ACS Chemical Biology最新文献

Mammalian cells synthesize thousands of distinct lipids, and lipid metabolism fundamentally influences signaling in both homeostatic and disease states. Despite considerable interest in the biological functions of diverse lipid species, direct study of their effects in biological systems has historically been limited by a lack of methods to modulate discrete lipid species with precise spatiotemporal control. Several efforts in lipid chemical biology focus on the development of technologies to generate individual lipid species with spatial and temporal control in living cells. Emerging strategies are enabling the on-demand generation of lipids in response to various stimuli. In this Perspective, we discuss three mechanistic approaches for controlled lipid generation: light-triggered production of lipids from modified precursors, orthogonal chemical release, and triggerable chemoenzymatic or genetically encoded enzymatic generation of lipids. We highlight the strengths and weaknesses of each platform, and we outline the design strategies that will shape the next generation of tools.

Mycobacterial infections constitute a major public health burden, with antibiotic resistance contributing significantly to increasing mortality rates. “Anti-evolution” drugs that treat infections and limit new resistance represent promising next-generation antibiotics for these infections. The DNA replication machinery, or replisome, represents a prime anti-evolution drug target. Within the replisome, targeting the bacterial sliding clamp protein DnaN is particularly advantageous: it is essential for both normal and error-prone DNA replication and interacts with replication proteins through a short, conserved clamp binding motif (CBM) that is distinct from eukaryotic homologs. A promising starting point to target mycobacterial sliding clamp proteins is the cyclic depsipeptide natural product griselimycin (GM), which inhibits mycobacterial growth by interacting with the sliding clamp protein as a CBM mimic to inhibit DNA replication. Nonetheless, a detailed map of how GM’s chemical structure influences its function is lacking. Here, we prepared a systematic series of GM variants to probe functional group roles in GM’s interaction with bacterial sliding clamp proteins and in antibiotic activity against the model mycobacterial species Mycobacterium smegmatis. Overall, we determined that the N-terminal exocyclic region makes a limited contribution to both sliding clamp protein interaction and antibiotic activity. Modifications to the three native proline residues also contribute minimally to sliding clamp protein interaction but do improve antibiotic activity. Taken together, these studies reveal previously unknown functional group roles in GM and highlight the N-terminus and/or proline residues as favorable sites for GM optimization to improve antimycobacterial activity and/or reduce synthetic complexity.

Modulating aptamer–ligand interactions in response to pH is essential for developing nucleic acid sensors and molecular devices capable of functioning across diverse biological pH environments. Current approaches to engineer pH-responsive aptamers predominantly rely on appending pH-sensitive motifs to the termini of the intact aptamer sequence. However, such approaches often require substantial sequence engineering, limiting their applicability to aptamers with ligand-binding pockets distant from the termini. Here, we present a rational and modular approach to pH-dependent control of aptamer function by splitting the aptamer proximal to the ligand-binding site and fusing it to a pH-responsive motif. The engineered aptamer exhibited efficient ligand binding under acidic conditions, with minimal response at neutral or alkaline pH. The findings were validated using fluorescence spectroscopy, circular dichroism (CD), electrophoretic analysis, and isothermal titration calorimetry (ITC). Notably, the engineered aptamer retained its pH-responsive behavior in a cellular environment. The design principle is general in nature and can be extended to other aptamer systems. Collectively, this work provides a versatile framework that expands the toolkit for programmable engineering of pH-responsive aptamers utilizing split aptamers, with potential applications in diagnostics and even label-free imaging using light-up aptamers.

Antibody-directed enzyme prodrug therapy (ADEPT) utilizes a cancer-targeting antibody–enzyme conjugate to selectively activate a prodrug at the site of a tumor. We developed a modular plug-and-play approach for the generation of ADEPT antibody–enzyme conjugates that pairs a self-labeling SNAPtag-enzyme fusion with a benzylguanine-tagged antibody for rapid generation of a series of conjugates for testing. This approach enables the fusion of pharmacy-grade antibodies that are site-specifically modified with a recombinantly expressed enzyme using bioorthogonal self-labeling chemistry. The biorthogonality of the final assembly step of the ADEPT system even provides the ability for antibody–enzyme pairing in situ in the target cellular environment, improving ease of testing and potentially improving in vivo pharmokinetics. We have demonstrated conjugate formation and localization to antigen-positive cells, retained enzymatic activity upon conjugate cell surface binding, and ultimately selective killing of tumor cells. Additionally, this modular assembly allowed us to construct conditionally controlled ADEPT systems through site-specific incorporation of a light-cleavable protecting group, pairing a caged enzyme with a caged prodrug to create Boolean AND logic, potentially increasing the therapeutic window by providing conditional control over cytotoxic drug activation.

Toll-like receptor 2 (TLR2) recognizes a remarkably diverse range of microbial molecules and plays central roles in host defense, immune homeostasis, and cancer immunotherapy. Although bacterial lipoproteins and lipopeptides are canonical TLR2 agonists, recent studies have identified additional microbiota-derived ligands, including lipids, proteins, and polysaccharides. Advances in chemical biology, structural analysis, and activity-guided fractionation have revealed how subtle variations in ligand architecture influence TLR2 heterodimer usage, downstream signaling, and context-dependent immune outcomes. In this review, we summarize current knowledge of microbiota-derived TLR2 ligands and discuss recent progress in synthetic TLR2 agonists. We further highlight how ligand structure, receptor heterodimerization, and cellular context collectively shape TLR2-mediated immune responses.

Over the past two decades, engineering efforts have yielded aminoacyl-tRNA synthetases (aaRSs) capable of charging diverse noncanonical amino acids (ncAAs). Here, we report an unexpected and exploitable activity of these engineered aaRSs. In the absence of their cognate tRNA, aaRSs can release the ncAA-AMP intermediate, which labels proximal proteins via covalent acylation. Multiple different aaRSs exhibit this behavior, enabling both self-labeling and trans-labeling of proteins with ncAAs in vitro and in living cells. When using bioorthogonal ncAAs, the resulting labeled proteins can be further functionalized through bioorthogonal conjugation with a fluorophore for visualization or biotin for enrichment. Although labeling by existing aaRSs is slow, we provide preliminary evidence that improved activity is possible through further engineering. Taken together, these results establish a new potential approach for proximity-based protein labeling, analogous to BioID and TurboID. Many aaRSs are available to charge diverse ncAAs, creating opportunities for multiplexed labeling; i.e., different aaRS/ncAA pairs used simultaneously to probe multiple spatially/temporally distinct microenvironments.

Bispecific T cell engagers (BiTEs) are engineered immunotherapeutic molecules designed to direct the body’s immune system against malignant cells by physically connecting T cells to target cells, triggering cytotoxic responses. BiTEs have typically engaged T cells via αCD3 domains, such as OKT3 or UCHT1, which are linked to a second binding domain specific for a tumor-associated antigen. The risk of on-target off-tumor toxicity is a serious concern for BiTE immunotherapy due to the expression of tumor-associated antigen targets on healthy tissues. To address this concern, we have developed protease-responsive masked-BiTEs using OKT3 and UCHT1 mAb clones, that activate T cells only after exposure to tumor micro-environment-associated proteases and stimulate the T cell activation by their tumor-antigenic targets. Our protease-responsive masked BiTEs (mBiTEs) are comprised of αCD3 Fab proteins that are sterically prevented from binding CD3 via N’ terminal fusion of coiled-coil leucine zipper dimerization domains to each Fab chain, with further C’ terminal linkage to a tumoral-antigen targeting nanobody (either αPSMA or αHER2). The peptide linking each Fab chain and its coiled-coil masking zipper harbors a protease cleavable site, such that engagement of either recombinant or T cell displayed CD3 is significantly reduced until mBiTE protease activation. OKT3 and UCHT1-based masked T cell engagers (mTEs) showed ∼400-fold and ∼800-fold reduced affinity for CD3 protein compared to unmodified T cell engagers (TEs), with ∼222-fold and ∼457-fold restoration of binding upon proteolytic unmasking, respectively. mBiTEs did not activate Jurkat T cells while proteolytically unmasked proteins did, and mBiTEs harboring linkers cleavable by the cancer associated matric metalloprotease 2 (MMP2) enzyme enabled killing of PSMA + and HER2 + cancer cells in an MMP2-dependent manner in PBMC co-culture assays. The mBiTEs we describe here could be of broad utility towards helping constrain BiTE-mediated immune-activation to protease-high environments, such as the solid tumor microenvironment.

Gene order is a powerful design principle for protein nanomachines. In nature, gene organization ensures the precise assembly of functional protein nanostructures. We demonstrate how the genetic repositioning of the key structural gene pduN within the operon encoding a self-assembling protein nanocompartment sculpts the morphology and function of bacterial microcompartments (BMCs). Relocating pduN to new operonic positions dramatically altered the size, shape, and catalytic output of BMCs, despite identical protein sequences. These shifts reveal how gene order may control nanoscale assembly and compartmentalized function. Our findings establish operon architecture as a programmable genetic framework for nanostructure morphogenesis and provide a synthetic biology strategy to engineer self-assembling nanodevices with customized geometries and activities.

Pantocin A is a ribosomally synthesized, post-translationally modified peptide (RiPP) natural product that exhibits antibiotic activity through the inhibition of histidine biosynthesis. The gene cluster that gives rise to pantocin A contains genes for two putative biosynthetic enzymes, PaaA and PaaB. PaaA has previously been shown to modify the EEN core of the precursor peptide to generate a bicyclic structure similar to, but distinct from, the final structure of pantocin A; no characterization of PaaB has been reported to date. Herein, we determine the structure of the PaaA product, resolving the outstanding ambiguity in previous structural characterization. Moreover, we show that this species is accepted as a substrate by PaaB, an Fe(II)- and 2-oxoglutarate-dependent enzyme that catalyzes a unique 1,4-dehydrogenation to furnish the mature post-translational modification. These studies complete the biosynthetic pathway of pantocin A and the steps involved in its maturation.

We developed a genetically encoded system to perturb the methylation potential of living cells by exploiting the metabolic consumption of S-adenosylmethionine (SAM) through a plant methyltransferase, eugenol O-methyltransferase 1 (EOMT1). This system enables control of intracellular SAM levels with spatio- and temporal precision, regulated by the expression of EOMT1 and the timing of eugenol addition. EOMT1 expression allowed concentration-dependent reduction of SAM levels within 30 min. The system revealed distinct histone marks such as H3K4me3, H3K9me3, and H3K27me3, which responded differentially to SAM perturbation, revealing variable sensitivity between histone methyltransferases and demethylases. In addition, the spatially restricted expression of EOMT1 enabled the metabolic link of SAM between multiple organelles.