Magnetic resonance studies of fredericamycin A: evidence for O2-dependent free-radical formation

Structures, biosynthesis and biological activities of benastatins, anthrabenzoxocinones and fredericamycins

The fast spread of antibiotic resistance results in the requirement for a constant introduction of new candidates. Pentangular polyphenols, a growing family of actinomycetes-derived aromatic type II polyketides, have attracted considerable attention due to their intriguing polycyclic systems and potent antimicrobial activity. Among them, benastatins, anthrabenzoxocinones (ABXs), and fredericamycins, display unique variations in their polycyclic frameworks, yet concurrently share structural commonalities within their substitutions. The present review summarizes advances in the isolation, spectroscopic characteristics, biosynthesis, and biological activities of pentangular polyphenols benastatins (1-16), ABXs (17-39), and fredericamycins (40-42) from actinomycetes. The information presented here thus prompts researchers to further explore and discover additional congeners within these three small classes of pentangular polyphenols.

Synthetic Approach to the ABCD Ring System of Anticancer Agent Fredericamycin A via Claisen Rearrangement and Ring-Closing Metathesis as Key Steps

A new synthetic strategy to the ABCD ring system of the anticancer agent fredericamycin A (NSC-305263) was realized by the Diels-Alder reaction and olefin metathesis as key steps. The tactics developed here for the construction of the ABCD ring system also involve double Claisen rearrangement followed by a retro-Diels-Alder reaction and ring-closing metathesis. The metathesis approach performs a key role in the construction of A and D rings of the ABCD core unit. More importantly, ABCD fragment synthesis was accomplished without the involvement of protecting groups.

Antibacterial Paramagnetic Quinones from Actinoallomurus

Four metabolites, designated paramagnetoquinone A, B, C, and D (1-4), were isolated from three strains belonging to the actinomycete genus Actinoallomurus. Compounds 1 and 2 showed potent antibacterial activity with MIC values lower than 0.015 μg/mL against Gram-positive pathogens, including antibiotic-resistant strains. Since compounds 1 and 2 were NMR-silent due to the presence of an oxygen radical, structure elucidation was achieved through a combination of derivatizations, oxidations, and analysis of ¹³C-labeled compounds. The paramagnetoquinones share the same carbon scaffold as tetracenomycin but carry two quinones and a five-membered lactone fused to the aromatic system. Compounds 2 and 1 are identical except for an unprecedented replacement of a methoxy in 2 by a methylamino group in 1. Related compounds devoid of methyl group(s) and of antibacterial activity were isolated from a different Actinoallomurus strain. The likely pmq biosynthetic gene cluster was identified from strain ID145113. While the cluster encodes many of the expected enzymes involved in the formation of aromatic polyketides, it also encodes a dedicated ketoacid dehydrogenase complex and an unusual acyl carrier protein transacylase, suggesting that an unusual starter unit might prime the polyketide synthase.

Identification of fredericamycin E from Streptomyces griseus: Insights into fredericamycin A biosynthesis highlighting carbaspirocycle formation

Fredericamycin (FDM) A ( 1), a pentadecaketide featuring two sets of peri-hydroxy tricyclic aromatic moieties connected through a unique asymmetric carbaspiro center, exhibits potent cytotoxicity and represents a novel anticancer drug lead. We have localized previously the fdm gene cluster to a 33 kb DNA segment of Streptomyces griseus ATCC49344, the involvement of which in the biosynthesis of 1 was confirmed by gene inactivation, complementation, and heterologous expression experiments. We now report the isolation and characterization of FDM E ( 5), a heretofore undetected intermediate for 1 biosynthesis from S. griseus, shedding new insight into the mechanism of carbaspirocycle formation. The structure of 5 was elucidated through the combination of spectroscopic methods and isotope-labeling experiments. The core spiro[4.5]decane scaffold of 5 is characterized by a unique cyclohexa-1,2,4-triketone moiety. Transformation of the spiro[4.5]decane 5 into the spiro[4.4]nonane 1 can be rationalized by a biosynthetic benzilic acid-like rearrangement. This unusual rearrangement can be mimicked in vitro by proceeding under aerobic conditions in the absence of enzyme. FDM E displays cytotoxic activity on par with 1 against a selected set of cancer cells, a finding that further supports the unique molecular topology, resulting from the unprecedented carbaspirocycle as exemplified by 1 and 5, as a novel pharmacophore for this family of anticancer agents.

Cloning, Sequencing, Analysis, and Heterologous Expression of the Fredericamycin Biosynthetic Gene Cluster from Streptomyces griseus

Fredericamycin (FDM) A, a pentadecaketide featuring two sets of peri-hydroxy tricyclic aromatic moieties connected through a unique chiral spiro carbon center, exhibits potent cytotoxicity and has been studied as a new type of anticancer drug lead because of its novel molecular architecture. The fdm gene cluster was localized to 33-kb DNA segment of Streptomyces griseus ATCC 49344, and its involvement in FDM A biosynthesis was proven by gene inactivation, complementation, and heterologous expression experiments. The fdm cluster consists of 28 open reading frames (ORFs), encoding a type II polyketide synthase (PKS) and tailoring enzymes as well as several regulatory and resistance proteins. The FDM PKS features a KSalpha subunit with heretofore unseen tandem cysteines at its active site, a KSbeta subunit that is distinct phylogenetically from KSbeta of hexa-, octa-, or decaketide PKSs, and a dedicated phosphopantetheinyl transferase. Further study of the FDM PKS could provide new insight into how a type II PKS controls chain length in aromatic polyketide biosynthesis. The availability of the fdm genes, in vivo characterization of the fdm cluster in S. griseus, and heterologous expression of the fdm cluster in Streptomyces albus set the stage to investigate FDM A biosynthesis and engineer the FDM biosynthetic machinery for the production of novel FDM A analogues.

Asymmetric Total Synthesis of Fredericamycin A: An Intramolecular Cycloaddition Pathway

The asymmetric total synthesis of the potent antitumor antibiotic fredericamycin A ((S)-1) was achieved by the intramolecular [4+2] cycloaddition of the silylene-protected styrene derivative (S)-7 followed by the aromatic Pummerer-type reaction of the sulfoxide (S)-5. Although we had already succeeded in the total synthesis of racemic 1 by the same approach, synthesis of its asymmetric version was more complicated than we had expected due to the difficulties involved in constructing the quaternary carbon center and the tendency of this center to undergo facile racemization. Racemization of this center during the installation of the acetylene moiety on the dione (R)-8 was the most serious aspect. Systematic studies of its DE-ring analogue (R)-25 revealed that racemization of the quaternary carbon center proceeded by a retro-aldol-aldol reaction of the initial adduct, (1R)-39 a-Li, and that the degree of racemization was dependent on the reaction temperature. The racemization process could be completely depressed by keeping the reaction temperature at -78 degrees C. The construction of the stereogenic quaternary carbon center was achieved by the lipase-catalyzed desymmetrization of the prochiral 1,3-diol 9 a bearing the DEF-ring moiety. These studies enabled us to attain the asymmetric total synthesis of (S)-1 while completely retaining the chiral integrity created by the enzymatic reactions.

Do ethoxy radicals reduce 3H-nitrendipine binding in rat cardiac membranes?

The incubation of xanthine-oxidase (XOD) with rat cardiac membranes induced the formation of free radicals; they were identified by electron spin resonance spectroscopy (ESR) studies using 0.1 M 5,5-dimethyl-1-pyrroline-1-oxide (DMPO) for spin-trapping. The following DMPO adducts were measured: DMPO-OH, which was produced during the first minute of incubation, DMPO-R radicals, which were present after 4 min of incubation and where the signal intensity remained constant for at least 20 min. The binding studies performed after 15 min incubation showed that XOD (1-50 mU) dose dependently reduced 3H-nitrendipine binding. This reduction was caused by a decrease in the density of binding sites while the affinity remained unchanged. These results suggest that ethoxy radical formation may be an important step in the regulation of L type calcium channels.

Singlet oxygen interaction with Ca(2+)-ATPase of cardiac sarcoplasmic reticulum

We investigated the role of singlet oxygen (generated from photoactivation of rose bengal) on the calcium transport and Ca(2+)-ATPase activity of cardiac sarcoplasmic reticulum (SR). Isolated cardiac SR exposed to rose bengal (10 nM) irradiated at 560 nm resulted in significant inhibition of Ca2+ uptake (from 2.27 +/- 0.05 to 0.62 +/- 0.05 mumol Ca2+/mg.min [mean +/- SEM], p less than 0.01) and Ca(2+)-ATPase activity (from 2.08 +/- 0.05 to 0.28 +/- 0.04 mumol Pi/min.mg [mean +/- SEM], p less than 0.01). The inhibition of calcium uptake and Ca(2+)-ATPase activity by rose bengal-derived activated oxygen (singlet oxygen) was dependent on the duration of exposure and intensity of light. Singlet oxygen scavengers ascorbic acid and histidine significantly protected SR Ca(2+)-ATPase against rose bengal-derived activated oxygen species, but superoxide dismutase and catalase did not attenuate the inhibition. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of SR exposed to photoactivated rose bengal for up to 14 minutes demonstrated complete loss of the Ca(2+)-ATPase monomer band, which was significantly protected by histidine. The addition of dithiothreitol (5 mM) had a slight protective effect, showing that new disulfide bond formation was not a major cause of aggregation. The results were also confirmed by high-performance liquid chromatography of the SR exposed to irradiated rose bengal. Irradiation of rose bengal also caused an 18% loss of total sulfhydryl groups of SR. On the other hand, superoxide radical (generated from xanthine oxidase action on xanthine) and hydroxyl radical (in the presence of Fe(3+)-EDTA or 0.5 mM H2O2 plus Fe(2+)-EDTA) as well as H2O2 (0.25-12 mM) were without any effect on the 97,000-d Ca(2+)-ATPase band of SR. Generation of radical species (superoxide and hydroxyl radical) from rose bengal was studied by electron paramagnetic resonance spectroscopy using the spin trap 5,5-dimethyl-1-pyrroline-N-oxide (DMPO). The results showed that irradiation of rose bengal formed a 1:2:2:1 quartet, characteristic of the DMPO-OH adduct, which was scavenged by ethanol but not by superoxide dismutase, catalase, or histidine. No radical species could be detected from irradiated rose bengal or irradiated DMPO under the assay conditions used. Peroxy adducts of DMPO might be produced but would be observed only at very low temperatures. Similarly, we could not detect any measurable.O2- anion from irradiation of rose bengal as indicated by either cytochrome c reduction at 550 nm or nitro blue tetrazolium reduction at 560 nm. These results show that SR is damaged most likely by singlet oxygen derived from rose bengal.(ABSTRACT TRUNCATED AT 400 WORDS)

Inhibition of topoisomerases by fredericamycin A

Fredericamycin is an antibiotic product of Streptomyces griseus that exhibits modest antitumor activity in vivo and in vitro. Because of its unique structure and the absence of a clearly defined mechanism of action, we examined the effects of this compound on L1210 cells in culture as well as on several enzymes that bind to DNA. Fredericamycin exhibits an IC50 of 4.4 microM toward L1210 cells, and its cytotoxicity is a function of the time of exposure as well as drug dose. No DNA breakage was observed in L1210 cells or isolated nuclei following exposure to highly lethal concentrations of fredericamycin. As a first step toward understanding its mechanism of action, we examined the effect of fredericamycin on several enzymes involved in DNA metabolism. The catalytic activity of both DNA topoisomerases I and II were totally inhibited by fredericamycin concentrations of 4.4 and 7.4 microM, respectively. Fredericamycin blocked etoposide-stimulated DNA cleavage by topoisomerase II both in vitro and in isolated nuclei. In addition, the drug inhibits DNA polymerase a in vitro, exhibiting an IC50 of 93 microM. These diverse actions of fredericamycin do not enable us to draw conclusions regarding its mechanism of antitumor effect but clearly identify it as a compound of pharmacologic interest.

A novel mechanism for the generation of superoxide anions in hematoporphyrin derivative-mediated cutaneous photosensitization. Activation of the xanthine oxidase pathway

Prior studies, both in vitro and in vivo, have suggested that cutaneous porphyrin photosensitization requires the generation of superoxide anion (.O2-) and various other reactive oxygen metabolites. No unifying concept has emerged, however, that unequivocally demonstrates the source of generation of these species. Since xanthine oxidase is known to generate .O2- in reperfused ischemic tissue and in certain inflammatory disorders, we attempted to assess its role in porphyrin photosensitization. C3H mice were rendered photosensitive by the intraperitoneal administration of dihematoporphyrin ether (DHE) (5 mg/kg) followed by irradiation with visible light. Murine ear swelling was used as a marker of the acute photosensitization response and involvement of oxygen radicals was evaluated using electron spin resonance (ESR) spectroscopy. The administration of allopurinol, a potent inhibitor of xanthine oxidase, afforded 90% protection against DHE-mediated acute photosensitivity in vivo. Furthermore, xanthine oxidase activity was twofold higher in the skin of photosensitized mice than in unirradiated animals. ESR spectra of 5,5-dimethyl-1-pyrroline N-oxide-trapped radicals from the skin of photosensitized mice verified the presence of .O2- and .OH, while neither of these species was detected in the skin of control mice or mice receiving allopurinol. The administration of a soybean trypsin inhibitor or verapamil before irradiation also partially blocked the photosensitivity response, suggesting that calcium-dependent proteases play a role in the activation of xanthine oxidase in this photodynamic process. These data provide in vivo evidence for the involvement of .O2- in DHE-mediated cutaneous photosensitization and suggest that these radicals are generated through the activation of the xanthine oxidase pathway. The administration of allopurinol and calcium channel blockers may thus offer new approaches for the treatment of cutaneous porphyrin photosensitization.

Measurement of endothelial cell free radical generation: evidence for a central mechanism of free radical injury in postischemic tissues

Oxygen free radicals have been demonstrated to be important mediators of postischemic reperfusion injury in a broad variety of tissues; however, the cellular source of free radical generation is still unknown. In this study, electron paramagnetic resonance measurements with the spin trap 5,5'-dimethyl-1-pyrroline-N-oxide (DMPO) demonstrate that bovine endothelial cells subjected to anoxia and reoxygenation become potent generators of superoxide and hydroxyl free radicals. A prominent DMPO-OH signal aN = aH = 14.9 G is observed on reoxygenation after 45 min of anoxic incubation. Quantitative measurements of this free radical generation and the time course of radical generation are performed. Both superoxide dismutase and catalase totally abolish this radical signal, suggesting that O2 is sequentially reduced from O2-. to H2O2 to OH.. Addition of ethanol resulted in trapping of the ethoxy radical, further confirming the generation of OH.. Endothelial radical generation was shown to cause cell death, as evidenced by trypan blue uptake. Radical generation was partially inhibited and partially scavenged by the xanthine oxidase inhibitor allopurinol. Marked inhibition of radical generation was observed with the potent xanthine oxidase inhibitor oxypurinol. These studies demonstrate that endothelial cells subjected to anoxia and reoxygenation, conditions observed in ischemic and reperfused tissues, generate a burst of superoxide-derived hydroxyl free radicals that in turn cause cell injury and cell death. Most of this free radical generation appears to be from the enzyme xanthine oxidase. Thus, endothelial cell free radical generation may be a central mechanism of cellular injury in postischemic tissues.

Improvement of postischemic myocardial function and metabolism induced by administration of deferoxamine at the time of reflow: the role of iron in the pathogenesis of reperfusion injury

Reperfusion of ischemic myocardium has been postulated to result in a specific oxygen radical-mediated component of tissue injury. In a previous study we demonstrated improved recovery of ventricular function and metabolism when the superoxide radical scavenger superoxide dismutase was administered at the time of postischemic reflow. Studies in vitro, have suggested that superoxide toxicity might be mediated via the generation of more reactive hydroxyl radicals in an iron-catalyzed reaction. The present study was designed to test the hypothesis that myocardial reperfusion injury might be reduced by administration of the iron chelator deferoxamine at the time of reflow, most likely by preventing hydroxyl radical formation. Sixteen isolated Langendorff rabbit hearts, perfused within the bore of a superconducting magnet, were subjected to 30 min of normothermic (37 degrees C) total global ischemia followed by 45 min of reperfusion. At reflow eight treated hearts received a 10 ml bolus containing 50 mumol of deferoxamine followed by an infusion of 11 mumol/min for the first 15 min of reflow. The hearts were then perfused with standard perfusate for an additional 30 min. Eight untreated control hearts received a similar bolus of perfusate followed by 45 min of standard reperfusion. Serial 5 min 31P nuclear magnetic resonance spectra were recorded. Myocardial phosphocreatine (PCr) content fell to 5% to 7% of control during ischemia in both groups of hearts. Deferoxamine-treated hearts recovered 99 +/- 10% of control PCr content, while untreated hearts recovered 60 +/- 16% (p less than .05). Intracellular pH fell to 5.9 during ischemia in both groups, before showing more rapid and complete recovery in treated hearts (p less than .01). Recovery of developed pressure reached 70 +/- 6% of control in treated hearts compared with 35 +/- 10% in untreated hearts (p less than .05). Iron content of the perfusate was 7 microM, and by electron paramagnetic resonance spectroscopy was in the form of Fe3+-EDTA complexes. In the effluent of treated hearts iron was in the form of Fe3+-deferoxamine chelates. In summary, administration of the iron chelator deferoxamine at the time of postischemic reflow results in greater recovery of myocardial function and energy metabolism, which supports the hypothesis that iron plays an important role in the pathogenesis of reperfusion injury.