Steven E. Rokita

A potent lead for covalent inhibition of regulatory proteins is protected by alkylation of its 5-hydroxy group and released by hNQO1-catalyzed reduction.

Mosquito-borne diseases remain a major global health burden, and novel approaches to vector control are urgently needed as resistance to conventional insecticides spreads. Most mosquitoes breed in freshwater but can tolerate low levels of halogen salts. Whether specific halides exhibit selective toxicity toward mosquito larvae has not been systematically investigated. We now report on a survey focusing on Anopheles gambiae, the primary vector for malaria in sub-Saharan Africa. Bromide is selectively and potently toxic to Anopheles larvae with an LC₅₀ below 10 µM for its sodium potassium salt. This sensitivity is more than 10-fold greater than that observed for the sodium salts of fluoride, chloride and iodide. Bromide toxicity is stage-specific with An. gambiae larvae displaying 300-fold greater sensitivity than adults while eggs and pupae remain unaffected by 5 mM sodium bromide. Anopheles stephensi larvae exhibit similar sensitivity, whereas Aedes aegypti and Culex quinquefasciatus larvae are approximately 10-fold more tolerant of sodium bromide. Mechanistically, bromide toxicity operates independently of pathways targeted by conventional insecticides. Co-administration of diflubenzuron or spinosad neither suppressed nor enhanced bromide's effects. Bromide exposure induced overexpression of GluCl and CCC-2 ion channels without affecting Rdl, Best, or KCC expression. Chloride also protected against bromide toxicity, suggesting these halides share a common receptor. Thoracic swelling in dying larvae indicates that bromide disrupts osmotic or hydration balance. These findings are unexpected given that bromide is generally considered non-toxic to aquatic and terrestrial organisms, revealing a previously unknown vulnerability specific to Anopheles mosquitoes. The selective toxicity of bromide toward Anopheles larvae, combined with its independence from existing insecticide resistance mechanisms, may serve as a new lead for controlling malaria vectors. Targeting this bromide-sensitive pathway may complement current larvicidal approaches and provide an alternative for populations resistant to standard insecticides. The use of bromide itself will still require significant environmental and field-based studies prior to its possible application.

Flavin mononucleotide (FMN)-dependent nitroreductases offer a mild and selective route to reduce nitroaromatic compounds, yet these typically fail to generate the corresponding amines. This incomplete transformation has been attributed to the two-electron redox potential of FMN. To test this hypothesis, the major nitroreductase NfsA from Escherichia coli was reconstituted with a series of FMN variants spanning midpoint potentials of -215 to -307 mV. Product profiles were examined with four substrates covering a range of electron affinities (nitrofurazone, 1,3-dinitrobenzene, 4-nitroacetophenone, and nitrobenzene), and all combinations of enzymes and substrates were found to yield only the hydroxylamine products. No amines were detected under any condition, and as a confirmation, 4-hydroxylaminacetophenone was shown to be inert to treatment with reduced nicotinamide adenine dinucleotide phosphate and NfsA reconstituted with a low-potential FMN variant (-295 mV). The inability of NsfA to generate amines is consequently not a function of the reducing potential. However, this is a determinant of the catalytic efficiency. The k_cat for nitroacetophenone turnover decreased almost 240-fold for NfsA containing an FMN variant with an E_m of -307 mV relative to that containing native FMN (-215 mV). 1,3-Dinitrobenzene experienced the smallest decrease of 52-fold in the same comparison. Redox tuning of NfsA can therefore be detrimental to catalytic efficiency and fails to generate the desired amine products. A renewed focus on active site properties is recommended for engineering new catalysts to promote nitroaromatic to arylamine conversion.

The generality of delivering a bisquinone methide precursor to the major groove of DNA was tested with two naphthalimide conjugates differing in a single amide versus ester linkage. Both demonstrated an ability to cross‐link duplex DNA. Although a difference in their efficiency was not anticipated, the amide‐containing derivative consistently generated more cross‐linking than its ester‐containing analogue when tested with four different sequences of duplex DNA and one analogous RNA duplex. Both conjugates exhibited a specificity for a 5′‐…GTGTGT…‐3′ sequence in the absence of the preferred target of 5′‐GC and 5′‐CG. A corresponding conjugate containing acridine in place of naphthalimide did not express a sequence specificity and cross‐linked the test duplexes with an average efficiency similar to the median of that produced by the amide containing naphthalimide conjugate. With the plethora of naphthalimides available for cellular imaging and chemotherapy, quinone methide delivery may now be designed for many targets of choice. However, a variety of linkers should be surveyed to account for their surprising influence on cross‐linking.

Hydropersulfides (RSSH) are increasingly recognized for their potent redox-modulating and cytoprotective properties, yet their therapeutic potential remains underexplored due to their chemical instability. Here, we report the design and optimization of azoreductase (AzoR)-responsive RSSH donors for programmed release. We explore azo-linked precursors that undergo AzoR-mediated reduction to form phenylamino intermediates, which are designed to trigger RSSH release via spontaneous 1,6-elimination. A series of precursors was synthesized to evaluate the structure-activity relationships governing elimination efficiency. Direct attachment of aliphatic or aromatic RSSH moieties at the benzylic position of the azobenzene core (azodisulfides, AzDS-1, AzDS-2, AzDS-3) resulted in stable intermediates that failed to eliminate RSSH under physiological conditions. Even an electron-rich substituent such as a methoxy group on the azobenzene core was not sufficient to drive the elimination. To improve leaving group ability, we synthesized an azoperthiocarbonate donor (AzPTC) that enabled AzoR-triggered RSSH release but also underwent undesired nonenzymatic hydrolysis. Finally, incorporation of a hydrolytically stable perthiocarbamate yielded the precursor azoperthiocarbamate (AzPTB) that released RSSH selectively upon AzoR activation via 1,6-elimination, decarboxylation, and an intramolecular cyclization cascade. AzPTB demonstrates high enzymatic turnover and excellent stability under neutral aqueous conditions. These results provide key structural parameters that govern RSSH release via 1,6-elimination and establish AzPTB as a robust platform for site-selective delivery. This work expands the chemical biology toolkit for probing RSSH signaling and supports future efforts in redox-based therapeutic development.

Iodotyrosine deiodinases (IYDs) generally suffer from substrate inhibition and none more acutely than a homologue from Thermotoga neapolitana (TnIYD) that is an attractive target for engineering a robust catalysis to degrade halophenols that are persistent in the environment. The mechanism of this inhibition for TnIYD and the human homologue is now shown to derive from the formation of a nonproductive complex between the substrate 3-iodo-l-tyrosine and the enzyme in its oxidized form. This complex prevents subsequent reduction required for catalysis. Here, we describe a split gene approach to overcome this constraint. By coexpression, fragments of TnIYD assemble into an active catalyst with a discontinuity in the polypeptide backbone and deletion of residues from the active site lid. Three sets of enzymes were generated, and at least one representative of each overcomes substrate inhibition and demonstrates a gain in k_cat. Our results suggest that a significant fraction of the active site lid is dispensable and not necessary for promoting reductive dehalogenation. This approach offers a complement to circular permutation for manipulating the dynamics and accessibility of active sites.

Flavin-dependent azoreductases have been applied to a wide range of tasks from decolorizing numerous azo dyes to releasing azo-conjugated prodrugs. A general narrative reiterated in much of the literature suggests that this enzyme promotes sequential reduction of both the azo-containing substrate and its corresponding hydrazo product to release the aryl amine components while consuming two equivalents of NAD(P)H. Indeed, such aryl amines can be formed by incubation of certain azo compounds with azoreductases, but the nature of the substrates capable of this apparent azo bond lysis remained unknown. We have now prepared a set of azobenzene derivatives and characterized their turnover and products after treatment with azoreductase from Escherichia coli to discover the structural basis regulating aryl amine formation. Without resonance donation by aryl substituents, reduction ceases at the hydrazo product. This indicates that azoreductases do not act on the hydrazo bond. Instead, aryl amine formation depends on a spontaneous hydrazo bond lysis that is promoted by resonance stabilization and subsequent reduction of the quinone-like intermediate by azoreductase. Experimental and computational approaches confirm the substituent dependence of this process. With knowledge of this requirement, full release of aryl amines from azo-conjugates can now be designed and applied with confidence.

Flavin cofactors offer a wide range of chemical mechanisms to support a great diversity in catalytic function. As a corollary, such diversity necessitates careful control within each flavoprotein to limit its function to an appropriate subset of possible reactions and substrates. This task falls to the protein environment surrounding the flavin in most enzymes. For iodotyrosine deiodinase that catalyzes a reductive dehalogenation of halotyrosines, substrates can dictate the chemistry available to the flavin. Their ability to stabilize the necessary one-electron reduced semiquinone form of flavin strictly depends on a direct coordination between the flavin and alpha-ammonium and carboxylate groups of its substrates. While perturbations to the carboxylate group do not significantly affect binding to the resting oxidized form of the deiodinase, dehalogenation (k (cat)/K- m) is suppressed by over 2000-fold. Lack of the alpha-ammonium group abolishes detectable binding and dehalogenation. Substitution of the ammonium group with a hydroxyl group does not restore measurable binding but does support dehalogenation with an efficiency greater than those of the carboxylate derivatives. Consistent with these observations, the flavin semiquinone does not accumulate during redox titration in the presence of inert substrate analogues lacking either the alpha-ammonium or carboxylate groups. As a complement, a nitroreductase activity based on hydride transfer is revealed for the appropriate substrates with perturbations to their zwitterion.

Active site lids are common features of enzymes and typically undergo conformational changes upon substrate binding to promote catalysis. Iodotyrosine deiodinase is no exception and contains a lid segment in all of its homologues from human to bacteria. The solution-state dynamics of the lid have now been characterized using F-19 NMR spectroscopy with a CF3-labeled enzyme and CF3O-labeled ligands. From two-dimensional F-19-F-19 NMR exchange spectroscopy, interconversion rates between the free and bound states of a CF3O-substituted tyrosine (45 +/- 10 s(-1)) and the protein label (40 +/- 3 s(-1)) are very similar and suggest a correlation between ligand binding and conformational reorganization of the lid. Both occur at rates that are similar to 100-fold faster than turnover, and therefore these steps do not limit catalysis. A simple CF3O-labeled phenol also binds to the active site and induces a conformational change in the lid segment that was not previously detectable by crystallography. Exchange rates of the ligand (130 +/- 20 s(-1)) and protein (98 +/- 8 s(-1)) in this example are faster than those above but remain self-consistent to affirm a correlation between ordering of the lid and binding of the ligand. Both ligands also protect the protein from limited proteolysis, as expected from their ability to stabilize a compact lid structure. However, the minimal turnover of simple phenol substrates indicates that such stabilization may be necessary but is not sufficient for efficient catalysis.

Biological regulation often depends on reversible reactions such as phosphorylation, acylation, methylation, and glycosylation, but rarely halogenation. A notable exception is the iodination and deiodination of thyroid hormones. Here, we report detection of bromotyrosine and its subsequent debromination during Drosophila spermatogenesis. Bromotyrosine is not evident when Drosophila express a native flavin-dependent dehalogenase that is homologous to the enzyme responsible for iodide salvage from iodotyrosine in mammals. Deletion or suppression of the dehalogenase-encoding condet ( cdt ) gene in Drosophila allows bromotyrosine to accumulate with no detectable chloro- or iodotyrosine. The presence of bromotyrosine in the cdt mutant males disrupts sperm individualization and results in decreased fertility. Transgenic expression of the cdt gene in late-staged germ cells rescues this defect and enhances tolerance of male flies to bromotyrosine. These results are consistent with reversible halogenation affecting Drosophila spermatogenesis in a process that had previously eluded metabolomic, proteomic, and genomic analyses.