Generation of Broad-Spectrum Antifungal Drug Candidates from the Natural Product Compound Aureobasidin A.

The natural product aureobasidin A (AbA) is a potent, well-tolerated antifungal agent with robust efficacy in animals. Although native AbA is active against a number of fungi, it has little activity against Aspergillus fumigatus, an important human pathogen, and attempts to improve the activity against this organism by structural modifications have to date involved chemistries too complex for continued development. This report describes novel chemistry for the modification of AbA. The key step involves functionalization of the phenylalanine residues in the compound by iridium-catalyzed borylation. This is followed by displacement of the pinacol boron moiety to form the corresponding bromide or iodide and substitution by Suzuki biaryl coupling. The approach allows for synthesis of a truly wide range of derivatives and has produced compounds with A. fumigatus minimal inhibitory concentrations (MIC) of <0.5 µg/mL. The approach is readily adaptable to large-scale synthesis and industrial production.

Cloning and molecular characterization of the gene encoding the Aureobasidin A biosynthesis complex in Aureobasidium pullulans BP-1938.

The gene (aba1) encoding the NRPS complex responsible for the synthesis of the cyclic peptide antibiotic Aureobasidin A (AbA) in Aureobasidium pullulans BP-1938, was cloned using a combination of PCR and library screening approaches. The aba1 gene was found to consist of a single, intronless open reading frame (ORF) of 34,980 bp, encoding an 11,659 amino acid protein with a calculated molecular mass of 1,286,254 Da. Putative promoter and translation start elements were identified upstream from the putative ATG in the aba1 gene, and a consensus poly(A) addition signal (AATAAA) was identified 191 bp downstream of the translation termination codon (TGA). As predicted by the structure AbA, the aba1 gene encodes an enzyme composed of nine biosynthetic modules, eight of which contain adenylation domains with recognizable amino acid specificity-conferring code elements, and four of which contain embedded methylation domains. The biosynthetic module located at position one in the aba1 gene lacks recognizable specificity-conferring code elements, consistent with it being responsible for incorporation of the 2-hydroxy-3-methylpentanoic acid residue at that position in AbA. An unusual feature of the aba1 gene sequence is a very high degree of shared identity among eight of the biosynthetic modules, at both the nucleotide and amino acid level. The majority of the modules share better than 70% nucleotide identity with another module in the complex, and modules with the same amino acid adenylation specificity share up to 95% identity. Insertion of a hygromycin B phosphotransferase (HPT) gene cassette in place of the module 4 sequence in aba1 resulted in a cessation of AbA production, thus validating that the isolated gene encodes the AbA biosynthesis complex.

A cluster of novel serotonin receptor 3-like genes on human chromosome 3.

The ligand-gated ion channel family includes receptors for serotonin (5-hydroxytryptamine, 5-HT), acetylcholine, GABA, and glutamate. Drugs targeting subtypes of these receptors have proven useful for the treatment of various neuropsychiatric and neurological disorders. To identify new ligand-gated ion channels as potential therapeutic targets, drafts of human genome sequence were interrogated. Portions of four novel genes homologous to 5-HT(3A) and 5-HT(3B) receptors were identified within human sequence databases. We named the genes 5-HT(3C1)-5-HT(3C4). Radiation hybrid (RH) mapping localized these genes to chromosome 3q27-28. All four genes shared similar intron-exon organizations and predicted protein secondary structure with 5-HT(3A) and 5-HT(3B). Orthologous genes were detected by Southern blotting in several species including dog, cow, and chicken, but not in rodents, suggesting that these novel genes are not present in rodents or are very poorly conserved. Two of the novel genes are predicted to be pseudogenes, but two other genes are transcribed and spliced to form appropriate open reading frames. The 5-HT(3C1) transcript is expressed almost exclusively in small intestine and colon, suggesting a possible role in the serotonin-responsiveness of the gut.

A unique phenotype of 5-HT2C, agonist-induced GTPgamma35S binding, transferable to 5-HT2A and 5-HT2B, upon swapping intracellular regions.

1 The human 5-HT(2C) receptor, when expressed heterologously in various mammalian cell lines (HEK293, SH-EP and NIH-3T3) at various receptor densities (6 to 45 pmol mg(-1) protein), mediates robust agonist-induced GTPgamma(35)S binding from coupling to G(i) subtypes of G proteins, in addition to G(q/11). Such a phenotype, however, was not seen with the human 5-HT(2A) and 5-HT(2B) receptors, indicating their common pathway with 5-HT(2C) limited to G(q/11), not including G(i). 2 Because intracellular regions are largely responsible for signalling pathways, we prepared the chimeras of the 5-HT(2A) and 5-HT(2B) receptors where the second and third intracellular loops, and the C-terminal region were replaced with the 5-HT(2C) counterparts. 3 The chimeras showed robust agonist-induced GTPgamma(35)S binding. Relative intrinsic efficacies of agonists from the GTPgamma(35)S binding were nearly identical to the reported values for their parent receptors as measured with Ca(2+) or [(3)H]-inositol phosphate accumulation. Also the chimeras displayed the same ligand-binding properties as the parent receptors. 4 We conclude that the phenotype of agonist-induced GTPgamma(35)S binding is unique to 5-HT(2C) among the 5-HT(2) receptor family, and is transferable to 5-HT(2A) and 5-HT(2B), upon swapping intracellular sequences, without altering their receptor pharmacology.

Discovery of a protein sequence in the N-terminal region of the human neuronal alpha7 nicotinic acetylcholine receptor involved in homomeric interactions.

The neuronal nicotinic acetylcholine receptor subunit, alpha7, can form homopentameric receptor/ion channel complexes. Potential contributions of its N-terminal region to homomeric interactions were investigated, in comparison with the corresponding region of an analogous heteromeric subunit, alpha3. Recombinant chimeras were prepared upon engineering the N-terminal alpha7 (M1-V224) or alpha3 (M1-S232) sequence into the background of another homomeric mouse 5-hydroxytryptamine3 (5-HT)(3) receptor. The alpha7/5-HT(3) chimera, when expressed heterologously in a human epithelial cell line, SH-EP1, robustly expressed alpha-bungarotoxin binding sites as homooligomers while the alpha3/5-HT(3) did not produce epibatidine (non-selective ligand) binding sites, and did not interfere the alpha7/5-HT3 phenotype, upon co-expression. Yeast two hybrid assays with the N-terminal regions showed positive responses between alpha7:alpha7, but not between alpha7:alpha3 and alpha3:alpha3. Similar assays with the alpha7 N-terminal region and its five smaller fragments (G23-N46, D47-N90, V91-N133, S134-M182and Q183-V224) revealed that the G23-N46 sequence is involved in homomeric interactions. Replacement of the corresponding region of the alpha3/5-HT(3) chimera with the alpha7 G23-N46 sequence conferred a dominant negative role on the chimera, by abolishing the alpha7/5-HT(3) phenotype. These results support the view that the G23-N46 portion of the alpha7 N-terminal region may contribute to receptor homooligomerizations.