No abstract

The chemical reactions underlying the emission of light in fireflies and other bioluminescent beetles are some of the most thoroughly studied processes by scientists worldwide. Despite these remarkable efforts, fierce academic arguments continue around even some of the most fundamental aspects of the reaction mechanism behind the beetle bioluminescence. In an attempt to reach a consensus, we made an exhaustive search of the available literature and compiled the key discoveries on the fluorescence and chemiluminescence spectrochemistry of the emitting molecule, the firefly oxyluciferin, and its chemical analogues reported over the past 50+ years. The factors that affect the light emission, including intermolecular interactions, solvent polarity, and electronic effects, were analyzed in the context of both the reaction mechanism and the different colors of light emitted by different luciferases. The collective data points toward a combined emission of multiple coexistent forms of oxyluciferin as the most probable explanation for the variation in color of the emitted light. We also highlight realistic research directions to eventually address some of the remaining questions related to firefly bioluminescence. It is our hope that this extensive compilation of data and detailed analysis will not only consolidate the existing body of knowledge on this important phenomenon but will also aid in reaching a wider consensus on some of the mechanistic details of firefly bioluminescence.

New crystal structures of red- and green blue–shifted beetle luciferases reveal that the color emission mechanism is dependent on the active site microenvironment affected by the conformation of loop regions.

Chemiluminescence, a process of transduction of energy stored within chemical bonds of ground-state reactants into light via high-energy excited intermediates, is known in solution, but has remained undetected in macroscopic crystalline solids. By detecting thermally induced chemiluminescence from centimeter-size crystals of an organic peroxide here we demonstrate direct transduction of heat into light by thermochemiluminescence of bulk crystals. Heating of crystals of lophine hydroperoxide to ~115 °C results in detectable emission of blue-green light with maximum at 530 nm with low chemiluminescent quantum yield [(2.1 ± 0.1) × 10 ‒7 E mol ‒1 ]. Spectral comparison of the thermochemiluminescence in the solid state and in solution revealed that the solid-state thermochemiluminescence of lophine peroxide is due to emission from deprotonated lophine. With selected 1,2-dioxetane, endoperoxide and aroyl peroxide we also establish that the thermochemiluminescence is common for crystalline peroxides, with the color of the emitted light varying from blue to green to red.

Camphorsultam-based lithium enolates referred to colloquially as Oppolzer enolates are examined spectroscopically, crystallographically, kinetically, and computationally to ascertain the mechanism of alkylation and the origin of the stereoselectivity. Solvent-and substrate-dependent structures include tetramers for alkyl-substituted enolates in toluene, unsymmetric dimers for arylsubstituted enolates in toluene, substrate-independent symmetric dimers in THF and THF/toluene mixtures, HMPA-bridged trisolvated dimers at low HMPA concentrations, and disolvated monomers for the aryl-substituted enolates at elevated HMPA concentrations. Extensive analyses of the stereochemistry of aggregation are included. Rate studies for reaction with allyl bromide implicate an HMPA-solvated ion pair with a + Li(HMPA) 4 counterion. Dependencies on toluene and THF are attributed to exclusively secondary-shell (medium) effects. Aided by density functional theory (DFT) computations, a stereochemical model is presented in which neither chelates nor the lithium gegenion serves roles. The stereoselectivity stems from the chirality within the sultam ring and not the camphor skeletal core.

The preparation, structure, physical properties, and reactivities of sodium isopropyl(trimethylsilyl)amide (NaPTA) are described. The solubilities at room temperature range from n-heptane (0.55 M), n-hexane (0.60 M), toluene (0.65 M), MTBE (1.7 M), Et3N (3.2 M), and THF (>6.0 M). The half-life to destruction in neat THF is >1 year at 25 °C and 7 days at 70 °C, which compares favorably to 2.5 months and 1.5 days, respectively, for LDA in neat THF. This study focuses on NaPTA in THF. 29Si NMR spectroscopy shows exclusively a mixture of cis and trans stereoisomeric dimers in 0.10–12 M THF in hexane. Density functional theory (DFT) computations suggest that the pK b is intermediate between dimeric sodium diisopropylamide (NaDA) and dimeric sodium hexamethyldisilazide (NaHMDS). Metalations of arenes, epoxides, ketones, hydrazones, alkenes, and alkyl halides show higher reactivities than LDA (k NaPTA/LDA = 1–30). While the rates of arene metalation are high, the lower pK b of NaPTA limits the substrates. Metalation of pseudoephedrate-based carboxamides to form disodiated Myers enolates solves several challenging technical problems.

The emitter of the firefly bioluminescence, oxyluciferin, and its derivatives were incorporated in agarose matrix to obtain self-supporting, lightweight fluorescent acidochromic thin films. This study demonstrates an alternative approach to investigating environmental effects on bioluminescent molecules.

Deep generative models for molecular graphs offer a new avenue for property optimization in drug discovery. Optimizing differentiable models that generate molecular graphs is certainly faster, cheaper, and much more accessible than traditional methods of chemical synthesis. Recent advances in generative modeling have managed to address many of the challenges surrounding generation of chemically-valid molecular graphs from latent representations, however the question of generating high-quality molecules remains. Herein we introduce MoFlowGAN a tandem normalizing flow model that can also be trained on both adversarial and reward objectives. We train our model on QM9 to generate high-quality and drug-like compounds. Our experiments show that MoFlowGAN is competitive with current state-of-the-art generative models while requiring far fewer training resources.