were clustered in the αF pocket and ß1-ß2 strand regions of Src, which have yet to be described in driving Hsp90 dependence. Hsp90 dependence was also strongly correlated with kinase activity. We found that a combination of activation, global extension, and general conformational flexibility, primarily induced by variants at the αF pocket and ß1-ß2 strands, was necessary to render Src functionally dependent on Hsp90. Moreover, the degree of activation and flexibility required to transform Src into a functionally dependent client varied with variant location, suggesting that a combination of regulatory domain disengagement and catalytic domain flexibility are required for chaperone dependence. Thus, by studying the chaperone dependence of a massive number of variants, we highlight factors driving Hsp90 client specificity and propose a model of chaperone-kinase interactions.

Integrating thousands of PTEN variant activity and abundance measurements reveals variant subgroups and new dominant negatives in cancers.

BACKGROUND: PTEN is a multi-functional tumor suppressor protein regulating cell growth, immune signaling, neuronal function, and genome stability. Experimental characterization can help guide the clinical interpretation of the thousands of germline or somatic PTEN variants observed in patients. Two large-scale mutational datasets, one for PTEN variant intracellular abundance encompassing 4112 missense variants and one for lipid phosphatase activity encompassing 7244 variants, were recently published. The combined information from these datasets can reveal variant-specific phenotypes that may underlie various clinical presentations, but this has not been comprehensively examined, particularly for somatic PTEN variants observed in cancers. METHODS: Here, we add to these efforts by measuring the intracellular abundance of 764 new PTEN variants and refining abundance measurements for 3351 previously studied variants. We use this expanded and refined PTEN abundance dataset to explore the mutational patterns governing PTEN intracellular abundance, and then incorporate the phosphatase activity data to subdivide PTEN variants into four functionally distinct groups. RESULTS: This analysis revealed a set of highly abundant but lipid phosphatase defective variants that could act in a dominant-negative fashion to suppress PTEN activity. Two of these variants were, indeed, capable of dysregulating Akt signaling in cells harboring a WT PTEN allele. Both variants were observed in multiple breast or uterine tumors, demonstrating the disease relevance of these high abundance, inactive variants. CONCLUSIONS: We show that multidimensional, large-scale variant functional data, when paired with public cancer genomics datasets and follow-up assays, can improve understanding of uncharacterized cancer-associated variants, and provide better insights into how they contribute to oncogenesis.

Multiplexed measurement of variant abundance and activity reveals VKOR topology, active site and human variant impact.

Vitamin K epoxide reductase (VKOR) drives the vitamin K cycle, activating vitamin K-dependent blood clotting factors. VKOR is also the target of the widely used anticoagulant drug, warfarin. Despite VKOR's pivotal role in coagulation, its structure and active site remain poorly understood. In addition, VKOR variants can cause vitamin K-dependent clotting factor deficiency or alter warfarin response. Here, we used multiplexed, sequencing-based assays to measure the effects of 2,695 VKOR missense variants on abundance and 697 variants on activity in cultured human cells. The large-scale functional data, along with an evolutionary coupling analysis, supports a four transmembrane domain topology, with variants in transmembrane domains exhibiting strongly deleterious effects on abundance and activity. Functionally constrained regions of the protein define the active site, and we find that, of four conserved cysteines putatively critical for function, only three are absolutely required. Finally, 25% of human VKOR missense variants show reduced abundance or activity, possibly conferring warfarin sensitivity or causing disease.

Suppression of unwanted CRISPR-Cas9 editing by co-administration of catalytically inactivating truncated guide RNAs.

CRISPR-Cas9 nucleases are powerful genome engineering tools, but unwanted cleavage at off-target and previously edited sites remains a major concern. Numerous strategies to reduce unwanted cleavage have been devised, but all are imperfect. Here, we report that off-target sites can be shielded from the active Cas9•single guide RNA (sgRNA) complex through the co-administration of dead-RNAs (dRNAs), truncated guide RNAs that direct Cas9 binding but not cleavage. dRNAs can effectively suppress a wide-range of off-targets with minimal optimization while preserving on-target editing, and they can be multiplexed to suppress several off-targets simultaneously. dRNAs can be combined with high-specificity Cas9 variants, which often do not eliminate all unwanted editing. Moreover, dRNAs can prevent cleavage of homology-directed repair (HDR)-corrected sites, facilitating scarless editing by eliminating the need for blocking mutations. Thus, we enable precise genome editing by establishing a flexible approach for suppressing unwanted editing of both off-targets and HDR-corrected sites.

High-throughput, microscope-based sorting to dissect cellular heterogeneity.

Microscopy is a powerful tool for characterizing complex cellular phenotypes, but linking these phenotypes to genotype or RNA expression at scale remains challenging. Here, we present Visual Cell Sorting, a method that physically separates hundreds of thousands of live cells based on their visual phenotype. Automated imaging and phenotypic analysis directs selective illumination of Dendra2, a photoconvertible fluorescent protein expressed in live cells; these photoactivated cells are then isolated using fluorescence-activated cell sorting. First, we use Visual Cell Sorting to assess hundreds of nuclear localization sequence variants in a pooled format, identifying variants that improve nuclear localization and enabling annotation of nuclear localization sequences in thousands of human proteins. Second, we recover cells that retain normal nuclear morphologies after paclitaxel treatment, and then derive their single-cell transcriptomes to identify pathways associated with paclitaxel resistance in cancers. Unlike alternative methods, Visual Cell Sorting depends on inexpensive reagents and commercially available hardware. As such, it can be readily deployed to uncover the relationships between visual cellular phenotypes and internal states, including genotypes and gene expression programs.

An improved platform for functional assessment of large protein libraries in mammalian cells.

Multiplex genetic assays can simultaneously test thousands of genetic variants for a property of interest. However, limitations of existing multiplex assay methods in cultured mammalian cells hinder the breadth, speed and scale of these experiments. Here, we describe a series of improvements that greatly enhance the capabilities of a Bxb1 recombinase-based landing pad system for conducting different types of multiplex genetic assays in various mammalian cell lines. We incorporate the landing pad into a lentiviral vector, easing the process of generating new landing pad cell lines. We also develop several new landing pad versions, including one where the Bxb1 recombinase is expressed from the landing pad itself, improving recombination efficiency more than 2-fold and permitting rapid prototyping of transgenic constructs. Other versions incorporate positive and negative selection markers that enable drug-based enrichment of recombinant cells, enabling the use of larger libraries and reducing costs. A version with dual convergent promoters allows enrichment of recombinant cells independent of transgene expression, permitting the assessment of libraries of transgenes that perturb cell growth and survival. Lastly, we demonstrate these improvements by assessing the effects of a combinatorial library of oncogenes and tumor suppressors on cell growth. Collectively, these advancements make multiplex genetic assays in diverse cultured cell lines easier, cheaper and more effective, facilitating future studies probing how proteins impact cell function, using transgenic variant libraries tested individually or in combination.

Elucidating the Molecular Determinants of Aß Aggregation with Deep Mutational Scanning.

Despite the importance of Aß aggregation in Alzheimer's disease etiology, our understanding of the sequence determinants of aggregation is sparse and largely derived from in vitro studies. For example, in vitro proline and alanine scanning mutagenesis of Aß40 proposed core regions important for aggregation. However, we lack even this limited mutagenesis data for the more disease-relevant Aß42 Thus, to better understand the molecular determinants of Aß42 aggregation in a cell-based system, we combined a yeast DHFR aggregation assay with deep mutational scanning. We measured the effect of 791 of the 798 possible single amino acid substitutions on the aggregation propensity of Aß42 We found that ∼75% of substitutions, largely to hydrophobic residues, maintained or increased aggregation. We identified 11 positions at which substitutions, particularly to hydrophilic and charged amino acids, disrupted Aß aggregation. These critical positions were similar but not identical to critical positions identified in previous Aß mutagenesis studies. Finally, we analyzed our large-scale mutagenesis data in the context of different Aß aggregate structural models, finding that the mutagenesis data agreed best with models derived from fibrils seeded using brain-derived Aß aggregates.

A Combined Approach Reveals a Regulatory Mechanism Coupling Src's Kinase Activity, Localization, and Phosphotransferase-Independent Functions.

Multiple layers of regulation modulate the activity and localization of protein kinases. However, many details of kinase regulation remain incompletely understood. Here, we apply saturation mutagenesis and a chemical genetic method for allosterically modulating kinase global conformation to Src kinase, providing insight into known regulatory mechanisms and revealing a previously undiscovered interaction between Src's SH4 and catalytic domains. Abrogation of this interaction increased phosphotransferase activity, promoted membrane association, and provoked phosphotransferase-independent alterations in cell morphology. Thus, Src's SH4 domain serves as an intramolecular regulator coupling catalytic activity, global conformation, and localization, as well as mediating a phosphotransferase-independent function. Sequence conservation suggests that the SH4 domain regulatory interaction exists in other Src-family kinases. Our combined approach's ability to reveal a regulatory mechanism in one of the best-studied kinases suggests that it could be applied broadly to provide insight into kinase structure, regulation, and function.

Multiplex assessment of protein variant abundance by massively parallel sequencing.

Determining the pathogenicity of genetic variants is a critical challenge, and functional assessment is often the only option. Experimentally characterizing millions of possible missense variants in thousands of clinically important genes requires generalizable, scalable assays. We describe variant abundance by massively parallel sequencing (VAMP-seq), which measures the effects of thousands of missense variants of a protein on intracellular abundance simultaneously. We apply VAMP-seq to quantify the abundance of 7,801 single-amino-acid variants of PTEN and TPMT, proteins in which functional variants are clinically actionable. We identify 1,138 PTEN and 777 TPMT variants that result in low protein abundance, and may be pathogenic or alter drug metabolism, respectively. We observe selection for low-abundance PTEN variants in cancer, and show that p.Pro38Ser, which accounts for ~10% of PTEN missense variants in melanoma, functions via a dominant-negative mechanism. Finally, we demonstrate that VAMP-seq is applicable to other genes, highlighting its generalizability.

Rheostatic Control of Cas9-Mediated DNA Double Strand Break (DSB) Generation and Genome Editing.

We recently reported two novel tools for precisely controlling and quantifying Cas9 activity: a chemically inducible Cas9 variant (ciCas9) that can be rapidly activated by small molecules and a ddPCR assay for time-resolved measurement of DNA double strand breaks (DSB-ddPCR). Here, we further demonstrate the potential of ciCas9 to function as a tunable rheostat for Cas9 function. We show that a new highly potent and selective small molecule activator paired with a more tightly regulated ciCas9 variant expands the range of accessible Cas9 activity levels. We subsequently demonstrate that ciCas9 activity levels can be dose-dependently tuned with a small molecule activator, facilitating rheostatic time-course experiments. These studies provide the first insight into how Cas9-mediated DSB levels correlate with overall editing efficiency. Thus, we demonstrate that ciCas9 and our DSB-ddPCR assay permit the time-resolved study of Cas9 DSB generation and genome editing kinetics at a wide range of Cas9 activity levels.

Rapidly inducible Cas9 and DSB-ddPCR to probe editing kinetics.

We developed a chemically inducible Cas9 (ciCas9) and a droplet digital PCR assay for double-strand breaks (DSB-ddPCR) to investigate the kinetics of Cas9-mediated generation and repair of DSBs in cells. ciCas9 is a rapidly activated, single-component Cas9 variant engineered by replacing the protein's REC2 domain with the BCL-xL protein and fusing an interacting BH3 peptide to the C terminus. ciCas9 can be tunably activated by a compound that disrupts the BCL-xL-BH3 interaction within minutes. DSB-ddPCR demonstrates time-resolved, highly quantitative, and targeted measurement of DSBs. Combining these tools facilitated an unprecedented exploration of the kinetics of Cas9-mediated DNA cleavage and repair. We find that sgRNAs targeting different sites generally induce cleavage within minutes and repair within 1 or 2 h. However, we observe distinct kinetic profiles, even for proximal sites, and this suggests that target sequence and chromatin state modulate cleavage and repair kinetics.

A platform for functional assessment of large variant libraries in mammalian cells.

Sequencing-based, massively parallel genetic assays have revolutionized our ability to quantify the relationship between many genotypes and a phenotype of interest. Unfortunately, variant library expression platforms in mammalian cells are far from ideal, hindering the study of human gene variants in their physiologically relevant cellular contexts. Here, we describe a platform for phenotyping variant libraries in transfectable mammalian cell lines in two steps. First, a landing pad cell line with a genomically integrated, Tet-inducible cassette containing a Bxb1 recombination site is created. Second, a single variant from a library of transfected, promoter-less plasmids is recombined into the landing pad in each cell. Thus, every cell in the recombined pool expresses a single variant, allowing for parallel, sequencing-based assessment of variant effect. We describe a method for incorporating a single landing pad into a defined site of a cell line of interest, and show that our approach can be used generate more than 20 000 recombinant cells in a single experiment. Finally, we use our platform in combination with a sequencing-based assay to explore the N-end rule by simultaneously measuring the effects of all possible N-terminal amino acids on protein expression.

Measuring the activity of protein variants on a large scale using deep mutational scanning.

Deep mutational scanning marries selection for protein function to high-throughput DNA sequencing in order to quantify the activity of variants of a protein on a massive scale. First, an appropriate selection system for the protein function of interest is identified and validated. Second, a library of variants is created, introduced into the selection system and subjected to selection. Third, library DNA is recovered throughout the selection and deep-sequenced. Finally, a functional score for each variant is calculated on the basis of the change in the frequency of the variant during the selection. This protocol describes the steps that must be carried out to generate a large-scale mutagenesis data set consisting of functional scores for up to hundreds of thousands of variants of a protein of interest. Establishing an assay, generating a library of variants and carrying out a selection and its accompanying sequencing takes on the order of 4-6 weeks; the initial data analysis can be completed in 1 week.

Engineering A-kinase anchoring protein (AKAP)-selective regulatory subunits of protein kinase A (PKA) through structure-based phage selection.

PKA is retained within distinct subcellular environments by the association of its regulatory type II (RII) subunits with A-kinase anchoring proteins (AKAPs). Conventional reagents that universally disrupt PKA anchoring are patterned after a conserved AKAP motif. We introduce a phage selection procedure that exploits high-resolution structural information to engineer RII mutants that are selective for a particular AKAP. Selective RII (RSelect) sequences were obtained for eight AKAPs following competitive selection screening. Biochemical and cell-based experiments validated the efficacy of RSelect proteins for AKAP2 and AKAP18. These engineered proteins represent a new class of reagents that can be used to dissect the contributions of different AKAP-targeted pools of PKA. Molecular modeling and high-throughput sequencing analyses revealed the molecular basis of AKAP-selective interactions and shed new light on native RII-AKAP interactions. We propose that this structure-directed evolution strategy might be generally applicable for the investigation of other protein interaction surfaces.

Suppression of statin effectiveness by copper and zinc in yeast and human cells.

Lovastatin and other statins inhibit HMG-CoA reductase, which carries out an early step in the sterol biosynthesis pathway. Statins lower cholesterol and are widely prescribed to prevent heart disease, but like many drugs, they can interact with nutritionally acquired metabolites. To probe these interactions, we explored the effect of a diverse library of metabolites on statin effectiveness using a Saccharomyces cerevisiae model. In yeast, treatment with lovastatin results in reduced growth. We combined lovastatin with the library of metabolites, and found that copper and zinc ions impaired the ability of the statin to inhibit yeast growth. Using an integrated genomic and metabolomic approach, we found that lovastatin plus metal synergistically upregulated some sterol biosynthesis genes. This altered pattern of gene expression resulted in greater flux through the sterol biosynthesis pathway and an increase in ergosterol levels. Each sterol intermediate level was correlated with expression of the upstream gene. Thus, the ergosterol biosynthetic response induced by statin is enhanced by copper and zinc. In cultured mammalian cells, these metals also rescued statin growth inhibition. Because copper and zinc impair the ability of statin to reduce sterol biosynthesis, dietary intake of these metals could have clinical relevance for statin treatment in humans.

High-resolution mapping of protein sequence-function relationships.

We present a large-scale approach to investigate the functional consequences of sequence variation in a protein. The approach entails the display of hundreds of thousands of protein variants, moderate selection for activity and high-throughput DNA sequencing to quantify the performance of each variant. Using this strategy, we tracked the performance of >600,000 variants of a human WW domain after three and six rounds of selection by phage display for binding to its peptide ligand. Binding properties of these variants defined a high-resolution map of mutational preference across the WW domain; each position had unique features that could not be captured by a few representative mutations. Our approach could be applied to many in vitro or in vivo protein assays, providing a general means for understanding how protein function relates to sequence.

CD8+ T cells from SIV elite controller macaques recognize Mamu-B*08-bound epitopes and select for widespread viral variation.

BACKGROUND: It is generally accepted that CD8+ T cell responses play an important role in control of immunodeficiency virus replication. The association of HLA-B27 and -B57 with control of viremia supports this conclusion. However, specific correlates of viral control in individuals expressing these alleles have been difficult to define. We recently reported that transient in vivo CD8+ cell depletion in simian immunodeficiency virus (SIV)-infected elite controller (EC) macaques resulted in a brief period of viral recrudescence. SIV replication was rapidly controlled with the reappearance of CD8+ cells, implicating that these cells actively suppress viral replication in ECs. METHODS AND FINDINGS: Here we show that three ECs in that study made at least seven robust CD8+ T cell responses directed against novel epitopes in Vif, Rev, and Nef restricted by the MHC class I molecule Mamu-B*08. Two of these Mamu-B*08-positive animals subsequently lost control of SIV replication. Their breakthrough virus harbored substitutions in multiple Mamu-B*08-restricted epitopes. Indeed, we found evidence for selection pressure mediated by Mamu-B*08-restricted CD8+ T cells in all of the newly identified epitopes in a cohort of chronically infected macaques. CONCLUSIONS: Together, our data suggest that Mamu-B*08-restricted CD8+ T cell responses effectively control replication of pathogenic SIV(mac)239. All seven regions encoding Mamu-B*08-restricted CD8+ T cell epitopes also exhibit amino acid replacements typically seen only in the presence of Mamu-B*08, suggesting that the variation we observe is indeed selected by CD8+ T cell responses. SIV(mac)239 infection of Indian rhesus macaques expressing Mamu-B*08 may therefore provide an animal model for understanding CD8+ T cell-mediated control of HIV replication in humans.

Gag-specific CD8+ T lymphocytes recognize infected cells before AIDS-virus integration and viral protein expression.

CD8(+) T cells are a key focus of vaccine development efforts for HIV. However, there is no clear consensus as to which of the nine HIV proteins should be used for vaccination. The early proteins Tat, Rev, and Nef may be better CD8(+) T cell targets than the late-expressed structural proteins Gag, Pol, and Env. In this study, we show that Gag-specific CD8(+) T cells recognize infected CD4(+) T lymphocytes as early as 2 h postinfection, before proviral DNA integration, viral protein synthesis, and Nef-mediated MHC class I down-regulation. Additionally, the number of Gag epitopes recognized by CD8(+) T cells was significantly associated with lower viremia (p = 0.0017) in SIV-infected rhesus macaques. These results suggest that HIV vaccines should focus CD8(+) T cell responses on Gag.

The antiviral efficacy of simian immunodeficiency virus-specific CD8+ T cells is unrelated to epitope specificity and is abrogated by viral escape.

CD8(+) T lymphocytes appear to play a role in controlling human immunodeficiency virus (HIV) replication, yet routine immunological assays do not measure the antiviral efficacy of these cells. Furthermore, it has been suggested that CD8+ T cells that recognize epitopes derived from proteins expressed early in the viral replication cycle can be highly efficient. We used a functional in vitro assay to assess the abilities of different epitope-specific CD8+ T-cell lines to control simian immunodeficiency virus (SIV) replication. We compared the antiviral efficacies of 26 epitope-specific CD8+ T-cell lines directed against seven SIV epitopes in Tat, Nef, Gag, Env, and Vif that were restricted by either Mamu-A*01 or Mamu-A*02. Suppression of SIV replication varied depending on the epitope specificities of the CD8+ T cells and was unrelated to whether the targeted epitope was derived from an early or late viral protein. Tat(28-35)SL8- and Gag(181-189)CM9-specific CD8+ T-cell lines were consistently superior at suppressing viral replication compared to the other five SIV-specific CD8+ T-cell lines. We also investigated the impact of viral escape on antiviral efficacy by determining if Tat(28-35)SL8- and Gag(181-189)CM9-specific CD8+ T-cell lines could suppress the replication of an escaped virus. Viral escape abrogated the abilities of Tat(28-35)SL8- and Gag(181-189)CM9-specific CD8+ T cells to control viral replication. However, gamma interferon (IFN-gamma) enzyme-linked immunospot and IFN-gamma/tumor necrosis factor alpha intracellular-cytokine-staining assays detected cross-reactive immune responses against the Gag escape variant. Understanding antiviral efficacy and epitope variability, therefore, will be important in selecting candidate epitopes for an HIV vaccine.