C26 H17 N O6
University of British Columbia (Originator)
uncialamycin, an enediyne natural product isolated from the Streptomyces uncialis, bacteria present on the surface of the lichen Cladonia uncialis.
Laboratory cultures of an undescribed streptomycete obtained from the surface of a British Columbia lichen produce uncialamycin (1), a new enediyne antibiotic.Uncialamycin exhibits potent in vitro antibacterial activity against Gram-positive and Gram-negative human pathogens, including Burkholderia cepacia, a major cause of morbidity and mortality in patients with cystic fibrosis.
Uncialamycin is an enediyne antibiotic with some unprecedented activity. The isolationists have filed a patent application almost right away. The total synthesis by Nicolaou [ACIE2007, 46, 4704] goes along nearly the same lines that have been predicted, and similar to Myers’ synthesis of dynemicin A [JACS 1997, 119, 6072], only it is not paper chemistry but the real one.
They have easily constructed the quinoline system with required functionality and subjected it to AllocCl-assisted acetylide addition (if I interpreted correctly “92% yield based on 80% conversion”). 5-alkoxyquinoline system was later advanced to iminoquinone and the two remaining rings were again attached by Hauser annulation with 3-cyanophthalide. The final product turned out be different from the one reported, more precisely, it was a C26-epimer. It is funny that I have accidentally drawn the correct structure with R-configuration at C-26 in the previous post.
The synthetic scheme allowed to easily invert this stereocenter via oxidation/reduction sequence on the last compound shown on the scheme below. The spectral properties of the final product thus obtained matched the reported data, and the structure of uncialamycin was confirmed by X-ray, despite it was isolated as an oil. The structure on the right is the revised one. The remaining details, including the chemistry behind DNA-cleaving Bergmann cycloaromatization,
Total Synthesis and Stereochemistry of Uncialamycin
K. C. Nicolaou, Hongjun Zhang, Jason S. Chen, James Crawford, Laxman Dasunoori
1Department of Chemistry and, The Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA
2Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093, USA
A new tot synth of Uncialamycin by Nicolaou. This is a natural occurring enediyne. Because the stereochemistry of C26 was unknown, both diastereomers as shown were synthesized. The retrosyn led back to simpler fragments 2, 3, and 4.
The following scheme illustrates the route to fragment 2. The key transformation was the two-step Friedlander quinoline synthesis (7 to 9).
Then fragment 2 was used in the following sequence. The key steps in the sequence involved installation of enediyne fragment 3 to give 11, the closure of the macrocycle to give 15, and the Hauser annulation in the last step to give 1a from 16.
In this case, it was found that the final product’s spectrum (1a) did not match the reported value. And therefore, the other isomer was synthesized. This was easily done using fragment 12 through oxidation-reduction sequence to give 18 with the opposite stereochemistry at C26. Sequence in Scheme 3 was then repeated on this fragment.
And 1b was found to match spectrum of the natural isomer. This natural compound was found to be stable as a solid and as solutions in a variety of solvents. But in presence of dray HCl in CH2Cl2 solution at rt, it rapidly converts to hexacyclic 19 through a cascade of Bergman cycloaromatization reaction. This cascade of reactions is believed to be responsible mode of action in damaging DNA and killing cells.
The enediynes are a family of antibiotics that possess a distinctive strained nine- or ten-member ring system comprising a Z-carbon-carbon double bond and two carbon- carbon triple bonds, usually arranged with the latter two flanking the former. The enediynes are potent damagers of DNA, causing single and double strand cuts. Their potency is attributed to their ability to bind to DNA and undergo a Bergmann rearrangement in which the strained ring system is converted into a highly reactive 1 ,4-benzenoid diradical, which damages the DNA by abstracting hydrogens from it.
Uncialamycin is an enediyne isolated from a Streptomyces strain found on the lichen Cladonia uncialis (Davies et al. 2005; 2007). (Full citations for references cited in this specification by first named author or inventor and year are provided in the section entitled “REFERENCES” later herein.)
The structure of uncialamycin has been confirmed by total synthesis (Nicolaou et al. 2007a; 2007b). In the course of the synthesis, it was noted that the unnatural 26(S) epimer was almost as active as the natural 26(R) epimer – that is, the stereochemistry of the C27 methyl had a minor effect on biological activity. Both epimers were active against several ovarian tumor cell lines. The IC50 values rang ed from 9 x 10“12 to 1 x 10“10, depending on the epimer and cell line or sub-line (Nicolaou et al, 2008).
Conjugates are an important method for the delivery of anti-cancer drugs, which are often highly cytotoxic and might otherwise be problematic to administer due to the risk of systemic toxicity. In a conjugate, the drug is conjugated (covalently linked) to a targeting moiety that specifically or preferentially binds to a chemical entity characteristic of the cancer cell, thus delivering the drug there with high specificity. Further, the drug is held in an inactive form until released from the conjugate, usually by cleavage of the covalent linker.
Typically, the targeting moiety is an antibody or an antigen-binding portion thereof, whose antigen is overexpressed or uniquely expressed by a cancer cell (“tumor associated antigen”). In such instances, the resulting conjugate is sometimes refered to as an “immunoconjugate” or an “antibody-drug conjugate” (ADC). Preferably the tumor associated antigen is located on the surface of the cancer cell, but also can be one that is secreted into the vicinal extracellular space. Upon binding, the antigen-conjugate complex is internalized and eventually finds its way inside a vesicular body such as a lysosome, where the covalent linker is cleaved, liberating active drug to exert its chemotherapeutic effect.
Advantageously, the covalent linker is designed such that cleavage is caused by a factor prevalent inside a cancer cell but not in plasma. One such factor is the low lysosomal pH, so that the covalent linker can be an acid-sensitive group such as a hydrazone. Another such factor is the generally higher intracellular concentration of glutathione, allowing for the cleavage of a disulfide covalent linker by a disulfide exchange mechanism. Yet another such factor is the presence of lysosomal enzymes such as cathepsin B, which can cleave peptide linkers designed to be preferred substrates (Dubowchik et al. 2002).
Conjugates have been used to deliver enediyne drugs in oncology. Gemtuzumab ozogamicin (Mylotarg®) is a conjugate of an anti-CD33 monoclonal antibody and a derivative of the enediyne calicheamicin. It was approved for treatment of acute
myelogenous leukemia but was later withdrawn from the market. Several other enediyne drugs, especially in the conjugated form, have been the subject of development efforts
If handled carefully, enediynes make powerful cancer drugs.
Inventors N. S. Chowdari, S. Gangwar, and B. Sufi synthesized enediyne compounds with general formula 1 that are based on the natural enediyne uncialamycin (2) scaffold (Figure 1). These compounds, used alone or in conjugates, are potent cytotoxins that may be useful in cancer chemotherapy.
Enediynes are a class of natural antibiotics that are characterized by 9- or 10-membered rings that contain two C≡C bonds separated by a cis (Z)-substituted C=C bond. Enediynes can undergo Bergman cyclization to form 1,4-benzenoid diradicals, which abstract hydrogen atoms from other molecules. When the diradical is generated near DNA, it abstracts hydrogen atoms from the sugar backbone of the DNA molecule and results in single- and double-strand lesions.
The high reactivity of enediynes toward DNA makes them very toxic. Their potent activity may be beneficial, however, if they are used to target the DNA of cancerous tumors. Most enediynes inhibit the proliferation of various cancer cells, including those that resist other chemotherapeutic drugs. Several naturally occurring enediynes are in clinical trials against cancer.
Both epimers at C26 of the natural enediyne uncialamycin are active against several ovarian tumor cell lines, with IC50 values ranging from 9 × 10–12 to 1 × 10–10 M, depending on the epimer and the cell line or subline. The synthetic enediynes described by the inventors are derivatives of uncialamycin.
Using these toxic molecules demands specific delivery systems. Conjugates are innovative drug-delivery systems designed to target tumor cells precisely and minimize the risk of systemic toxicity. Typically, drugs are linked covalently to conjugates that act as targeting moieties, which specifically or preferentially bind to a chemical entity characteristic of the cancer cell.
The covalent linker is designed to be cleaved only by a factor that exists inside a cancer cell and not in plasma, so that the drug remains in an inactive form until it is released from the conjugate. A typical targeting moiety may be a polymer or an antibody. Polymer-conjugated and antibody-linked enediyne drugs such as gemtuzumab ozogamicin (Mylotarg) were used to deliver enediyne drugs to cancer cells. Mylotarg, however, has been withdrawn from the market because of high patient mortality.
Compounds of structure 1 may be conjugated to a targeted moiety through a chemical bond to substituent R1. Compounds 3 and 4, shown in Figure 2, are examples of the synthetic enediynes with structure 1.
The investors tested the antiproliferative activities of several compounds against cancer cell lines. EC50 data for compounds 3 and 4 against 786-0 renal cancer cells and H226 lung cancer cells are shown in the table:
Several assays were also conducted on conjugates derived from other compounds of formula 1. (Bristol-Myers Squibb [Princeton, NJ]. WIPO Publication 2013122823, Aug 22, 2013;
DAVIES ET AL.: ‘UNCIALAMYCIN, A NEW ENEDIYNE ANTIBIOTIC‘ ORGANIC LETTERS vol. 7, no. 23, 13 October 2005, pages 5233 – 5236
Table 1. 13C and 1H NMR Assignments for Uncialamycin (1). Data were Recorded in DMSO-d6 at 600 MHz for 1H
|position||δ 13C||δ 1H (mult., J (Hz))|
|1||10.0 (d, 4.6)|
|6||126.1c||8.23 (dd, 1.4, 7.6)c|
|7||133.6d||7.88 (ddd, 1.4, 7.6, 7.6)d|
|8||134.9d||7.94 (ddd, 1.4, 7.6, 7.6)d|
|9||126.6c||8.24 (dd, 1.4, 7.6)c|
|17||63.0||5.14 (d, 3.3)|
|20||123.4||6.05 (dd, 0.8, 10)|
|21||124.0||5.97 (ddd, 1.4, 1.5, 10)|
|24||43.2||5.04 (dd, 1.5, 4.6)|
|26||63.6||4.31 (qd, 6.0, 6.0)|
|27||22.1||1.30 (d, 6.0)|
a−d May be interchanged.http://pubs.acs.org/doi/suppl/10.1021/ol052081f/suppl_file/ol052081fsi20051004_065853.pdf
Isolation of Uncialamvcin
 As part of a screening program aimed at discovering new antibiotics active against Bcc, it was found that crude organic extracts of cultures of a previously undescribed Streptomycete showed potent in vitro inhibition of Bcc. Bioassay guided fractionation of the crude extracts led to the identification of uncialamycin (1), a new enediyne antibiotic, as the active component. Bioactivity-guided fractionation involves thin layer chromatography of the extracts and fractions thereof and detection of the activity by overlaying a sensitive tester strain. A zone of inhibition identifies the active fraction containing the active compound.
The producing strain was extracted from the surface of the lichen Cladonia uncialis collected near Pitt River, British Columbia. Characterisation by 16S RNA sequencing showed the strain to be related, but not identical, to Streptomyces cyanogenus. Antibiotic activity of the strain was assayed by cutting plugs from solid agar cultures of the strain and placing them on lawns of tester strains of bacteria. Good inhibitory activity was detected against Gram-positive and Gram-negative bacteria (including Bcc), but not against yeasts.
Production cultures of the producing strain were grown as lawns on solid agar medium ISP4 for 14 to 21 days at room temperature. The solid agar cultures were lyophilized and extracted repeatedly with EtOAc. Concentration of the combined EtOAc extracts in vacuo gave a gummy residue that was partitioned between EtOAc and H2O. The EtOAc soluble material was fractionated by sequential application of flash C- 18 reversed-phase chromatography (eluent: step gradient from H2O to MeOH) and reversed-phase HPLC (column-Inertsil ODS-2; eluent: CH3CN/H2O 40:60) to give pure uncialamycin (1) (~ 300 μg) as a bright purple [UV(MeOH): λmaxnm (ε) 206 (25,000), 254 (33,000), 280 (shoulder), 320 (shoulder), 539 (9,400)], optically active ([α]D +3,300 (c 0.005, MeOH)) oil.
Chemical Characterization of Uncialamycin
Uncialamycin (1) gave a [M + Na]+ ion at m/z 462.0956 in the
HRESIMS appropriate for a molecular formula Of C26H17NO6 (calc’d for C26H17NO6Na 462.0954) requiring 19 sites of unsaturation. NMR data for uncialamycin was recorded in DMSO-^6 at 600 MHz using a cryoprobe. The 13C NMR spectrum (Table 1) showed well-resolved resonances for 26 carbon atoms and the 1H NMR spectrum contained resonances integrating for 17 protons in agreement with the HRMS data. Inspection of the HMQC data revealed that four of the protons (δ 5.39, 6.66, 10.0, and 13.2) were not attached to carbon atoms. Two major fragments A and B (Figure 1) of uncialamycin could be identified from analysis of the COSY, HMQC, and HMBC data obtained for the molecule.
Position δ 1W WH^mult, J(Hz)) ,
1 10.0 (d, 4.6)
6 126.1 8.23 (dd, 1.4, 7.6)
7 133.6 7.88 (ddd, 1.4, 7.6, 7.6)
8 134.9 7.94 (ddd, 1.4, 7.6, 7.6)
9 126.6 8.24 (dd, 1.4, 7.6)
14 129.9 8.51 (s)
17 63.0 5.14 (d, 3.3)
20 123.4 6.05 (dd, 0.8, 10)
21 124.0 5.97 (ddd, 1.4, 1.5, 10)
24 43.2 5.04 (dd, 1.5, 4.6)
26 63.6 4.31 (qd, 6.0, 6.0)
27 22.1 1.30 (d, 6.0)
13-OH 13.2 (brd.s)
17-OH 6.66 (brd.s)
26-OH 5.39 (d,6.0)
Table 1. C and H NMR assignments for uncialamycin (1). Data were recorded in OMSO-d6 at 600 MHz for 1H.  A pair of olefinic resonances at δ 5.97 (H-21 ) and 6.05 (H-20), that were strongly correlated to each other in the COSY spectrum and had a coupling constant of 10 Hz, were assigned to a cis disubsituted olefin. The upfield olefinic resonance at δ 5.97 (H-21) showed strong HMBC correlations to non-protonated carbon resonances at δ 89.7 (C- 19) and 98.9 (C-23), and the downfield olefinic resonance at δ 6.05 (H-20) showed strong correlations to non-protonated carbon resonances at δ 87.4 (C-22) and 100.4 (C- 18). This suite of HMBC correlations identified an enediyne substructure in 1 (see Fragment A in Figure 1). The olefinic resonance at δ 5.97 (H-21) showed a long range COSY correlation to a methine resonance at δ 5.04 (H- 24), indicating that the carbon bearing the methine proton (C-24: δ 43.2) was attached to the C-23 alkyne carbon. A COSY correlation observed between the methine (δ 5.04, H-24) and a broad singlet at 10.0, that was not correlated to a carbon in the HMQC spectrum, and the chemical shift of the methine carbon (C-24, δ 43.2) suggested that C-24 had an NH substituent. HMBC correlations observed between the H-24 methine (δ 5.04) and the two alkyne carbon resonances at δ 87.4 (C-22) and 98.9 (C-23) confirmed the attachment of C-24 to the C-23 alkyne carbon.
A singlet methine resonance at δ 5.14 (H- 17) showed HMBC correlations to the alkyne carbon resonances at δ 89.7 (C- 19) and 100.4 (C- 18), which demonstrated that the methine carbon (C- 17: δ 63.0) was linked to the second alkyne at C-18. Both of the methine resonances at δ 5.04 (H- 24) and 5.14 (H- 17) showed HMBC correlations to a pair of deshielded resonances at δ 63.5 (C- 16) and 76.0 (C-25), assigned to non-protonated oxygen bearing carbons. This set of four HMBC correlations indicated that the two oxygenated carbons bridged the C- 17 and C-24 carbons to form a ten membered ring (C- 16 to C-25) containing the enediyne substructure. A COSY correlation between the methine resonance at δ 5.14 and a broad singlet at 6.66 (17-OH) revealed an alcohol funtionality attached to the methine carbon.
A methyl doublet at δ 1.30 (Me-27, J = 6 Hz) was correlated in the COSY spectrum to a methine at 4.31 (H-26, q, J = 6.0 Hz)) that was further correlated to a broad singlet at 5.39 (OH-26), assigned to an alcohol. The methyl resonance (δ 1.30, Me-27) showed an HMBC correlation to the carbon resonance at 76.0 (C-25), indicating that the hydroxyethyl fragment (C-26 and C-27) was the fourth subsituent on the non-protonated carbon C- 25. Both the NH-I proton (δ 10.0) and the H-17 methine (5.14) were correlated to a carbon at δ 135.6 (C- 15), and the H-24 methine (δ 5.04) was correlated to a carbon at 143.6 (C-2) in the HMBC spectrum indicating that the NH and C- 16 were vicinal substituents on an olefin or aromatic ring. A deshielded singlet at δ 8.51 showed strong HMBC correlations into carbon resonances at δ 63.5 (C-16), 143.6 (C-2), and 112.7 (C- 12) and a weak correlation into the carbon resonance at 154.9 (C- 13). This set of HMBC correlations confirmed that the NH and C-16 were attached to a benzene ring. Based on the assumption that the intense HMBC correlations were through three bonds, these correlations also indicated that the aromatic methine (δ 8.51, H-14) was ortho to C-16 (δ 63.5) and meta to the NH (C-2, δ 143.6). The weak HMBC correlation between δ 8.51 and 154.9 was attributed to a two bond coupling, placing the carbon at 154.9 (C-13) ortho to the methine carbon (C- 14) and its chemical shift required an oxygen substituent.  The second fragment B of uncialamycin contained an isolated
1H spin system comprised of four contiguous aromatic protons (δ 8.23, dd, J = 1.4, 7.6 Hz H-6; 7.88, ddd, 1.4, 7.6, 7.6 Hz H-7; 7.94, ddd, J = 1.4, 7.6, 7.6 Hz H-8; 8.24, dd, J = 1.4, 7.6 Hz H-9). HMBC correlations observed between the proton resonance at δ 8.23 (H-6) and a carbon resonance at 187.0 (C-4) and between the proton resonance at 8.24 (H-8) and a carbon resonance at 182.2 (C-11) suggested that the other two subsituents on the benzene ring were quinone carbonyls. Fragments A and B shown in Figure 1 accounted for all of the carbon, hydrogen, and nitrogen atoms in the molecular formula of uncialamycin (1), but contained one extra oxygen atom. In order to complete the quinone and satisfy the remaining aromatic valences in Fragment A, the two carbonyl carbons of fragment B (C-4 and C-I l) had to be attached to the two substituted aromatic carbons (C-3 and C- 12) of fragment A. Finally, it was apparent that the two oxygentated carbons C- 16 and C-25 had to be bridged by an epoxide to account for the number of oxygen atoms and sites of unsaturation required by the molecular formula of 1. This implied that the C- 13 oxygen substituent had to be part of a phenol functionality that would engage in intramolecular hydrogen bonding with the C-I l carbonyl consistent with the observed OH chemical shift of δ 13.2.
A ROESY correlation between δ 5.14 (H- 17) and 4.31 (H-26) showed that C-26 and C- 17 were cis oriented about the C-16/C-25 epoxide and also defined the relative stereochemistry of H- 17 as shown. Molecular models revealed that due to steric and bond angle strain the C- 17 to C-23 enediyne containing bridge could only reasonably be cis fused to the piperidine ring. Uncialamycin (1) shares structural features with dynemicin A (2) and deoxydynemicin A (3) isolated from Micromonospora chersina. The H-24 resonance in uncialamycin (1) has a chemical shift of δ 5.04 and a 4.6 Hz coupling to the NH-I proton, which is nearly identical to the chemical shift (δ 5.05) and coupling (J = 4.3 Hz) of the corresponding methine proton (H-2) in dynemicin A (2), in agreement with the relative stereochemical assigment at C-24 in 1. Comparison of the additional NMR assigments reported for dynemicin A (2) and its triacetate derivative provided further strong support for the assigned structure of uncialamycin
Angewandte Chemie – International Edition, 2008 , vol. 47, 1 p. 185 – 189
The highly potent DNA-cleaving molecule uncialamycin (1) was prepared in an asymmetric total synthesis featuring an enantioselective Noyori reduction. Compound 1 and its C26 epimer exhibit impressive broad-spectrum antibacterial properties and highly potent antitumor activities against a variety of cell lines.