Band 75, Heft 2-3

Studies in our laboratory focus on problems in chemistry (new reactions and synthesis), biology (novel modes of action), and medicine (new therapeutic leads and drug delivery systems). These interconnected and often synergistic activities are inspired by an interest in novel structures, frequently from nature, that possess unique modes of action and significant clinical potential. Described herein are some examples of recent work from our laboratory that have led to new transition metal-catalyzed reactions, a new and remarkably potent therapeutic lead, and new drug delivery systems that are in clinical trials.

Heparan sulfate (HS) is a glycosaminoglycan (GAG) widely distributed as a proteoglycan on the cell surface and in the extracellular matrix of animal tissues. Among other important physiological functions, its polysaccharide chains mediate cell proliferation by binding growth factors [fibroblast growth factor (FGF), vascular endothelial growth factor (VEGF)], which are released in active form through the action of the enzyme heparanase overexpressed by tumor cells. HS is constituted of alternating disaccharide sequences of variously sulfated uronic acid (d-glucuronic, GlcA, or l-iduronic, IdoA) and glucosamine ( N -acetylated, GlcNAc, or N -sulfated, GlcNSO 3 ). HS mimics can be obtained by chemical modification of heparin, a more highly sulfated GAG clinically used as an anticoagulant and antithrombotic drug. With the aim of obtaining antagonists of FGFs as potential inhibitors of tumor neo-vascularization (angiogenesis), arrays of short FGF-binding sequences have been obtained by generating sulfation gaps within the prevalent (IdoA2SO 3 –GlcNSO 3 6SO 3 ) n sequences of heparin, by controlled base-catalyzed removal of 2- O -sulfate groups of IdoA2SO 3 residues.The C(2)–C(3) bond of all nonsulfated uronic acid residues have then been split with periodate, to generate flexible joints along the polysaccharide chain. The novel heparin derivative (poly-PST.sU), chiefly made up of the repeating tetrasaccharide units –GlcNSO 3 6SO 3 – IdoA2SO 3 – GlcNSO 3 6SO 3 – sU– (where sU is a glycol-split and reduced uronic acid residue) binds FGF2 as strongly as heparin. However, it is a poor inducer of formation of FGF2 dimers and of complexes with FGF receptors, required for triggering mitogenic signals. NMR and molecular modeling studies indicate that formation of these higher-order complexes is prevented by the unfavorable conformation induced by glycol-split residues. In a parallel study, partially N -acetylated heparins have been obtained that efficiently inhibit heparanase upon glycol-splitting. It is noteworthy that glycol-splitting involves inactivation of the active site for antithrombin, with consequent loss of anticoagulant activity. In contrast, poly-PST.sU and some of its analogs show potent antiangiogenic activity in in vivo models in which heparin is either proangiogenic or inactive.

The structure-activity relationships of epothilones indicate that major modifications are only tolerated in the western ring segment. In particular, C2 methyl of the thiazole ring appears to be most flexible. Its broad modification started from epothilone F, which was obtained from natural epothilone B by hydroxylation via the N -oxide. Some of the prepared derivatives exhibit improved esterase stability in addition to high cytotoxic activity. For these and other favorable properties, amine (BMS-310705) was recently introduced in clinical trials. In an alternative approach, modified side chains were introduced by replacement of the C12,C15 ring segment via ring-opening olefin metathesis (ROM) of epothilone C in the presence of ethylene to 12,13- seco -epothilone C, introduction of a synthetic building block followed by ring-closing olefin metathesis (RCM), and epoxidation to the 16-alkyne analog of epothilone A. The structure of the tetrapeptide tubulysin D was confirmed by total hydrolysis to N -methyl d-pipecolic acid, l-isoleucine, tubuvaline (Tuv), tubuphenylalanine (Tup), formaldehyde, and 3-methylbutyric acid. Mild acidic hydrolysis to cyclo-tubulysin andoxidative degradation to l-valine allowed the assignment of the stereocenters of Tuv, hydrazinolysis, and comparison with synthetic reference samples to that of Tup. The absolute configuration of tubulysin D is: ( R )-Mep, (2,3 S )-Ile, (1' R ,3' R )-Tuv, and (2 S ,4 R )-Tup.

Tackling the problem of carbohydrate recognition is a subject of current interest. Our strategies in this regard include development of new methods for the synthesis of saccharides and sugar arrays for screening, design of d-peptides to target bacterial cell-surface sugars, preparation of mechanism-based inhibitors of the enzymes involved in the biosynthesis of saccharides and the synthesis of aminoglycoside mimetics to target RNA.

Many natural and synthetic anticancer agents with the ability to interact with DNA have been discovered, but most have little sequence-specificity and often exhibit severe toxicity to normal tissues. Thus, there has been considerable interest in molecular biology and human medicine to find small molecules that can alkylate the DNA in a sequence-specific manner and modify the function of nucleic acids irreversibly. Analogs of naturally occurring antitumor agents, such as distamycin A, which bind in the minor groove of DNA, represent a new class of anticancer compounds currently under investigation. Distamycin A has driven researchers' attention not only for its biological activity, but also for its nonintercalative binding to the minor groove of double-stranded B-DNA, where it forms a strong reversible complex preferentially at the nucleotide sequences consisting of 4-5 adjacent adenine-thymine (AT) base pairs. The pyrrole-amide skeleton of distamycin A has been also used as DNA sequence-selective vehicles for the delivery of alkylating functions to DNA targets, leading to a sharp increase of its cytotoxicity, in comparison to that, very weak, of distamycin itself. In the last few years, several hybrid compounds, in which derivatives of naturally occurring antitumor agents, such as anthramycin or the alkylating unit of the antibiotic CC-1065, have been tethered to distamycin frames. The DNA alkylating and cytotoxic activities against several tumor cell lines are reported and discussed in terms of their structural differences in relation to both the number of N -methyl pyrrole rings and the type of alkylating unit tethered to the oligopeptidic frame.

Combination therapy using reverse transcriptase (RT) and protease (PR) inhibitors is currently the best clinical approach in combatting acquired immunodeficiency syndrome (AIDS), caused by infection from the human immunodeficiency virus type 1 (HIV-1). However, the emergence of resistant strains calls urgently for research on inhibitors of further viral targets such as integrase (IN), the enzyme that catalyzes the integration of the proviral DNA into the host chromosomes. Recently, we started studies on new IN inhibitors as analogs of natural products, characterized by one or two 3,4-dihydroxycinnamoyl moieties, which were proven to be IN inhibitors in vitro. Then, we designed and synthesized a number of derivatives sharing 3,4 dihydroxycinnamoyl groups, obtaining potent IN inhibitors active at submicromolar concentrations. Unfortunately, these derivatives lacked antiretroviral activity, probably owing to their high cytotoxicity. So we designed a number of 3,4,5-trihydroxycinnamoyl derivatives as less-cytotoxic IN inhibitors, which were proven to be antiretrovirals in cell-based assays. Finally, we designed and synthesized a number of aryldiketohexenoic acids, strictly related to the aryldiketo acid series recently reported by Merck Company, which were shown to be potent antiretroviral agents endowed with anti-IN activities either in 3' processing or in strand transfer steps.

The total synthesis of dysinosin A, a novel member of the aeruginosin group of marine natural products is discussed. The stereocontrolled synthesis also confirms the proposed structure and absolute stereochemistry of the natural product.

The enantiomerically pure cis -1,2-dihydrocatechols 2 , which are generated by enzymatic dihydroxylation of the corresponding aromatic, engage in regio- and stereo-controlled Diels-Alder cycloaddition reactions to give a range of synthetically useful bicyclo[2.2.2]octenes. Certain examples of the latter type of compound have been used as starting materials in the synthesis of the sesquiterpenoids (−)-patchoulenone and (−)-hirsutene.

(−)-Clavukerin A has been synthesized via intramolecular Julia coupling and intramolecular sulfone ester cyclization starting from (+)-limonene oxide.

After a brief account of the syntheses previously described in literature, several approaches of the antiangiogenic sesquiterpene fumagillin are described. Particularly, a Claisen–Ireland ring-closing metathesis strategy allowed the stereoselective preparation of several advanced intermediates in the fumagillin synthesis.

11-Deoxytetrodotoxin and 8,11-dideoxytetrodotoxin were synthesized from the same intermediate in the enantiomerically pure form.

Acetogenins isolated from the Annonaceae family of tropical trees have drawn considerable attention owing to their broad spectrum of biological activities. They are structurally characterized by the presence of one to three tetrahydrofuran rings in the center of a long (partly hydroxylated) hydrocarbon chain that ends in a (functionalized) butenolide moiety. Here we describe some of our results toward the first asymmetric total synthesis of cis- gigantrionenin, a principal acetogenin. In this approach, an enzyme-catalyzed epoxide hydrolysis and an enzyme-triggered double cyclization are crucial and give stereoselective access to essential chiral building blocks.

A convenient and efficient application of microencapsulated/methylrhenium trioxide (MTO) and MTO/pyridine systems for the selective oxidation of cardanol derivatives is reported. Environmentally friendly and low-cost H 2 O 2 was used as the oxygen atom donor. The catalysts were stable systems for at least five recycling experiments. In the oxidation of C2 alkyl-substituted cardanol derivatives, high conversion and yields of the corresponding para -benzoquinones were obtained. By contrast, in the absence of the C2 substituent, an unprecedented oxidative degradation to di-γ-lactone derivative was observed.

A short and efficient synthesis of enantiomerically pure α-hydroxylated lactone lignans starting from commercially available diisopropyl malate is presented. Stereoselective alkylation with various benzyl bromides and saponification yielded the corresponding succinic acids. Acetalization afforded the dioxolanones, which were stereoselectively alkylated. Reduction (and deprotection, where required) yielded the lactone lignans in up to 30 % overall yield. The inhibition of the proliferation of HT29 colon cancer cells was investigated. One lignane, bis-2,4,6 trimethylbenzyllactone lignan, was active (IC 50 = 35 μM), whereas all other tested lignans were inactive within the investigated concentration range.

Cystine is regarded as a four-atom bridged bis(α-amino acid). The bridge between the two glycine moieties in cystine has been replaced with all-carbon C4-bridges between the α-carbon of the glycines. The products are dicarba analogs of cystine. Conformational constraints have been conferred on these molecules by insertion of cis- and trans- double bonds in the bridge, by alkylidene substituents, by insertion of a triple bond, by insertion of aromatic or heteroaromatic rings, or otherwise by ring formation. Particularly rigidified dicarba analogs of cystine have been prepared where the α-amino acid carbon is quaternary and part of tricyclic ring structures. Ru(II)-catalyzed ring-closing metathesis (RCM) reactions have been widely used in the preparation of cyclic amino acids. Conformational constraints in acyclic structures result from substitution by simple alkyl, alkenyl, or alkynyl groups at the α-amino acid carbon in the formation of α-quaternary amino acid derivatives.

The stereochemical study of flexible stereogenic carbon chains, such as those of many novel natural products, is a particularly challenging task. Recent applications of our group on the so-called " J -based approach", a methodology relying on a detailed analysis of homonuclear (H-H) and heteronuclear (C-H) 2,3 J couplings, include the study of the sphinxolide family of antitumor macrolides, a group of molecules characterized by a flexible macrocyclic framework bearing a number of oxygenated and methylated undetermined stereocenters, and of ascaulitoxin, a nitrogen-containing phytotoxin with herbicidal activity produced by a phytopathogenic fungus. An extension of the original procedure, relying on a Hartree-Fock (HF) ab initio calculation of conformational equilibrium and an estimate of the Boltzmann averaged 2,3 J HH and 2,3 J CH couplings, has been applied to the stereochemical study of sapinofuranone A, where the conformational equilibrium among existing rotamers had initially led to controversial results. 13 C NMR chemical shifts are additional useful parameters in the study of complex organic molecules. Along these lines, we have lately proposed the use of Hartree–Fock gauge including atomic orbitals (GIAO) calculated 13 C NMR chemical shift values as a supporting tool for the validation of the structure of new natural products and the determination of the relative stereochemistry of diastereomeric flexible compounds that are characterized by multiple conformer equilibria.

Mass spectrometry plays a central role in structural proteomics, particularly in highly intensive structural genomics projects. This review paper reports some examples taken from recent work from the authors' laboratory and is aimed at showing that modern proteomics strategies are instrumental in the integration of structural genomic projects in fields such as: (i) protein–protein interactions, (ii) protein-DNA interactions, (iii) protein–ligand interactions, and (iv) protein-folding intermediates.

Prion diseases or transmissible spongiform encephalopathies (TSEs) are fatal neurodegenerative pathologies characterized by the formation in the central nervous system of the amyloid protein PrP Sc , which derives from a cellular precursor called PrP c . Epidemiological and laboratory studies have shown that in species where the PrP c gene is polymorphic, the genotype composition is an important factor for the development of the disease. Identification of PrP Sc allotypes accumulated in the brain during the disease proved valuable to investigate whether these polymorphisms are critical for the pathological conversion. These analyses are complicated by the heterogeneity and the insolubility of the prion amyloid extracted from affected brains, which have been obviated by extensive digestion of extracted fractions and analysis of peptide fragment composition. We have developed an optimized protocol of liquid chromatography/mass spectrometry (LC/MS) that can reliably map PrP peptides in digested fractions with a low PrPSc/contaminants ratio. This approach has been successfully applied to the analysis of amyloidogenesis in experimentally infected PrP-heterozygous laboratory animals.

A research program was initiated in 1990 to carefully examine the toxin profiles in mussels from the northern Adriatic Sea. Since then, a number of polyether toxins have been isolated and characterized, some of which represent new additions to the diarrhetic shellfish poisoning (DSP) class of biotoxins and seem to be peculiar to the Adriatic Sea. During our investigation of toxic Adriatic mussels, we also isolated a new type of toxin, whose structure was elucidated by extensive use of 1D and 2D NMR techniques. Some of them could represent a further alarm for public health owing to their cytotoxic activity. The recent application of liquid chromatography/mass spectrometry (LC/MS) methods for detection of Adriatic marine biotoxins allowed us to hasten the analysis of toxic samples and to advance effective structural hypothesis even when full structure elucidation of new toxins by NMR spectroscopy is hampered by the limited amount of available material.

Absolute stereochemistry of amphidinolides G and H, potent cytotoxic 27- and 26-membered macrolides, respectively, isolated from a marine dinoflagellate Amphidinium sp., was determined by X-ray diffraction analysis, synthesis of a degradation product, and chemical interconversion. Six new macrolides, amphidinolides H2~H5, G2, G3, and W, have been isolated from a marine dinoflagellate Amphidinium sp. (strain Y-42), and the structures were elucidated by 2D NMR data and chemical means. The structure-activity relationship of amphidinolide H-type macrolides for cytotoxicity was examined. The biosynthetic origins of amphidinolides B, C, H, J, T1, and W were investigated on the basis of 13 C NMR data of 13 C-enriched samples obtained by feeding experiments with [1- 13 C], [2- 13 C], and [1,2- 13 C 2 ] sodium acetates in cultures of the dinoflagellates. Five novel long-chain polyhydroxyl compounds, colopsinols A~E, were obtained from the Amphidinium sp. (strain Y-5).

This review article presents our group's recent research findings with regard to bioactive natural products from marine sponges and tunicates, as well as from sponge derived fungi. The organisms discussed originate in the Indopacific region, which has an exceptionally rich marine biodiversity. Major topics that are covered in our review include the chemical ecology of sponges, focusing on defense against fishes, as well as the isolation and identification of new bioactive constituents from sponges and tunicates. Sponge derived fungi are introduced as an emerging source for new bioactive metabolites, reflecting the currently growing interest in natural products from marine microorganisms.

Most leguminous plants close their leaves in the evening, as if to sleep, and open them early in the morning according to the circadian rhythm controlled by a biological clock. Extensive studies on nyctinastic plants led to the isolation of a variety of leaf-closing and -opening substances. And we found that the biological clock regulates the balance of concentration between leaf-opening and -closing substances in the plant body during the day.

In the last few years, molecular modeling studies have been published that are devoted to a better understanding of the structural flexibility of hyaluronan (HA). Further conformational investigations, however, are needed on this polysaccharide, such as the application of statistical methods to perform enhanced one-step conformational analyses of its subunits. Moreover, the adjustment of assisted model building and energy refinement (AMBER) force field could provide the appropriate computational tool to study the interactions of HA and its derivatives with proteins. The present paper reports a combined Monte Carlo (MC) and molecular dynamics (MD) approach applied to the conformational study of HA, using an adjusted version of AMBER force field and the generalized Born solvent-accessible surface area (GB/SA) continuum solvation model. The MC approach turned out to be extremely effective to outline a conformational survey of the disaccharides constituting HA. Complete sets of conformations of the monomers were provided for the first time, some of which had never been predicted. MD technique, integrating the MC results, correctly reproduced the unusual stiffness of HA and predicted the existence of a minor skew-boat conformation of the β-d-glucuronic moiety. The computational approach, as a whole, improved the comprehension of the dynamic behavior of HA and offered a clear causal explanation of the relative dynamics of the glycosidic linkages.

The Arabidopsis thaliana genome-sequencing project has identified the presence of 13 oxidosqualene cyclase homologs in this plant. In addition to the already identified clones, namely, CAS1 cycloartenol synthase, LUP1 lupeol synthase, and YUP8H12R.43 multifunctional triterpene synthase, two new cDNAs of the putative oxidosqualene cyclase genes, F1019.4 and T30F21.16, were obtained by polymerase chain reaction (PCR) and functionally expressed in yeast. Liquid chromatography/mass spectrometry (LC/MS) analysis led to the identification of some of their reaction products. Interestingly, except for CAS1 for sterol biosynthesis of primary metabolism, so-far-obtained all triterpene synthases of this plant are multifunctional, producing more than one cyclization product. A feeding experiment of 13 C-labeled acetate with LUP1 lupeol synthase transformant demonstrated the stereospecific water addition to lupenyl cation intermediate, yielding 3β,20 dihydroxylupane, which accounts for the multiproduct nature of this synthase.

A long-overlooked metabolic pathway for isoprenoid biosynthesis, the mevalonate-independent methylerythritol phosphate (MEP) pathway, is present in many bacteria and in the chloroplasts of all phototrophic organisms. It represents an alternative to the well known mevalonate pathway, which is present in animals, fungi, plant cytoplasm, archaebacteria, and some eubacteria. This contribution summarizes key steps of its elucidation and the state-of-the-art knowledge of this biosynthetic pathway, which represents a novel target for antibacterial and antiparasitic drugs.

The biotransformation of triptonide by cell suspension cultures of Platycodon grandiflorum was investigated. After six days of incubation, five products were obtained. On the basis of chemical and spectral evidence, their structures were elucidated as epitriptolide-14- O -β-d-glucoside, 5α-hydroxytriptonide, triptolide, triptodiolide, and 2β-hydroxytriptonide, among which epitriptolide-14- O -β-d-glucoside and 5α-hydroxytriptonide are new compounds.

Isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) serve as the universal precursors for the biosynthesis of terpenes. Besides the well-known mevalonate pathway, a second biosynthetic pathway conducive to IPP and DMAPP via 1-deoxy-d-xylulose-5-phosphate and 2 C -methyl-d-erythritol-4-phosphate has been discovered recently in plants and certain eubacteria. 2 C -Methyl-d-erythritol-4-phosphate, the first committed intermediate of the deoxyxylulose phosphate pathway, is converted into 2 C -methyl-d-erythritol 2,4-cyclodiphosphate by the catalytic action of three enzymes specified by the ispDEF genes. The cyclic diphosphate is reductively opened by the IspG protein affording 1-hydroxy-2-methyl-2-( E )-butenyl-4-diphosphate. This compound can be converted into IPP as well as DMAPP by the catalytic action of IspH protein. The enzymes of this pathway are potential targets for novel antibacterial, antimalarial, and herbicide agents.

An overview is presented of the experimental results obtained by our group in the design, realization, and applications of high-performance liquid chromatography (HPLC) enantioselective stationary phases based on surface-linked macrocyclic receptors. Macrocyclic synthetic receptors and macrocyclic glycopeptide receptors have been grafted on silica particles, and the resulting chiral stationary phases are used to monitor differential associations of the enantiomers of selected chiral guests with the immobilized hosts. Practical implications of the chiral phases in enantioselective analysis and in the study of recognition events are highlighted.

A new synthetic method based on combination of enzymatic desymmetrization of 3-phenylglutaric anhydrides with multicomponent condensation is outlined. Enzymatic monoesterification of glutaric anhydrides was performed, and the derived monoacids were used as substrates for subsequent Ugi and Passerini multicomponent condensations. Both reactions were combined as two-step, one-pot processes. The choice of solvent and enzyme type has an appreciable effect on the course of reactions.