Mapping the single-cell transcriptomic response of murine diabetic kidney disease to therapies.

Diabetic kidney disease (DKD) occurs in ∼40% of patients with diabetes and causes kidney failure, cardiovascular disease, and premature death. We analyzed the response of a murine DKD model to five treatment regimens using single-cell RNA sequencing (scRNA-seq). Our atlas of ∼1 million cells revealed a heterogeneous response of all kidney cell types both to DKD and its treatment. Both monotherapy and combination therapies targeted differing cell types and induced distinct and non-overlapping transcriptional changes. The early effects of sodium-glucose cotransporter-2 inhibitors (SGLT2i) on the S1 segment of the proximal tubule suggest that this drug class induces fasting mimicry and hypoxia responses. Diabetes downregulated the spliceosome regulator serine/arginine-rich splicing factor 7 (Srsf7) in proximal tubule that was specifically rescued by SGLT2i. In vitro proximal tubule knockdown of Srsf7 induced a pro-inflammatory phenotype, implicating alternative splicing as a driver of DKD and suggesting SGLT2i regulation of proximal tubule alternative splicing as a potential mechanism of action for this drug class.

A small molecule agonist of EphA2 receptor tyrosine kinase inhibits tumor cell migration in vitro and prostate cancer metastasis in vivo.

During tumor progression, EphA2 receptor can gain ligand-independent pro-oncogenic functions due to Akt activation and reduced ephrin-A ligand engagement. The effects can be reversed by ligand stimulation, which triggers the intrinsic tumor suppressive signaling pathways of EphA2 including inhibition of PI3/Akt and Ras/ERK pathways. These observations argue for development of small molecule agonists for EphA2 as potential tumor intervention agents. Through virtual screening and cell-based assays, we report here the identification and characterization of doxazosin as a novel small molecule agonist for EphA2 and EphA4, but not for other Eph receptors tested. NMR studies revealed extensive contacts of doxazosin with EphA2/A4, recapitulating both hydrophobic and electrostatic interactions recently found in the EphA2/ephrin-A1 complex. Clinically used as an α1-adrenoreceptor antagonist (Cardura®) for treating hypertension and benign prostate hyperplasia, doxazosin activated EphA2 independent of α1-adrenoreceptor. Similar to ephrin-A1, doxazosin inhibited Akt and ERK kinase activities in an EphA2-dependent manner. Treatment with doxazosin triggered EphA2 receptor internalization, and suppressed haptotactic and chemotactic migration of prostate cancer, breast cancer, and glioma cells. Moreover, in an orthotopic xenograft model, doxazosin reduced distal metastasis of human prostate cancer cells and prolonged survival in recipient mice. To our knowledge, doxazosin is the first small molecule agonist of a receptor tyrosine kinase that is capable of inhibiting malignant behaviors in vitro and in vivo.

Prediction of organ toxicity endpoints by QSAR modeling based on precise chemical-histopathology annotations.

The ability to accurately predict the toxicity of drug candidates from their chemical structure is critical for guiding experimental drug discovery toward safer medicines. Under the guidance of the MetaTox consortium (Thomson Reuters, CA, USA), which comprised toxicologists from the pharmaceutical industry and government agencies, we created a comprehensive ontology of toxic pathologies for 19 organs, classifying pathology terms by pathology type and functional organ substructure. By manual annotation of full-text research articles, the ontology was populated with chemical compounds causing specific histopathologies. Annotated compound-toxicity associations defined histologically from rat and mouse experiments were used to build quantitative structure-activity relationship models predicting subcategories of liver and kidney toxicity: liver necrosis, liver relative weight gain, liver lipid accumulation, nephron injury, kidney relative weight gain, and kidney necrosis. All models were validated using two independent test sets and demonstrated overall good performance: initial validation showed 0.80-0.96 sensitivity (correctly predicted toxic compounds) and 0.85-1.00 specificity (correctly predicted non-toxic compounds). Later validation against a test set of compounds newly added to the database in the 2 years following initial model generation showed 75-87% sensitivity and 60-78% specificity. General hepatotoxicity and nephrotoxicity models were less accurate, as expected for more complex endpoints.

Ligand recognition by A-class Eph receptors: crystal structures of the EphA2 ligand-binding domain and the EphA2/ephrin-A1 complex.

Ephrin (Eph) receptor tyrosine kinases fall into two subclasses (A and B) according to preferences for their ephrin ligands. All published structural studies of Eph receptor/ephrin complexes involve B-class receptors. Here, we present the crystal structures of an A-class complex between EphA2 and ephrin-A1 and of unbound EphA2. Although these structures are similar overall to their B-class counterparts, they reveal important differences that define subclass specificity. The structures suggest that the A-class Eph receptor/ephrin interactions involve smaller rearrangements in the interacting partners, better described by a 'lock-and-key'-type binding mechanism, in contrast to the 'induced fit' mechanism defining the B-class molecules. This model is supported by structure-based mutagenesis and by differential requirements for ligand oligomerization by the two subclasses in cell-based Eph receptor activation assays. Finally, the structure of the unligated receptor reveals a homodimer assembly that might represent EphA2-specific homotypic cell adhesion interactions.

Peroxynitrite inactivation of human cytochrome P450s 2B6 and 2E1: heme modification and site-specific nitrotyrosine formation.

This study examined the reaction of peroxynitrite (PN) with two human cytochrome P450s, P450 2B6 (2B6) and P450 2E1 (2E1). After the reaction with PN, the NADPH/reductase-supported 7-ethoxy-4-(trifluoromethyl)coumarin (EFC) deethylation activity of both P450s was decreased in a concentration-dependent manner. HPLC analysis revealed that the prosthetic heme group of 2B6 was modified but to a lesser extent than the decrease in enzymatic activity. In contrast, the heme moiety of 2E1 was not altered. These results suggest that protein modification by PN contributed to the loss in enzymatic activity of 2B6 and 2E1 but to different extents. After trypsin digestion of the control and PN-inactivated P450s, tyrosine nitration was used as a biomarker for protein modification and the addition of the nitro group was determined using electrospray ionization-liquid chromatography-tandem mass spectrometry, allowing site-specific assignment of the tyrosine residues nitrated. Tyrosine residues 354, 244, 268, and 380 in 2B6 and tyrosine residues 317, 422, 69, and 380 in 2E1 were found to be nitrated. Tyrosine 354 is the primary site of nitration in 2B6, and tyrosine residues 422 and 317 are the primary targets for nitration in 2E1. After PN exposure, the EFC catalytic activity of 2E1 supported by tert-butylhydroperoxide was not affected, and the activity of 2B6 supported by tert-butylhydroperoxide was decreased to a lesser extent than that supported by NADPH/reductase. Following exposure to PN, the levels of the reduced-CO complex were less than the content of native heme remaining. These results suggest that PN-mediated protein modification has no effect on substrate binding but may impair the interaction of the reductase with P450s, thereby inhibiting electron transfer. Homology modeling shows that Tyr422 of 2E1 is in close proximity to the FMN domain of reductase, suggesting that Tyr422 may be involved in transferring electrons from the reductase to the heme and thus may play a critical structural and functional role in the extensive activity loss following PN exposure.

Photo processes on self-associated cationic porphyrins and plastocyanin complexes 1. Ligation of plastocyanin tyrosine 83 onto metalloporphyrins and electron-transfer fluorescence quenching.

The spectroscopic properties of the self-associated complexes formed between the anionic surface docking site of spinach plastocyanin and the cationic metalloporphyrins, in which the tyrosine 83 (Y83) moiety is placed just below the docking site, tetrakis(N-methyl-4-pyridyl)porphyrin (Pd(II)TMPyP(4+) and Zn(II)TMPyP(4+)), have been studied and reported herein. The fluorescence quenching phenomenon of the self-assembled complex of Zn(II)TMPyP(4+)/plastocyanin has also been discovered. The observed red-shifting of the Soret and Q-bands of the UV-visible spectra, ca. 9 nm for Pd(II)TMPyP(4+)/plastocyanin and ca. 6 nm for the Zn(II)TMPyP(4+)/plastocyanin complexes, was explained in terms of exciton theory coupled with the Gouterman model. Thus, the hydroxyphenyl terminus of the Y83 residue of the self-associated plastocyanin/cationic porphyrin complexes was implicated in the charge-transfer ligation with the central metal atoms of these metalloporphyrins. Moreover, ground-state spectrometric-binding studies between Pd(II)TMPyP(4+) and the Y83 mutant plastocyanin (Y83F-PC) system proved that Y83 moiety of plastocyanin played a critical role in the formation of such ion-pair complexes. Difference absorption spectra and the Job's plots showed that the electrostatic attractions between the cationic porphyrins and the anionic patch of plastocyanin, bearing the nearby Y83 residue, led to the predominant formation of a self-associated 1:1 complex in the ground-state with significantly high binding constants (K = (8.0 +/- 1.1) x 10(5) M(-1) and (2.7 +/- 0.8) x 10(6) M(-1) for Pd(II)TMPyP(4+) and zinc variant, respectively) in low ionic strength buffer, 1 mM KCl and 1 mM phosphate buffer (pH 7.4). Molecular modeling calculations supported the formation of a 1:1 self-associated complex between the porphyrin and plastocyanin with an average distance of ca. 9 A between the centers of mass of the porphyrin and Y83 positioned just behind the anionic surface docking site on the protein surface. The photoexcited singlet state of Zn(II)TMPyP(4+) was quenched by the Y83 residue of the self-associated plastocyanin in a static mechanism as evidenced by steady-state and time-resolved fluorescence experiments. Even when all the porphyrin was complexed (more than 97%), significant residual fluorescence from the complex was observed such that the amplitude of quenching of the singlet state of uncomplexed species was enormously obscured.

Role of cytochrome b5 in catalysis by cytochrome P450 2B4.

Cytochrome b5 has been shown to stimulate, inhibit or have no effect on catalysis by P450 cytochromes. Its action is known to depend on the isozyme of cytochrome P450, the substrate, and experimental conditions. Cytochrome P450 2B4 (CYP 2B4) has been used in our laboratory as a model isozyme to study the role of cytochrome b5 in cytochrome P450 catalysis using two substrates, methoxyflurane and benzphetamine. One substrate is the volatile anesthetic, methoxyflurane, whose metabolism is consistently markedly stimulated by cytochrome b5. The other is benzphetamine, whose metabolism is minimally modified by cytochrome b5. Determination of the stoichiometry of the metabolism of both substrates showed that the amount of product formed is the net result of the simultaneous stimulatory and inhibitory actions of cytochrome b5 on catalysis. Site-directed mutagenesis studies revealed that both cytochrome b5 and cytochrome P450 reductase interact with cytochrome P450 on its proximal surface on overlapping but non-identical binding sites. Comparison of the rate of reduction of oxyferrous CYP 2B4 and the rate of substrate oxidation by cyt b5 and reductase with stopped-flow spectrophotometric and rapid chemical quench experiments has demonstrated that although cytochrome b5 and reductase reduce oxyferrous CYP 2B4 at the same rate, substrate oxidation proceeds more slowly in the presence of the reductase.

Adapter protein for site-specific conjugation of payloads for targeted drug delivery.

High-affinity interactions of two fragments of human RNase I (1-15-aa Hu-tag and 21-125-aa HuS adapter protein) can be used for assembly of targeting drug delivery complexes. In this approach, a targeting protein is expressed as a fusion protein with a 15-aa Hu-tag, while HuS is conjugated to a drug (or a drug carrier) creating a "payload" module, which is then bound noncovalently to the Hu-tag of the targeting protein. Although this approach eliminates chemical modifications of targeting proteins, the payload modules are still constructed by random cross-linking of drugs or drug carriers to an adapter protein that might lead to functional heterogeneity of the complexes. To avoid this problem, we engineered an adapter protein HuS(N88C) with an unpaired cysteine in position 88 that can be directly modified without interference with activity of assembled targeting complexes. HuS(N88C) binds Hu-tagged annexin V with K(D) of 50 +/- 6 nM, which is comparable to that of wild-type HuS. To demonstrate the utility of HuS(N88C) for developing uniform payload modules, we constructed a HuS(N88C)-lipid conjugate and inserted it into preformed liposomes loaded with a fluorescent dye. Targeting proteins, Hu-tagged vascular endothelial growth factor or Hu-tagged annexin V, were docked to liposomes decorated with HuS, and the assembled complexes delivered liposomes selectively to target cells.

Chimeric ribonuclease as a source of human adapter protein for targeted drug delivery.

Assembled modular complexes for targeted drug delivery can be based on strong non-covalent interactions between a cargo module containing an adapter protein and a docking tag fused to a targeting protein. We have recently constructed a completely humanized adapter/docking tag system based on interactions between 15 amino acid (Hu-tag) and 110 amino acid (HuS) fragments of human ribonuclease I (RNase I). Although recombinant HuS can be expressed and refolded into a functionally active form, the purification procedure is cumbersome and expensive, and more importantly, it yields a significant proportion of improperly folded proteins. Here we describe engineering, high-yield expression, and purification of a chimeric bovine/human RNase (BH-RNase) comprising 1-29 N-terminal amino acids of bovine ribonuclease A and 30-127 amino acids of human RNase I. Unlike RNase I, the chimeric BH-RNase can be cleaved by either subtilisin or proteinase K between A20 and S21, providing a functionally active HuS. The HuS obtained from chimeric BH-RNase differs from wild-type HuS by an N24T substitution; therefore, we have reverted this substitution by mutating N24 to T24 in BH-RNase. This BH-RNase mutant can also be cleaved by subtilisin or proteinase K yielding wild-type HuS. The affinity of HuS obtained from BH-RNase to Hu-tag is approximately five times higher than that for recombinant HuS, reflecting a higher percentage of properly folded proteins.

Chemometrical classification of ephrin ligands and Eph kinases using GRID/CPCA approach.

Eph receptor tyrosine kinases are divided on two subfamilies based on their affinity for ephrin ligands and play a crucial role in the intercellular processes such as angiogenesis, neurogenesis, and carcinogenesis. As such, Eph kinases represent potential targets for drug design, which requires the knowledge of structural features responsible for their specific interactions. To overcome the existing gap between available sequence and structure information we have built 3D models of eight ephrins and 13 Eph kinase ligand-binding domains using homology modeling techniques. The interaction energies for several molecular probes with binding sites of these models were calculated using GRID and subjected to chemometrical classification based on consensus principal component analysis (CPCA). Despite inherent limitations of the homology models, CPCA was able to successfully distinguish between ephrins and Eph kinases, between Eph kinase subfamilies, and between ephrin subfamilies. As a result we have identified several amino acids that may account for selectivity in ephrin-Eph kinase interactions. In general, although the difference in charge between ephrin and Eph kinase binding domains creates an attractive long-range electrostatic force, the hydrophobic and steric interactions are highly important for the short-range interactions between two proteins. The chemometrical analysis also provides the pharmacophore model, which could be used for virtual screening and de novo ligand design.

Mutagenesis of prochlorothrix plastocyanin reveals additional features in photosystem I interactions.

Three surface residues of plastocyanin from Prochlorothrix hollandica have been modified by site-directed mutagenesis. Changes have been made in methionine 33, located in the hydrophobic patch of the copper protein, and in arginine 86 and proline 53, both located in the eastern hydrophilic area. The reactivity toward photosystem I of single mutants M33N, P53A, P53E, R86Q, R86E, and the double mutant M33N/P14L has been studied by laser flash absorption spectroscopy. All the mutations yield increased reactivity of plastocyanin toward photosystem I as compared with wild type plastocyanin, thus indicating that in Prochlorothrix electron donation to photosystem I is not optimized. The most drastic increases in the intracomplex electron transfer rate are obtained with mutants in methionine 33, whereas replacing arginine 86 only modestly affects the plastocyanin-photosystem I equilibrium constant for complex formation. Mutations at position 53 also promote major changes in the association of plastocyanin with photosystem I, yielding a change from a mechanism involving complex formation to a simpler collisional interaction. Molecular dynamics calculations indicate that mutations at position 33 promote changes in the H-bond network around the copper center. The comparative kinetic analysis of the reactivity of Prochlorothrix plastocyanin mutants toward photosystem I from other cyanobacteria reveals that mutations M33N, P53A, and P53E result in enhanced general reactivity.

Computational simulation of the docking of Prochlorothrix hollandica plastocyanin to potosystem I: modeling the electron transfer complex.

We have used several docking algorithms (GRAMM, FTDOCK, DOT, AUTODOCK) to examine protein-protein interactions between plastocyanin (Pc)/photosystem I (PSI) in the electron transfer reaction. Because of the large size and complexity of this system, it is faster and easier to use computer simulations than conduct x-ray crystallography or nuclear magnetic resonance experiments. The main criterion for complex selection was the distance between the copper ion of Pc and the P700 chlorophyll special pair. Additionally, the unique tyrosine residue (Tyr(12)) of the hydrophobic docking surface of Prochlorothrix hollandica Pc yields a specific interaction with the lumenal surface of PSI, thus providing the second constraint for the complex. The structure that corresponded best to our criteria was obtained by the GRAMM algorithm. In this structure, the solvent-exposed histidine that coordinates copper in Pc is at the van der Waals distance from the pair of stacked tryptophans that separate the chlorophylls from the solvent, yielding the shortest possible metal-to-metal distance. The unique tyrosine on the surface of the Prochlorothrix Pc hydrophobic patch also participates in a hydrogen bond with the conserved Asn(633) of the PSI PsaB polypeptide (numbering from the Synechococcus elongatus crystal structure). Free energy calculations for complex formation with wild-type Pc, as well as the hydrophobic patch Tyr(12)Gly and Pro(14)Leu Pc mutants, were carried out using a molecular mechanics Poisson-Boltzman, surface area approach (MM/PBSA). The results are in reasonable agreement with our experimental studies, suggesting that the obtained structure can serve as an adequate model for P. hollandica Pc-PSI complex that can be extended for the study of other cyanobacterial Pc/PSI reaction pairs.