A Biomimetic Polymer for the Extraction and Purification of Superior Analogues of Amphotericin B.

Amphotericin B has been an essential drug in the fight against leishmaniasis and fungal pathogens for decades, and has more recently gained attention for the very limited microbial resistance displayed against it. However, its toxicity has restricted its use to only the most severe cases of disease, and attempts to reduce these ill effects via formulation have had only minor success. Genetic engineering has allowed the development of superior amphotericin analogues, notably 16-descarboxyl-16-methyl amphotericin B (MeAmB), which shows a ten-fold reduction in toxicity in addition to a slight improvement in therapeutic activity. However, MeAmB is difficult to extract from its bacterial source and purify. Presented here is an alternative method of MeAmB purification. A biomimetic polymer with a high affinity for MeAmB was designed via computational modelling and synthesised. Prepared as a separation column, the polymer was able to retain the target MeAmB whilst allowing the removal of cell debris from the bacterial extract. Starting with a simple bacterial extract, the relatively simple process allowed the purification of an MeAmB salt complex at approximately 70% MeAmB, and likely higher purification from further extraction. The mean MeAmB recovery between the pre-purification extract sample and the final product was 81%. This is the first successful demonstration of extraction or purification of any amphotericin molecule with any polymeric material. The biomimetic polymer was additionally reusable and simple to fabricate, giving this technique significant advantages over traditional methods of extraction and purification of valuable compounds.

Engineered biosynthesis and characterisation of disaccharide-modified 8-deoxyamphoteronolides.

Several polyene macrolides are potent antifungal agents that have severe side effects. Increased glycosylation of these compounds can improve water solubility and reduce toxicity. Three extending glycosyltransferases are known to add hexoses to the mycosaminyl sugar residues of polyenes. The Actinoplanes caeruleus PegA enzyme catalyses attachment of a D-mannosyl residue in a ß-1,4 linkage to the mycosamine of the aromatic heptaene 67-121A to form 67-121C. NppY from Pseudonocardia autotrophica adds an N-acetyl-D-glucosamine to the mycosamine of 10-deoxynystatin. NypY from Pseudonocardia sp. P1 adds an extra hexose to a nystatin, but the identity of the sugar is unknown. Here, we express the nypY gene in Streptomyces nodosus amphL and show that NypY modifies 8-deoxyamphotericins more efficiently than C-8 hydroxylated forms. The modified heptaene was purified and shown to be mannosyl-8-deoxyamphotericin B. This had the same antifungal activity as amphotericin B but was slightly less haemolytic. Chemical modification of this new disaccharide polyene could give better antifungal antibiotics.

Role of polyol moiety of amphotericin B in ion channel formation and sterol selectivity in bilayer membrane.

Amphotericin B (AmB) is a polyene macrolide antibiotic widely used to treat mycotic infections. In this paper, we focus on the role of the polyol moiety of AmB in sterol selectivity using 7-oxo-AmB, 7α-OH-AmB, and 7ß-OH-AmB. The 7-OH analogs were prepared from 7-oxo-AmB. Their K(+) flux activity in liposomes showed that introduction of an additional ketone or hydroxy group on the polyol moiety reduces the original activity. Conformational analyses of these derivatives indicated that intramolecular hydrogen-bonding network possibly influenced the conformational rigidity of the macrolactone ring, and stabilized the active conformation in the membrane. Additionally, the flexible polyol leads to destabilization of the whole macrolactone ring conformation, resulting in a loss of sterol selectivity.

Engineered biosynthesis of disaccharide-modified polyene macrolides.

Recent work has uncovered genes for two glycosyltransferases that are thought to catalyze mannosylation of mycosaminyl sugars of polyene macrolides. These two genes are nypY from Pseudonocardia sp. strain P1 and pegA from Actinoplanes caeruleus. Here we analyze these genes by heterologous expression in various strains of Streptomyces nodosus, producer of amphotericins, and in Streptomyces albidoflavus, which produces candicidins. The NypY glycosyltransferase converted amphotericins A and B and 7-oxo-amphotericin B to disaccharide-modified forms in vivo. The enzyme did not act on amphotericin analogs lacking exocyclic carboxyl or mycosamine amino groups. Both NypY and PegA acted on candicidins. This work confirms the functions of these glycosyltransferases and provides insights into their acceptor substrate tolerance. Disaccharide-modified polyenes may have potential as less toxic antibiotics.

Versatility of enzymes catalyzing late steps in polyene 67-121C biosynthesis.

Actinoplanes caeruleus produces 67-121C, a heptaene macrolide modified with a D-mannosyl-D-mycosaminyl disaccharide. Draft genome sequencing revealed genes encoding mycosaminyltransferase, mycosamine synthase, a cytochrome P450 that modifies the macrolactone core, and the extending mannosyltransferase. Only the mycosamine synthase and P450 were active in the biosynthesis of amphotericins in Streptomyces nodosus, the amphotericin producer.

Streptomyces nodosus host strains optimized for polyene glycosylation engineering.

The AmphDI glycosyltransferase transfers a mycosaminyl sugar residue from GDP onto 8-deoxyamphoteronolide B, the aglycone of the antifungal amphotericin B. In this study the amphDI gene was inactivated in Streptomyces nodosus strains lacking the AmphN cytochrome P450. The new mutants produced 8-deoxy-16-methyl-16-descarboxyl amphoteronolides in high yield. These strains and aglycones should prove valuable for in vivo and in vitro glycosylation engineering.

A labile point in mutant amphotericin polyketide synthases.

Streptomyces nodosus produces the antifungal polyene amphotericin B. Numerous modifications of the amphotericin polyketide synthase have yielded new analogues. However, previous inactivation of the ketoreductase in module 10 resulted in biosynthesis of truncated polyketides. Here we show that modules downstream of this domain remain intact. Therefore, loss of ketoreductase-10 activity is sufficient to cause early chain termination. This modification creates a labile point in cycle 11 of the polyketide biosynthetic pathway. Non-extendable intermediates are released to accumulate as polyenyl-pyrones.

Isolation and characterisation of amphotericin B analogues and truncated polyketide intermediates produced by genetic engineering of Streptomyces nodosus.

Amphotericin B is a powerful but toxic drug used against fungal infections and leishmaniases. These diseases would be treated more effectively if non-toxic amphotericin derivatives could be produced on a large scale at low cost. Genetic manipulation of the amphotericin B producer, Streptomyces nodosus, has previously led to the detection and partial characterisation of 8-deoxyamphotericin B, 16-descarboxyl-16-methyl-amphotericin B, 15-deoxy-16-descarboxyl-16-methyl-15-oxo-amphotericin B, 7-oxo-amphotericin B and pentaene analogues. Here we report improved production and purification protocols that have allowed detailed chemical analyses of these compounds. The polyketide synthase product 8-deoxy-16-descarboxyl-16-methyl-amphoteronolide B was identified for the first time. In addition, the ketoreductase 10 domain of the polyketide synthase was specifically inactivated by targeted gene replacement. The resulting mutants produced truncated polyketide intermediates as linear polyenyl-pyrones.

Redesign of polyene macrolide glycosylation: engineered biosynthesis of 19-(O)-perosaminyl-amphoteronolide B.

Most polyene macrolide antibiotics are glycosylated with mycosamine (3,6-dideoxy-3-aminomannose). In the amphotericin B producer, Streptomyces nodosus, mycosamine biosynthesis begins with AmphDIII-catalyzed conversion of GDP-mannose to GDP-4-keto-6-deoxymannose. This is converted to GDP-3-keto-6-deoxymannose, which is transaminated to GDP-mycosamine by the AmphDII protein. The glycosyltransferase AmphDI transfers mycosamine to amphotericin aglycones (amphoteronolides). The aromatic heptaene perimycin is unusual among polyenes in that the sugar is perosamine (4,6-dideoxy-4-aminomannose), which is synthesized by direct transamination of GDP-4-keto-6-deoxymannose. Here, we use the Streptomyces aminophilus perDII perosamine synthase and perDI perosaminyltransferase genes to engineer biosynthesis of perosaminyl-amphoteronolide B in S. nodosus. Efficient production required a hybrid glycosyltransferase containing an N-terminal region of AmphDI and a C-terminal region of PerDI. This work will assist efforts to generate glycorandomized amphoteronolides for drug discovery.

Engineered synthesis of 7-oxo- and 15-deoxy-15-oxo-amphotericins: insights into structure-activity relationships in polyene antibiotics.

Site-directed mutagenesis and gene replacement were used to inactivate two ketoreductase (KR) domains within the amphotericin polyketide synthase in Streptomyces nodosus. The KR12 domain was inactivated in the DeltaamphNM strain, which produces 16-descarboxyl-16-methyl-amphotericins. The resulting mutant produced low levels of the expected 15-deoxy-15-oxo analogs that retained antifungal activity. These compounds can be useful for further chemical modification. Inactivation of the KR16 domain in the wild-type strain led to production of 7-oxo-amphotericin A and 7-oxo-amphotericin B in good yield. 7-oxo-amphotericin B was isolated, purified, and characterized as the N-acetyl methyl ester derivative. 7-oxo-amphotericin B had good antifungal activity and was less hemolytic than amphotericin B. These results indicate that modification at the C-7 position can improve the therapeutic index of amphotericin B.

Autocatalytic formation of a covalent link between tryptophan 41 and the heme in ascorbate peroxidase.

Electronic spectroscopy, HPLC analyses, and mass spectrometry (MALDI-TOF and MS/MS) have been used to show that a covalent link from the heme to the distal Trp41 can occur on exposure of ascorbate peroxidase (APX) to H2O2 under noncatalytic conditions. Parallel analyses with the W41A variant and with APX reconstituted with deuteroheme clearly indicate that the covalent link does not form in the absence of either Trp41 or the heme vinyl groups. The presence of substrate also precludes formation of the link. Formation of a protein radical at Trp41 is implicated, in a reaction mechanism that is analogous to that proposed [Ghiladi, R. A., et al. (2005) Biochemistry 44, 15093-15105] for formation of a covalent Trp-Tyr-Met link in the closely related catalase peroxidase (KatG) enzymes. Collectively, the data suggest that radical formation at the distal tryptophan position is not an exclusive feature of the KatG enzymes and may be used more widely across other members of the class I heme peroxidase family.

Biosynthesis of amphotericin derivatives lacking exocyclic carboxyl groups.

Amphotericin B is a medically important antifungal antibiotic that is also active against human immunodeficiency virus, Leishmania parasites, and prion diseases. The therapeutic use of amphotericin B is restricted by severe side effects that can be moderated by liposomal formulation or structural alteration. Chemical modification has shown that suppression of charge on the exocyclic carboxyl group of amphotericin B substantially reduces toxicity. We report targeted deletions of the amphN cytochrome P450 gene from the chromosome of the amphotericin-producing bacterium Streptomyces nodosus. The mutant strains produced amphotericin analogues in which methyl groups replace the exocyclic carboxyl groups. These compounds retained antifungal activity and had reduced hemolytic activity.

Triol protection with 6-benzoyl-3,4-dihydro-(2H)-pyran.

6-Benzoyl-3,4-dihydro-(2H)-pyran will protect 1,2,3-triols such as glycerol as their corresponding spiro-[5-phenyl-3,6,8-trioxabicyclo[3.2.1]octane-4,2[prime or minute]-tetrahydropyran]s and 1,2,4-triols (less efficiently) as the corresponding trioxabicyclo[3.2.2]nonanes; the hexol mannitol is converted into the corresponding bis-protected product.

Biosynthesis of deoxyamphotericins and deoxyamphoteronolides by engineered strains of Streptomyces nodosus.

Amphotericin B is an antifungal antibiotic produced by Streptomyces nodosus. During biosynthesis of amphotericin, the macrolactone core undergoes three modifications: oxidation of a methyl branch to a carboxyl group, mycosaminylation, and hydroxylation. Gene disruption was undertaken to block two of these modifications. Initial experiments targeted the amphDIII gene, which encodes a GDP-D-mannose 4,6-dehydratase involved in biosynthesis of mycosamine. Analysis of products by mass spectrometry and NMR indicated that the amphDIII mutant produced 8-deoxyamphoteronolides A and B. This suggests that glycosylation with mycosamine normally precedes C-8 hydroxylation and that formation of the exocyclic carboxyl group can occur prior to both these modifications. Inactivation of the amphL cytochrome P450 gene led to production of novel polyenes with masses appropriate for 8-deoxyamphotericins A and B. These compounds retained antifungal activity and may be useful new antibiotics.