The biosynthetic machineries of ribosomally synthesized and post-translationally modified peptides (RiPPs) are often substrate tolerant. A remarkable example is the class II lanthipeptide synthetase ProcM, which naturally functions as a generalist enzyme that has not evolved to use a specific substrate during its evolutionary history. Although ProcM has been studied extensively, the sequence features associated with productive modification remain underexplored. In this study, we use the ultrahigh-throughput mRNA display technique to map the sequence compatibility of ProcM across a focused library. This approach expands the landscape of ProcM reactivity beyond native substrates and individually characterized variants. Machine learning (ML) analysis revealed that the dataset contains sequence patterns associated with ProcM reactivity, with models achieving a balanced accuracy of 0.73. This moderate predictive performance contrasts with the near-perfect accuracy reported for specialized enzymes. This finding suggests that predicting substrate compatibility for generalist enzymes is more challenging due to uneven representation of reactive and nonreactive sequences and limitations of the available data. Our results provide a high-throughput view of ProcM reactivity and highlight differences with previous high-throughput studies on substrate selectivity of RiPP modification enzymes. Future studies will need to assess whether these differences are common when comparing generalist with specialist enzymes.

Mammalian cells control a wide range of essential processes via protein-protein interactions (PPIs), including signal transduction, gene expression, cell cycle regulation, and metabolism. Pathologic alterations in PPIs have been linked to many human diseases. Thus, the ability to rapidly and selectively induce PPIs in cultured cells is crucial for many fields of biological research. One proven approach is the use of synthetic bivalent chemical ligands to selectively recruit protein domains with engineered mutations (i.e., "bump-and-hole"). Developing new PPI chemical inducers with unique protein domains enables multiplexing and potentially expands the breadth of recruitable proteins. Herein, we describe the design and characterization of a BromoTag- and dTAG-based toolkit to induce protein homo- and heterodimerization. Using a NanoBiT split-luciferase assay, we characterized a library of BromoTag-based homo- (BromoTag/BromoTag) and heterobivalent (BromoTag/dTAG) compounds for their ability to induce neo-PPIs. Compounds were evaluated for activity across multiple cellular applications, including caspase-9 and c-Kit tyrosine kinase activation, with comparable activity to other dimerizer systems. This BromoTag-based protein dimerization toolkit provides a useful alternative to current controlled dimerization systems that will enable robust prototyping of chemically induced proximity.

Copper is involved in healthy physiology and aberrant pathology with prominent roles in Wilson's and Menkes diseases. Tracking its reactivity in living systems is crucial to understanding its biological function and could be helpful in identifying targeted therapies. We developed a turn-on chemiluminescent copper probe CpCL, which was designed by linking a picolinic acid moiety to the phenolic substituent of a spiroadamantane 1,2-dioxetane. Reaction of CpCL with copper results in formation of a copper complex intermediate that releases the phenolate upon hydrolysis, followed by initiation of a chemiluminescence reaction for light emission. The probe displayed excellent selectivity for Cu2+ compared to other competitive metal ions, with a sensitive limit of detection of 360 nM. The mechanism was studied using initial rate kinetics to provide an estimated pseudo-first-order rate constant of 5.18 ± 0.38 × 10-5 s-1 for the hydrolysis of the copper-bound probe. The probe was then validated in living mice using the copper chelator ATN-224 as a negative control, demonstrating that it can be used to monitor copper ions in vivo.

Directed evolution is a powerful and versatile strategy for engineering proteins with enhanced affinity and tailored molecular recognition properties. Here, we applied phage display-guided directed evolution to enhance the binding of the wild-type Cbx1 chromodomain (Cbx1-CD) to an H3K9me3-containing peptide. Screening a soft-randomized library targeting residues surrounding the aromatic cage yielded C29, which bound the H3K9me3 peptide with a KD of 57 nM and exhibited an approximately 14-fold lower dissociation rate constant than Cbx1-CD. Across an H3K9me3-derived alanine-scanning series and a peptide panel, C29 showed stronger binding signals than Cbx1-CD for multiple peptides containing trimethyllysine (Kme3), suggesting lower flanking-sequence dependence for C29 than for Cbx1-CD. An AlphaFold3 model of the C29-H3K9me3 complex suggested local changes in the hydrophobic and electrostatic environment surrounding the Kme3-binding pocket, providing a mechanistic hypothesis for C29's enhanced affinity and reduced flanking-sequence dependence. When two Kme3-containing peptides were spiked into a 100-fold molar excess of unmodified background peptides, C29-conjugated beads effectively enriched both Kme3 peptides relative to the unmodified background. These findings define C29 as a high-affinity Kme3-binding Cbx1 chromodomain variant and support its further evaluation as an affinity reagent for Kme3-containing peptide enrichment.

Active DNA demethylation is a fundamental epigenetic process in which the iterative oxidation of 5-methylcytosine (5mC) by TET dioxygenases generates diverse combinations of oxidized 5-methylcytosines within CpG dinucleotides before their excision by thymine DNA glycosylase (TDG). Although heterogeneous CpG oxidation states are expected to arise frequently during this process, how the modification state of the opposing cytosine influences TDG activity remains poorly understood. Here, we developed a chemically defined panel of CpG substrates containing cytosine, 5mC, 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) opposite a target 5fC and systematically examined TDG-mediated excision in both naked DNA and nucleosome core particles (NCPs). In naked DNA, opposing-strand cytosine modifications exerted only modest effects on TDG activity when a single excisable base was present within the CpG dyad, whereas dual-substrate CpGs exhibited unexpected differences in strand utilization. In contrast, nucleosome assembly profoundly altered TDG activity, revealing modification-dependent effects that were largely absent in naked DNA and dramatically redistributing strand utilization within dual-substrate CpGs. Together, these findings demonstrate that the oxidation-state composition of a CpG dyad and nucleosome architecture cooperatively regulate TDG activity, establishing heterogeneous CpG oxidation states as functionally distinct substrates for TDG and providing a biochemical framework for understanding how chromatin structure regulates active DNA demethylation.

Though immune checkpoint blockade (ICB) therapy has been a promising advance in cancer treatment, response rates to ICB are low owing to other mechanisms that cancers employ to evade the immune system. SULT2B1b is an enzyme known to produce cholesterol sulfate (CS), an immunosuppressive metabolite that inhibits T-cell migration. Cancers expressing high levels of SULT2B1b thus generate a tumor microenvironment that excludes T cells, which leads to increased tumor growth and renders ICB therapy ineffective. Here, we identify a novel class of azasteroid SULT2B1b inhibitors with greater potency and selectivity than previously identified compounds. X-ray co-crystal structures indicate that these inhibitors bind in a conformation different from that of native substrates, enabling inhibition without acting as a substrate. Additionally, these new compounds effectively reduce CS levels in tumors in mice and, as such, have potential as cancer therapeutics in combination with ICB therapy.

Regulation of therapeutic cell response is important for safe and effective therapy, particularly for immunotherapy. Ideally, the regulators should be based on human proteins using compounds that have already been approved for human use. Regulation of protein degradation by small molecules enables fast cellular response and a small genetic footprint of genetic constructs. Here, we present a clinically compatible strategy for reversible pharmacological control of chimeric antigen receptor (CAR) T cell function using human estrogen receptor (ER)-based degron domains and FDA-approved small molecules. By fusing tamoxifen-responsive ER ligand-binding domains to CARs, we generate ligand-inducible ON-switch CARs whose stability, signaling, and effector functions are precisely controlled by 4-hydroxytamoxifen. Furthermore, we demonstrate that ER-tagged CARs can be selectively degraded using the FDA-approved ER-targeting PROTAC ARV-471, establishing a complementary OFF-switch mechanism that suppresses CAR expression and effector function. Notably, these results expand the ON-OFF switch platform for post-translational control of therapeutic proteins, offering new opportunities to improve the safety and precision of cellular immunotherapies.

The reaction of thioethers in biological macromolecules with oxaziridines enables their functionalization with moieties having utility in both basic and applied science. Exploitation of that reaction has advantages over common bioconjugations due to the relatively low abundance and weak reactivity of thioethers. Herein, we report a novel and improved route to 1-ureal piperidine,3-aryl oxaziridines that enables the preparation of bioconjugation reagents that are either electronically activated or deactivated. We leveraged the differential reactivities of the oxaziridine reagents and engineered methionine residues in distinct environments within a therapeutically relevant antibody fragment (Fab) to effect site-selective bioconjugations. We capitalized on this novel means of kinetic resolution to effect dual functionalization of a HER2-targeted Fab with distinct fluorophores for imaging receptor internalization. This work broadens the scope of what is possible in bioconjugations of thioethers and demonstrates that site-selective bioconjugations of the same residue at more than one site in a protein can be realized via reactivity-based reagent control.

Guanine (G)-rich DNA sequences can form unusual secondary structures known as G-quadruplexes (or G4s). The stability of these structures, whose folding is mainly driven by the formation and stacking of G-quartets, is influenced by several factors, foremost among them the nature of the sequence itself, particularly the nucleobases flanking the G-tracts, and the loops connecting them. A considerable effort has been devoted to understanding how the size, position, and composition of each loop and flanking sequences affect the stability of the resulting G4 fold. However, most of these studies have been restricted to sequences forming a three G-quartet G4 core. To better understand how the length of the G-tracts (from three to five Gs) and loops (from one to seven nucleotides) act in concert, we studied a set of 261 different sequences covering different possible combinations. A systematic biophysical analysis suggests that sequences with longer G-tracts tend to form more stable G4s, but also a mixture of species; it further shows that loop composition and position also influence the topology and stability of the resulting G4s. Collectively, this work provides an unprecedented view on how G4-forming sequences fold upon themselves and of the structural features that govern the stability of the resulting G4s.

A subset of lasso peptides function as antibiotics against Gram-negative bacteria by inhibition of the bacterial RNA polymerase (RNAP), including microcin J25 (MccJ25) and citrocin. These peptides interact with specific transporters in the inner and outer membranes of bacteria to gain access to the cytoplasm. The reliance on transporters limits the breadth of spectrum of these lasso peptide antibiotics and opens up paths to drug resistance. Here, we detail a strategy to overcome the transport limitations of lasso peptides: covalent conjugation of the lasso peptide to a cell penetrating peptide (CPP). Citrocin-CPP and MccJ25-CPP conjugates were synthesized by attaching the CPP to the loop of the lasso peptide and tested for antimicrobial activity against the ESKAPE pathogens. Citrocin-CPP and MccJ25-CPP have antimicrobial activity against all four Gram-negative ESKAPE pathogens, a broader spectrum of activity than the parent lasso peptides. Citrocin-CPP and MccJ25-CPP also exhibit more potent activity than the parent lasso peptides, with minimum inhibitory concentrations in the high nanomolar to single micromolar range. The uptake of citrocin-CPP and MccJ25-CPP was also studied in Escherichia coli revealing that the conjugates are internalized via both transporter-dependent and transporter-independent pathways.

Small GTPases are critical regulators of cellular processes such as cell migration. However, no generalizable method exists for directly controlling their activity with subcellular precision, limiting the ability to dissect long-range signaling effects. To address this, we pair fragmented small GTPases with improved light-inducible dimerization (iLID) domains, optimizing for low dark-state background and efficient light-induced reassembly, affording spLIT-small GTPases. Demonstrating modularity, we control filopodia formation (spLIT-Cdc42), cell movement (spLIT-Rac1), and membrane retraction (spLIT-RhoA) with subcellular precision. Applying this platform to long-range signaling, we provide the first demonstration that localized RhoA activation is sufficient to drive membrane retraction at the opposite cell pole, with a directional dependence set by actin cytoskeleton polarization, a result inaccessible to global perturbations used previously. These results suggest a mechanism for coordinated establishment of cell polarity and highlight spLIT-small GTPases as a versatile platform for the direct, spatial control of small GTPase signaling in living cells.

Cellular immunotherapies have transformed oncologic treatment in recent years, yet the inability to monitor cell migration post-transfusion remains a significant barrier to their development. Here, we explore the feasibility of a non-invasive cell-tracking method utilizing Cu(I)-catalyzed click chemistry in which cells are first labeled with alkynes and, after 1-2 weeks, are selectively targeted via an azide-conjugated fluorescent probe. We developed Click1, a novel tris-Cu-chelating azide capable of a rapid, robust click reaction directly on live cell surfaces with minimal cytotoxicity. When paired with Click1, two distinct alkyne-labeling reagents, N-(4-pentynoyl)-mannosamine-tetraacylated (Ac4ManNAl) and propargyl choline, were capable of detecting human T cells for over two weeks. Our findings elucidate the metabolic turnover of biochemical handles in primary immune cells, providing a potential framework for long-term cell tracking in immunotherapies.

Fungal nonreducing polyketide synthases (NRPKSs) produce therapeutically and ecologically important aromatic polyketides. The product template (PT) domain of fungal NRPKSs catalyzes the regioselective aldol cyclization of the reactive poly-β-ketone chain during polyketide maturation and is essential for product fidelity of the NRPKSs. Herein, we discovered the fungal NRPKS MecA of which the polyketide cyclization steps do not involve a functional PT domain. MecA is encoded in the biosynthetic gene cluster (BGC) responsible for the formation of 5-hydroxy-7-methoxy-2,3-dimethylchromone (1) from the biocontrol fungus Trichoderma afroharzianum T22. Heterologous expression and in vitro enzymatic assays fully established a biosynthetic pathway of 1 that involves three enzymes: MecA, dehydratase MecB, and methyltransferase MecC. The PT domain of MecA does not contain the conserved catalytic histidine that is essential for aldol cyclization activity. Consistent with the predicted inactive PT domain, MecA was demonstrated to catalyze a C1-C6 Claisen condensation as the first cyclization step of the pentaketide precursor, instead of the canonical aldol cyclization. Reconstitution of dissected MecA domains further demonstrated that the activity of MecA was not affected when the PT domain was removed. A Claisen Cyclase (CLC) domain that typically catalyzes Claisen cyclization in NRPKSs is also absent in MecA, implying the observed C1-C6 cyclization step is catalyzed by the remaining domains. This work contributes to the diversity of NRPKS cyclization chemistry and enables further genome mining and engineering of NRPKSs that may lead to biosynthesis of new aromatic polyketides.

Backbone-cyclized conotoxin Vc1.1 is a potent GABAB receptor agonist and analgesic drug lead for treating neuropathic pain. Here, we describe two yeast-based strategies for the direct recombinant production of this cyclic, disulfide-rich peptide. In the first approach, using Pichia pastoris, a fusion protein comprising the conotoxin precursor and a full-length asparaginyl ligase connected by a poly-His linker was secreted into the culture medium. Cyclization of the secreted α-conotoxin was initiated by a two-step pH adjustment protocol for generating the correctly folded conotoxin. In the second approach, Saccharomyces cerevisiae was engineered to display the activated ligase on its cell surface while secreting Vc1.1 precursors extracellularly. Cyclized peptides were detected in the culture supernatant without further processing. These two strategies provide efficient routes for the direct recombinant production of disulfide-rich and backbone-cyclized peptides.

A subset of endogenous steroids can be formed in the brain and exhibit brain-specific activities linked to neuronal health and disease. Despite the biological relevance of these so-called neurosteroids, molecular profiling of their effects is incomplete and their signature in the nuclear hormone receptor family that mediates signals of classical steroid hormones is elusive. Here, we present a comprehensive characterization of neurosteroids for nuclear receptor modulation and reveal pronounced inverse RORγ agonism as a relevant molecular mechanism of several abundant neurosteroids, with an impact on circadian rhythms and transcriptomic profiles in cellular settings. Structural analysis of the DHEA-RORγ interaction demonstrated orthosteric binding and pronounced rearrangements in the RORγ ligand binding domain to a fully repressed state, rationalizing the strong inverse agonist efficacy.