An atlas of protein homo-oligomerization across domains of life.

Protein structures are essential to understanding cellular processes in molecular detail. While advances in artificial intelligence revealed the tertiary structure of proteins at scale, their quaternary structure remains mostly unknown. We devise a scalable strategy based on AlphaFold2 to predict homo-oligomeric assemblies across four proteomes spanning the tree of life. Our results suggest that approximately 45% of an archaeal proteome and a bacterial proteome and 20% of two eukaryotic proteomes form homomers. Our predictions accurately capture protein homo-oligomerization, recapitulate megadalton complexes, and unveil hundreds of homo-oligomer types, including three confirmed experimentally by structure determination. Integrating these datasets with omics information suggests that a majority of known protein complexes are symmetric. Finally, these datasets provide a structural context for interpreting disease mutations and reveal coiled-coil regions as major enablers of quaternary structure evolution in human. Our strategy is applicable to any organism and provides a comprehensive view of homo-oligomerization in proteomes.

Computational design of soluble and functional membrane protein analogues.

De novo design of complex protein folds using solely computational means remains a substantial challenge1. Here we use a robust deep learning pipeline to design complex folds and soluble analogues of integral membrane proteins. Unique membrane topologies, such as those from G-protein-coupled receptors2, are not found in the soluble proteome, and we demonstrate that their structural features can be recapitulated in solution. Biophysical analyses demonstrate the high thermal stability of the designs, and experimental structures show remarkable design accuracy. The soluble analogues were functionalized with native structural motifs, as a proof of concept for bringing membrane protein functions to the soluble proteome, potentially enabling new approaches in drug discovery. In summary, we have designed complex protein topologies and enriched them with functionalities from membrane proteins, with high experimental success rates, leading to a de facto expansion of the functional soluble fold space.

De novo design of protein interactions with learned surface fingerprints.

Physical interactions between proteins are essential for most biological processes governing life1. However, the molecular determinants of such interactions have been challenging to understand, even as genomic, proteomic and structural data increase. This knowledge gap has been a major obstacle for the comprehensive understanding of cellular protein-protein interaction networks and for the de novo design of protein binders that are crucial for synthetic biology and translational applications2-9. Here we use a geometric deep-learning framework operating on protein surfaces that generates fingerprints to describe geometric and chemical features that are critical to drive protein-protein interactions10. We hypothesized that these fingerprints capture the key aspects of molecular recognition that represent a new paradigm in the computational design of novel protein interactions. As a proof of principle, we computationally designed several de novo protein binders to engage four protein targets: SARS-CoV-2 spike, PD-1, PD-L1 and CTLA-4. Several designs were experimentally optimized, whereas others were generated purely in silico, reaching nanomolar affinity with structural and mutational characterization showing highly accurate predictions. Overall, our surface-centric approach captures the physical and chemical determinants of molecular recognition, enabling an approach for the de novo design of protein interactions and, more broadly, of artificial proteins with function.

Antibody-peptide conjugates deliver covalent inhibitors blocking oncogenic cathepsins.

Cysteine cathepsins are a family of proteases that are relevant therapeutic targets for the treatment of different cancers and other diseases. However, no clinically approved drugs for these proteins exist, as their systemic inhibition can induce deleterious side effects. To address this problem, we developed a modular antibody-based platform for targeted drug delivery by conjugating non-natural peptide inhibitors (NNPIs) to antibodies. NNPIs were functionalized with reactive warheads for covalent inhibition, optimized with deep saturation mutagenesis and conjugated to antibodies to enable cell-type-specific delivery. Our antibody-peptide inhibitor conjugates specifically blocked the activity of cathepsins in different cancer cells, as well as osteoclasts, and showed therapeutic efficacy in vitro and in vivo. Overall, our approach allows for the rapid design of selective cathepsin inhibitors and can be generalized to inhibit a broad class of proteases in cancer and other diseases.

Bottom-up de novo design of functional proteins with complex structural features.

De novo protein design has enabled the creation of new protein structures. However, the design of functional proteins has proved challenging, in part due to the difficulty of transplanting structurally complex functional sites to available protein structures. Here, we used a bottom-up approach to build de novo proteins tailored to accommodate structurally complex functional motifs. We applied the bottom-up strategy to successfully design five folds for four distinct binding motifs, including a bifunctionalized protein with two motifs. Crystal structures confirmed the atomic-level accuracy of the computational designs. These de novo proteins were functional as components of biosensors to monitor antibody responses and as orthogonal ligands to modulate synthetic signaling receptors in engineered mammalian cells. Our work demonstrates the potential of bottom-up approaches to accommodate complex structural motifs, which will be essential to endow de novo proteins with elaborate biochemical functions, such as molecular recognition or catalysis.

Computational design of anti-CRISPR proteins with improved inhibition potency.

Anti-CRISPR (Acr) proteins are powerful tools to control CRISPR-Cas technologies. However, the available Acr repertoire is limited to naturally occurring variants. Here, we applied structure-based design on AcrIIC1, a broad-spectrum CRISPR-Cas9 inhibitor, to improve its efficacy on different targets. We first show that inserting exogenous protein domains into a selected AcrIIC1 surface site dramatically enhances inhibition of Neisseria meningitidis (Nme)Cas9. Then, applying structure-guided design to the Cas9-binding surface, we converted AcrIIC1 into AcrIIC1X, a potent inhibitor of the Staphylococcus aureus (Sau)Cas9, an orthologue widely applied for in vivo genome editing. Finally, to demonstrate the utility of AcrIIC1X for genome engineering applications, we implemented a hepatocyte-specific SauCas9 ON-switch by placing AcrIIC1X expression under regulation of microRNA-122. Our work introduces designer Acrs as important biotechnological tools and provides an innovative strategy to safeguard CRISPR technologies.

Single-molecule kinetic analysis of HP1-chromatin binding reveals a dynamic network of histone modification and DNA interactions.

Chromatin recruitment of effector proteins involved in gene regulation depends on multivalent interaction with histone post-translational modifications (PTMs) and structural features of the chromatin fiber. Due to the complex interactions involved, it is currently not understood how effectors dynamically sample the chromatin landscape. Here, we dissect the dynamic chromatin interactions of a family of multivalent effectors, heterochromatin protein 1 (HP1) proteins, using single-molecule fluorescence imaging and computational modeling. We show that the three human HP1 isoforms are recruited and retained on chromatin by a dynamic exchange between histone PTM and DNA bound states. These interactions depend on local chromatin structure, the HP1 isoforms as well as on PTMs on HP1 itself. Of the HP1 isoforms, HP1α exhibits the longest residence times and fastest binding rates due to DNA interactions in addition to PTM binding. HP1α phosphorylation further increases chromatin retention through strengthening of multivalency while reducing DNA binding. As DNA binding in combination with specific PTM recognition is found in many chromatin effectors, we propose a general dynamic capture mechanism for effector recruitment. Multiple weak protein and DNA interactions result in a multivalent interaction network that targets effectors to a specific chromatin modification state, where their activity is required.

Iron bioavailability from fresh cheese fortified with iron-enriched yeast.

PURPOSE: An iron-enriched yeast able to lyse at body temperature was developed for iron fortification of chilled dairy products. The aim was to evaluate iron (Fe) absorption from iron-enriched yeast or ferrous sulfate added to fresh cheese. METHODS: Two stable isotope studies with a crossover design were conducted in 32 young women. Fe absorption from fresh cheese fortified with iron-enriched yeast (2.5 mg 58Fe) was compared to that from ferrous sulfate (2.5 mg 57Fe) when ingested with fresh cheese alone or with fresh cheese consumed with bread and butter. Iron absorption was determined based on erythrocyte incorporation of isotopic labels 14 days after consumption of the last test meal. RESULTS: Geometric mean fractional iron absorption from fresh cheese fortified with iron-enriched yeast consumed alone was significantly lower than from the cheese fortified with FeSO4 (20.5 vs. 28.7 %; p = 0.0007). When the fresh cheese was consumed with bread and butter, iron absorption from both fortificants decreased to 6.9 % from the iron-enriched yeast compared to 8.4 % from ferrous sulfate. The relative bioavailability of the iron-enriched yeast compared to ferrous sulfate was 0.72 for the cheese consumed alone and 0.82 for cheese consumed with bread and butter (p = 0.157). CONCLUSIONS: Iron from iron-enriched yeast was 72-82 % as well absorbed as ferrous sulfate indicating that the yeast lysed during digestion and released its iron.

c-Abl phosphorylates α-synuclein and regulates its degradation: implication for α-synuclein clearance and contribution to the pathogenesis of Parkinson's disease.

Increasing evidence suggests that the c-Abl protein tyrosine kinase could play a role in the pathogenesis of Parkinson's disease (PD) and other neurodegenerative disorders. c-Abl has been shown to regulate the degradation of two proteins implicated in the pathogenesis of PD, parkin and α-synuclein (α-syn). The inhibition of parkin's neuroprotective functions is regulated by c-Abl-mediated phosphorylation of parkin. However, the molecular mechanisms by which c-Abl activity regulates α-syn toxicity and clearance remain unknown. Herein, using NMR spectroscopy, mass spectrometry, in vitro enzymatic assays and cell-based studies, we established that α-syn is a bona fide substrate for c-Abl. In vitro studies demonstrate that c-Abl directly interacts with α-syn and catalyzes its phosphorylation mainly at tyrosine 39 (pY39) and to a lesser extent at tyrosine 125 (pY125). Analysis of human brain tissues showed that pY39 α-syn is detected in the brains of healthy individuals and those with PD. However, only c-Abl protein levels were found to be upregulated in PD brains. Interestingly, nilotinib, a specific inhibitor of c-Abl kinase activity, induces α-syn protein degradation via the autophagy and proteasome pathways, whereas the overexpression of α-syn in the rat midbrains enhances c-Abl expression. Together, these data suggest that changes in c-Abl expression, activation and/or c-Abl-mediated phosphorylation of Y39 play a role in regulating α-syn clearance and contribute to the pathogenesis of PD.

Dissection of the BCR-ABL signaling network using highly specific monobody inhibitors to the SHP2 SH2 domains.

The dysregulated tyrosine kinase BCR-ABL causes chronic myelogenous leukemia in humans and forms a large multiprotein complex that includes the Src-homology 2 (SH2) domain-containing phosphatase 2 (SHP2). The expression of SHP2 is necessary for BCR-ABL-dependent oncogenic transformation, but the precise signaling mechanisms of SHP2 are not well understood. We have developed binding proteins, termed monobodies, for the N- and C-terminal SH2 domains of SHP2. Intracellular expression followed by interactome analysis showed that the monobodies are essentially monospecific to SHP2. Two crystal structures revealed that the monobodies occupy the phosphopeptide-binding sites of the SH2 domains and thus can serve as competitors of SH2-phosphotyrosine interactions. Surprisingly, the segments of both monobodies that bind to the peptide-binding grooves run in the opposite direction to that of canonical phosphotyrosine peptides, which may contribute to their exquisite specificity. When expressed in cells, monobodies targeting the N-SH2 domain disrupted the interaction of SHP2 with its upstream activator, the Grb2-associated binder 2 adaptor protein, suggesting decoupling of SHP2 from the BCR-ABL protein complex. Inhibition of either N-SH2 or C-SH2 was sufficient to inhibit two tyrosine phosphorylation events that are critical for SHP2 catalytic activity and to block ERK activation. In contrast, targeting the N-SH2 or C-SH2 revealed distinct roles of the two SH2 domains in downstream signaling, such as the phosphorylation of paxillin and signal transducer and activator of transcription 5. Our results delineate a hierarchy of function for the SH2 domains of SHP2 and validate monobodies as potent and specific antagonists of protein-protein interactions in cancer cells.

Computational design of soluble functional analogues of integral membrane proteins.

De novo design of complex protein folds using solely computational means remains a significant challenge. Here, we use a robust deep learning pipeline to design complex folds and soluble analogues of integral membrane proteins. Unique membrane topologies, such as those from GPCRs, are not found in the soluble proteome and we demonstrate that their structural features can be recapitulated in solution. Biophysical analyses reveal high thermal stability of the designs and experimental structures show remarkable design accuracy. The soluble analogues were functionalized with native structural motifs, standing as a proof-of-concept for bringing membrane protein functions to the soluble proteome, potentially enabling new approaches in drug discovery. In summary, we designed complex protein topologies and enriched them with functionalities from membrane proteins, with high experimental success rates, leading to a de facto expansion of the functional soluble fold space.

Rational Design of Chemically Controlled Antibodies and Protein Therapeutics.

Protein-based therapeutics, such as monoclonal antibodies and cytokines, are important therapies for various pathophysiological conditions such as oncology, autoimmune disorders, and viral infections. However, the wide application of such protein therapeutics is often hindered by dose-limiting toxicities and adverse effects, namely, cytokine storm syndrome, organ failure, and others. Therefore, spatiotemporal control of the activities of these proteins is crucial to further expand their application. Here, we report the design and application of small-molecule-controlled switchable protein therapeutics by taking advantage of a previously engineered OFF-switch system. We used the Rosetta modeling suite to computationally optimize the affinity between B-cell lymphoma 2 (Bcl-2) protein and a previously developed computationally designed protein partner (LD3) to obtain a fast and efficient heterodimer disruption upon the addition of a competing drug (Venetoclax). The incorporation of the engineered OFF-switch system into anti-CTLA4, anti-HER2 antibodies, or an Fc-fused IL-15 cytokine demonstrated an efficient disruption in vitro, as well as fast clearance in vivo upon the addition of the competing drug Venetoclax. These results provide a proof-of-concept for the rational design of controllable biologics by introducing a drug-induced OFF-switch into existing protein-based therapeutics.

A rational blueprint for the design of chemically-controlled protein switches.

Small-molecule responsive protein switches are crucial components to control synthetic cellular activities. However, the repertoire of small-molecule protein switches is insufficient for many applications, including those in the translational spaces, where properties such as safety, immunogenicity, drug half-life, and drug side-effects are critical. Here, we present a computational protein design strategy to repurpose drug-inhibited protein-protein interactions as OFF- and ON-switches. The designed binders and drug-receptors form chemically-disruptable heterodimers (CDH) which dissociate in the presence of small molecules. To design ON-switches, we converted the CDHs into a multi-domain architecture which we refer to as activation by inhibitor release switches (AIR) that incorporate a rationally designed drug-insensitive receptor protein. CDHs and AIRs showed excellent performance as drug responsive switches to control combinations of synthetic circuits in mammalian cells. This approach effectively expands the chemical space and logic responses in living cells and provides a blueprint to develop new ON- and OFF-switches.

Btk SH2-kinase interface is critical for allosteric kinase activation and its targeting inhibits B-cell neoplasms.

Bruton's tyrosine kinase (Btk) is critical for B-cell maturation and activation. Btk loss-of-function mutations cause human X-linked agammaglobulinemia (XLA). In contrast, Btk signaling sustains growth of several B-cell neoplasms which may be treated with tyrosine kinase inhibitors (TKIs). Here, we uncovered the structural mechanism by which certain XLA mutations in the SH2 domain strongly perturb Btk activation. Using a combination of molecular dynamics (MD) simulations and small-angle X-ray scattering (SAXS), we discovered an allosteric interface between the SH2 and kinase domain required for Btk activation and to which multiple XLA mutations map. As allosteric interactions provide unique targeting opportunities, we developed an engineered repebody protein binding to the SH2 domain and able to disrupt the SH2-kinase interaction. The repebody prevents activation of wild-type and TKI-resistant Btk, inhibiting Btk-dependent signaling and proliferation of malignant B-cells. Therefore, the SH2-kinase interface is critical for Btk activation and a targetable site for allosteric inhibition.

De novo protein design enables the precise induction of RSV-neutralizing antibodies.

De novo protein design has been successful in expanding the natural protein repertoire. However, most de novo proteins lack biological function, presenting a major methodological challenge. In vaccinology, the induction of precise antibody responses remains a cornerstone for next-generation vaccines. Here, we present a protein design algorithm called TopoBuilder, with which we engineered epitope-focused immunogens displaying complex structural motifs. In both mice and nonhuman primates, cocktails of three de novo-designed immunogens induced robust neutralizing responses against the respiratory syncytial virus. Furthermore, the immunogens refocused preexisting antibody responses toward defined neutralization epitopes. Overall, our design approach opens the possibility of targeting specific epitopes for the development of vaccines and therapeutic antibodies and, more generally, will be applicable to the design of de novo proteins displaying complex functional motifs.

Rapid Screen for Tyrosine Kinase Inhibitor Resistance Mutations and Substrate Specificity.

We present a rapid and high-throughput yeast and flow cytometry based method for predicting kinase inhibitor resistance mutations and determining kinase peptide substrate specificity. Despite the widespread success of targeted kinase inhibitors as cancer therapeutics, resistance mutations arising within the kinase domain of an oncogenic target present a major impediment to sustained treatment efficacy. Our method, which is based on the previously reported YESS system, recapitulated all validated BCR-ABL1 mutations leading to clinical resistance to the second-generation inhibitor dasatinib, in addition to identifying numerous other mutations which have been previously observed in patients, but not yet validated as drivers of resistance. Further, we were able to demonstrate that the newer inhibitor ponatinib is effective against the majority of known single resistance mutations, but ineffective at inhibiting many compound mutants. These results are consistent with preliminary clinical and in vitro reports, indicating that mutations providing resistance to ponatinib are significantly less common; therefore, predicting ponatinib will be less susceptible to clinical resistance relative to dasatinib. Using the same yeast-based method, but with random substrate libraries, we were able to identify consensus peptide substrate preferences for the SRC and LYN kinases. ABL1 lacked an obvious consensus sequence, so a machine learning algorithm utilizing amino acid covariances was developed which accurately predicts ABL1 kinase peptide substrates.

Structural and functional dissection of the DH and PH domains of oncogenic Bcr-Abl tyrosine kinase.

The two isoforms of the Bcr-Abl tyrosine kinase, p210 and p190, are associated with different leukemias and have a dramatically different signaling network, despite similar kinase activity. To provide a molecular rationale for these observations, we study the Dbl-homology (DH) and Pleckstrin-homology (PH) domains of Bcr-Abl p210, which constitute the only structural differences to p190. Here we report high-resolution structures of the DH and PH domains and characterize conformations of the DH-PH unit in solution. Our structural and functional analyses show no evidence that the DH domain acts as a guanine nucleotide exchange factor, whereas the PH domain binds to various phosphatidylinositol-phosphates. PH-domain mutants alter subcellular localization and result in decreased interactions with p210-selective interaction partners. Hence, the PH domain, but not the DH domain, plays an important role in the formation of the differential p210 and p190 Bcr-Abl signaling networks.

The SH2 domain of Abl kinases regulates kinase autophosphorylation by controlling activation loop accessibility.

The activity of protein kinases is regulated by multiple molecular mechanisms, and their disruption is a common driver of oncogenesis. A central and almost universal control element of protein kinase activity is the activation loop that utilizes both conformation and phosphorylation status to determine substrate access. In this study, we use recombinant Abl tyrosine kinases and conformation-specific kinase inhibitors to quantitatively analyse structural changes that occur after Abl activation. Allosteric SH2-kinase domain interactions were previously shown to be essential for the leukemogenesis caused by the Bcr-Abl oncoprotein. We find that these allosteric interactions switch the Abl activation loop from a closed to a fully open conformation. This enables the trans-autophosphorylation of the activation loop and requires prior phosphorylation of the SH2-kinase linker. Disruption of the SH2-kinase interaction abolishes activation loop phosphorylation. Our analysis provides a molecular mechanism for the SH2 domain-dependent activation of Abl that may also regulate other tyrosine kinases.