Marine-derived actinomycete bacteria are emerging as a valuable resource for bioactive natural products encompassing a variety of unique structural classes.[1] In our hands, early detection of cell growth inhibitors using in vitro cytotoxicity assays against the colon carcinoma cancer cell line HCT-116, followed by extensive mechanism of action studies, has proven to be an effective approach. As such, the HCT-116 assay has been instrumental in the identification of potentially important anticancer agents.[2]

In the course of our continued studies, Streptomyces strain CNR-698[3] was isolated from bottom sediments collected at a depth of 1618 meters in the Bahamas Islands in 2003. Cytotoxicity-guided (HCT-116) fractionation by C18 flash chromatography and RP-HPLC of crude extract led to the isolation of ammosamides A (1) and B (2) as blue and red solids, respectively (3 and 4 mg L−1). Structure assignments for 1 and 2 proved to be particularly difficult due to their inherent insolubility (soluble only in dimethyl sulfoxide (DMSO)) and a lack of descriptive NMR signals, ultimately requiring the integration of NMR spectral analysis, mass spectrometry data, and single crystal X-ray diffraction studies.

High-resolution (ESI) mass spectrometric analysis of ammosamide A (1) indicated a molecular formula C12H1035ClN5OS (m/z [M+H]+: 308.0303]. The molecular weight of ammosamide B (2) was found to be 16 amu lower (m/z [M+H]+: 292.0604) consistent with the molecular formula C12H1035ClN5O2. The UV/Vis spectrum of 1 was indicative of an unusually highly conjugated structure with absorptions at λmax=580, 430, 350, and 290 nm.

Inspection of the 1H NMR spectrum of 1 in [D6]DMSO revealed six singlets between δ=6.0 and 9.0 ppm and one methyl singlet at δ=4.03 ppm, while the 13C NMR spectra revealed the presence of eleven sp2 hybridized carbon atoms and a single sp3 hybridized carbon atom at δc=33.3 ppm (Table 1). The addition of D2O (20 μL) to the sample in [D6]DMSO resulted in the immediate disappearance of 1H NMR signals at δ=7.16 (1H), 6.63 (1H), 6.89 ppm (2H) and the slower disappearance of singlets at δ=8.92 (1H), and 7.68 ppm (1H) (less than 10 min). The exchangeable protons at δ=7.16, 6.63 and 6.89 ppm were assigned as aromatic amines at C-6 and C-8 (based on HMBC correlations), while the slowly exchanging protons at δ=8.92 and 7.68 ppm were assigned to a primary amide on the basis of COSY and HMBC correlations. The only non-exchangeable hydrogen atoms were the methyl singlet resonance at δ=4.03 ppm and a one-proton singlet at δ=8.47 ppm. The 13C NMR spectrum of 1 indicated the presence of two carbonyl groups (δc=177.2 and 166.0 ppm), as well as two upfield sp2 carbon atoms (δc=103.1 and 110.5 ppm). HMBC correlations between the downfield carbonyl (δc=177.2 ppm) and the proton methyl singlet at δ=4.03 ppm, we thought, defined an N-methyl amide, although a carbon chemical shift so far downfield would not be expected. In addition to correlations from the aromatic δ=8.47 ppm singlet, the only other HMBC correlations were from the exchangeable protons at δ=7.16/6.63 ppm to C-7 (δc=103.1 ppm) and from δ=6.89 ppm to C-7 and C-8a (δc=110.5 ppm).



Table 1

NMR spectral data for 1 and 2 ([D6]DMSO).





The spectral data for 1 suggested a highly unsaturated azaaromatic metabolite possessing three rings. However, the lack of definitive NMR assignments that could be used to link these features forced us to concentrate efforts toward obtaining an X-ray crystal structure. We were fortunate to obtain small crystals of 1 by the slow diffusion of H2O into a saturated solution in DMSO.[4] The X-ray assignment of ammosamide A (1) is shown in Figure 1. Once X-ray data became clear, the spectral data for 1 could be assigned.



Figure 1

X-ray crystal structure of ammosamide A (1). Red O, blue N, yellow Cl, black C, white H.



The structure assignment of ammosamide B (2) followed from analysis of spectral data and chemical interconversion. Comparison of the C-2 carbonyl chemical shifts in 1 (δC=177.2 ppm) and 2 (δC=164.0 ppm) revealed a difference of 13 ppm, consistent with the typical 13C chemical shift difference between a carbonyl and a thiocarbonyl (ca. 20 ppm).[5] In order to chemically confirm the presence of the thiolactam functionality, we used Lawesson's reagent [2,4-bis(p-methoxyphenyl)-1,3-dithiadiphosphetane-2,4-disulfide] to convert lactam 2 into thiolactam 1.[6] The low yield of this reaction is likely attributable the nucleophilic amines in 2.[7] Exposed to air during storage, 1 was gradually converted to ammosamide B (2). Notably, the transformation could also be accomplished in 10 min, upon treatment of 1 with hydrogen peroxide in aqueous methanol.[8] This reactivity has been previously observed in other thioamide-containing compounds.[9]

The structural similarities between the ammosamides and the microbial product lymphostin (3) are clear,[10] as is the relationship of the ammosamides to several sponge-derived pyrroloiminoquinone natural products, including batzelline A (4),[11a] isobatzelline D (5),[11b] and makaluvamine A (6)[11c] (Scheme 1). The sponge metabolites 4–6 possess different patterns of Cl and NH2 substitution and assume p-iminoquinone and o-quinone structures. The presence of an amino group at C-8 in the ammosamides results in a fundamentally different structure type in which the quinoline tautomer predominates. The pyrrole moiety in 3–6 is uniquely oxidized to the pyrrolidinone in ammosamide B (2). Finally, though methyl sulfides are present in 4 and 5, ammosamide A (1) is the first natural product to contain a thio-γ-lactam functionality.[12]



Scheme 1

Related metabolites from bacteria and sponges.



The fact that the ammosamides are highly colored (1: λmax=580 nm; 2: λmax=530 nm), yet lack quinone or iminoquinone functionalities, leads to speculation about the electronic character and reactivity of these metabolites. The intense colors of these compounds could reflect a strong charge separation between the two six-membered aromatic rings due to the effects of electron-donating groups on the chlorine-containing ring and electron-withdrawing substituents on the pyridine ring. It is, conceptually, also explained by the potential for ammosamide A to exist in an equilibrium with its bis-iminoquinone tautomer (Scheme 2). Furthermore, in 1 and 2 the chlorine atom at C-7 is poised to engage in nucleophilic aromatic substitution with a suitable nucleophile, particularly when the molecule exists as its bis-iminoquinone tautomer.[13] This reactivity may be relevant to the molecule's interaction with its protein target.[14]



Scheme 2

Possible tautomeric form of ammosamide A (1).



Ammosamides A (1) and B (2) exhibited significant in vitro cytotoxicity against HCT-116 colon carcinoma, each with IC50=320 nM. These compounds also demonstrated pronounced selectivity in a diversity of cancer cell lines with values ranging from 20 nM to 1 μM, indicating a specific target mechanism of action. To explore the intracellular target of the ammosamides, ammosamide B (2) was converted to a highly fluorescent molecule by conjugation.[14] Treatment of HCT-116 colon carcinoma or HeLa cells with this fluorescent molecule produced immediate and irreversible labeling of a specific protein in the cellular cytosol. Using a cell and molecular biology approach, the target of the ammosamides was identified as a member of the myosin family, important cellular proteins that are involved in numerous cell processes, including cell cycle regulation, cytokinesis, and cell migration.

/pdf/the-ammosamides-structures-of-cell-cycle-modulators-from-a-3qa7ssfw7v.pdf

The Ammosamides: Structures of Cell Cycle Modulators from a Marine-Derived Streptomyces Species†

Publisher Summary A quality development candidate compound after a positive proof of concept clinical trial has a good chance of making it to be a marketed drug. It is the role of medicinal chemists to choose the right compounds to design and synthesize. Although toxicity is now the joint primary cause of failure of clinical candidates, there is a growing body of information about structure–toxicity relationships even as those relate to complex bioactivation pathways. With this ever-growing knowledge, it should be possible for talented medicinal chemists to design inherently safer molecules. The introduction of a broader array of in vitro toxicology studies earlier in the drug discovery process should result in high quality development candidates entering the clinic and a richer database from which future structure–toxicity relationships may be discovered. As part of medicinal chemistry design, the properties of the molecules are predicted and decisions are taken based on those predictions. From measured physical and in vitro properties, it is also possible to predict pharmacokinetic and pharmacodynamic data to estimate the dose of development compounds. Good medicinal chemistry design, taking into account both potency at the target and also the predictive toxicology discussed in this chapter, increases the probability of a project's success and heavily influences speed of reaching a clear proof of concept outcome.

Designing Drugs to Avoid Toxicity

Information on bacterial thioamide metabolism has focused on transformation of the antituberculosis drug ethionamide and related compounds by Mycobacterium tuberculosis. To study this metabolism more generally, a bacterium that grew using thioacetamide as the sole nitrogen source was isolated via enrichment culture. The bacterium was identified as Ralstonia pickettii and designated strain TA. Cells grown on thioacetamide also transformed other thioamide compounds. Transformation of the thioamides tested was dependent on oxygen. During thioamide degradation, sulfur was detected in the medium at the oxidation level of sulfite, further suggesting an oxygenase mechanism. R. pickettii TA did not grow on thiobenzamide as a nitrogen source, but resting cells converted thiobenzamide to benzamide, with thiobenzamide S-oxide and benzonitrile detected as intermediates. Thioacetamide S-oxide was detected as an intermediate during thioacetamide degradation, but the only accumulating metabolite of thioacetamide was identified as 3,5-dimethyl-1,2,4-thiadiazole, a compound shown to derive from spontaneous reaction of thioacetamide and oxygenated thioacetamide species. This dead-end metabolite accounted for only ca. 12% of the metabolized thioacetamide. Neither acetonitrile nor acetamide was detected during thioacetamide degradation, but R. pickettii grew on both compounds as nitrogen and carbon sources. It is proposed that R. pickettii TA degrades thioamides via a mechanism involving consecutive oxygenations of the thioamide sulfur atom.

Metabolism of thioamides by Ralstonia pickettii TA.

Azo-coupling of pyrazole-3(5)-diazonium chlorides with cyanothioacetamide: a convenient synthesis of pyrazolo[5,1-c][1,2,4]triazine-3-carbothioamides

5.14 – Thionoamides and Their Selenium and Tellurium Analogues

Thioacetamide, thioacetamide S-oxide (1), and thiopivalamide S-oxide (8) were oxidized with H2O2 in H2 18O to generate the corresponding amides with at least 50% 18O incorporation; hydrolysis of the S-oxides to the amides or 18O exchange with the amides occurs much more slowly. When 1 and trifluorothioacetamide (6) were oxidized with three and four equivalents of H2O2, respectively, in the presence of benzylamine, N-acetylbemylamine (5) and N-benzyltrifluoroacetamide (7) were isolated in respective yields of 17% and 37%. These results are interpreted as evidence for an oxidative desulfurization mechanism involving nucleophilic attack at the carbon atom of an S,S-dioxide or trioxide intermediate as the major pathway; a minor pathway may involve the intermediacy of an oxathiirane S -oxide (3) or dioxide (4) species. Understanding is added to the behavior of S-oxides in aqueous solution, as well as to thermal stability in deuterochloroform. Also reported are studies of the preparation and properties...

John H. Hillhouse

Papers

Thiono compounds, 7. oxidation of thioamides in relation to adverse biological effects1