Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Nov 29.
Published in final edited form as: J Nat Prod. 2010 Nov 8;73(11):1927–1932. doi: 10.1021/np100600z

Scytonemides A and B, Cyclic Peptides with 20S Proteasome Inhibitory Activity from the Cultured Cyanobacterium Scytonema hofmanii

Aleksej Krunic 1, Armelle Vallat 1,, Shunyan Mo 1, Daniel D Lantvit 1, Steven M Swanson 1, Jimmy Orjala 1,*
PMCID: PMC3074188  NIHMSID: NIHMS251439  PMID: 21058727

Abstract

Two cyclic peptides, scytonemides A (1) and B (2), were isolated from the cultured fresh water cyanobacterium Scytonema hofmannii (UTEX 1834) by bioassay-guided-fractionation using a proteasome inhibition assay. The planar structures of the compounds were determined by a combination of MS and 1D- and 2D-NMR spectroscopy. The advanced Marfey’s method was used to determine the absolute configuration of both peptides. Scytonemide A possesses an unusual imino linkage, while scytonemide B is a depsipeptide containing 3-hydroxyoctanoic acid in the macrocycle. Both isolates were evaluated for their inhibition of the 20S proteasome, and scytonemide A displayed an IC50 value of 96 nM, while scytonemide B was inactive at 50 μM.


Cyanobacteria have been found to be an abundant source of secondary metabolites with diverse biological activities and chemical structures Numerous biologically active natural products have been reported from cyanobacteria, including alkaloids, polyketides, and linear as well as cyclic peptides.13 These cyanobacterial peptides often contain modified amino acid residues, and the addition of a lipid component. Many cyanobacterial peptides have been shown to be inhibitors of various proteases.4 As part of our ongoing effort to find novel bioactive natural products from cultured freshwater and terrestrial cyanobacteria, we have evaluated cyanobacterial extracts for the inhibition of the 20S proteasome. The 20S proteasome is the catalytic core of the proteasome complex, the major proteolytic system in the cytosol of eukaryotic cells, and catalyzes the degradation of regulatory proteins.5,6 The proteasome complex plays a central role in maintaining cellular homeostasis and controlling the cell cycle.79 Accordingly, proteasome inhibitors are of interest as potential anticancer agents.1015 Currently, there is one FDA-approved proteasome inhibitor on the market (bortezomib, Velcade)16, 17 and there is an ongoing research effort to identify additional inhibitors.

In our studies, the organic extract of the cultured cyanobacterium Scytonema hofmannii (UTEX 1834) showed significant inhibitory activity against the human 20S proteasome. Bioassay-guided fractionation resulted in the identification of two new cyclic peptides, scytonemides A (1) and B (2). Scytonemide A, which contained an unusual imine linkage in the cyclic peptide core, was found to be a potent inhibitor of the 20S proteasome. Scytonemide B, while inactive in 20S proteasome assay, was identified as a cyclic lipopetide with the unusual amino acid D-citrulline and a hydroxylated fatty acid (3-hydroxyoctanoic acid). Herein, we report the isolation, structure elucidation and biological activity of these compounds.

Results and Discussion

Lyophilized cell material of S. hofmannii was extracted with MeOH-CH2Cl2 (1:1), and the concentrated extract was fractionated by flash chromatography on silica gel using mixtures of organic solvents of increased polarity, concluding with 100% MeOH. Reversed-phase HPLC of the 100% MeOH fraction led to isolation of 1 and 2.

graphic file with name nihms251439u1.jpg

Scytonemide A was obtained as a white amorphous powder. The molecular formula was determined as C36H56N8O9, based on the HRMS data (m/z 74454.4186 [M+H]+). Inspection of the DEPTQ and 1H NMR spectra indicated nine quaternary, 12 methine, seven methylene, and six methyl carbons, corresponding to 34 carbons and 44 protons. The signal intensity at δC 131.4 and 115.2 suggested two additional methine carbons were present as part of a para-substituted phenyl ring, thus bringing the total number of carbon signals to 36 and protons to 46. Based on the molecular formula, eight nitrogens, nine oxygens, and 10 exchangeable protons were also present in 1.

The 1H and DEPTQ NMR spectra of 1 (Table 1) displayed signals characteristics of a peptide: six exchangeable amide NH protons (δH 8.55, 8.40, 8.17, 7.90, 7.78, and 7.36) as well as one NH2 group (δH 7.33 and 6.78) of a glutamine were evident. The amide protons correlated with α-hydrogen signals (δH 4.93, 4.37, 4.33, 4.32, 4.17, 4.01, and 3.53) in the COSY spectrum. In the upfield region of the spectra, six doublet or triplet methyl groups appeared with chemical shifts at δH 0.93, 0.90, 0.86, 0.83, 0.80, and 0.72. The two downfield doublets at δH 6.86 (d, H-5/6, J = 8.5 Hz) and 6.61 (d, H-8/9, J = 8.5 Hz) were assigned to a para-substituted phenyl ring, also apparent in the DEPTQ spectrum. The presence of seven amide carbonyl resonances (δC 173.68, 172.93, 172.58, 172.56, 171.28, 170.15, and 170.31) in the DEPTQ NMR spectrum further implied the peptidic nature of this compound. An unusual low-field resonance at δC 166.0 connected to a proton at 6.80 ppm was also observed (Table 1). This resonance was assigned as the carbon of an imino double bond in a modified Leu residue. This accounted for all eight nitrogens required by the molecular formula as well as eight of the exchangeable hydrogens and seven of the oxygens. The two remaining protons and oxygens were assigned to hydroxy functional groups.

Table 1.

NMR Data for Scytonemide A (1) in DMSO-d6a

position δC, mult. δH, mult., J(Hz) COSY HMBCb NOESY/ROESY
Tyr 1 172.93, C
2 74.0, CH 3.53, dd (11.3, 2.8) H-3a,3b 4, Leu 1 H-5/9, H-1 Leu
3a 40.1, CH2 3.04, dd (13.5, 2.4) H-2 1, 4 H-5/9
3b 2.48, ddd (13.5, 13.5, 2.4)
4 127.6, C
5/9 131.4, CH 6.86, d (8.5) H-6/8 3, 7, 9/5 H-3, 3′, H-6/8
6/8 115.3, CH 6.61, d (8.5) H-5/9 4, 8/6 H-5/9
7 156.4, C
OH
Gly 1 171.28, C
2a 40.6, CH2 4.32, m NH 1, Tyr 1 H-2b
2b 3.42, m 1, Tyr 1 H-2a, NH
NH 7.78, dd (10.0, 2.6) H-2 2, Tyr 1 H-2a, H-2 Gln, Gln-NH2, H-3b Ser, H-2 Tyr, H-2 Val
Gln 1 172.58, C
2 53.8, CH 4.33 H-3a,3b, NH 1 H-3a, 3b, H-4b, NH, Ile-NH
3a 25.7, CH2 1.85, m H-2, H-4a,4b 1, 4 H-2
3b 1.76, m 1, 5
4a 31.8, CH2 2.11 H-3a,3b 1, 5 H-2
4b 1.99
5 173.68, C
NH 8.55, d (6.3) H-2 1, 2, 3, Gly 1 H-2, H-2a, 2b Gly
NH2 7.33, br s 5 H-4
6.78, br s 4
Ile 1 170.15, C
2 58.0, CH 4.17, dd (9.2, 3.9) H-3, NH 1, 3, 4, Gln 1 H-3, NH, Ser-NH
3 35.7, CH 2.06, m H-4, 3-Me 2 H-2, 3-Me, NH
3-Me 16.4, CH3 0.86, d (7.0) H-3 2, 3, 4 H-3, NH
4 24.1, CH2 1.28, m H-3, 5 2, 3-Me
5 12.5, CH3 0.83, m H-4 3, 4
NH 8.17, d (9.3) H-2 2, 3, Gln 1 H-2, Gln-2, Ser-NH
Ser 1 172.56, C
2 53.0, CH 4.93, m H-3a, NH 1, 3 H-3a, 3b, NH, Val-NH
3a 62.8, CH2 3.87, dd (10.0, 10.0) H-2 1, 2 H-2, H-3b
3b 3.60, dd (9.8, 6.0) 1 H-2, NH, Gly-NH, H-3b Tyr, H-5/9 Tyr
OH
NH 7.36, d (9.0) H-2 Ile 1 H-2, 3, Ile-NH, Ile-2, Gln-2, Val-2
Val 1 170.31, C
2 59.7, CH 4.01, dd (7.7, 4.2) H-3, NH 1, 3, 3-Me, Ser 1 H-3, Leu-NH
3 29.0, CH 2.22, m H-4, 3-Me 1 H-2, H-4
3-Me 17.4, CH3 0.90d (7.0) H-3 2, 3, 4 H-3
4 19.7, CH3 0.93, d (7.0) H-3 2, 3, 3-Me H-3
NH 8.40, brs H-2 H-Ser-2
Leu 1 166.0, CH 6.80, d (1.8) H-2 3, Tyr 2 H-Tyr-2
2 49.6, CH 4.37, m H-1, 3, NH 1, 3, 4 H-1, H-3a, 3b, H-4, NH
3a 40.6, CH2 1.05 H-2, 4 1, 4-Me, 5 H-2, 4-Me
3b 0.81 1
4 24.8, CH 1.29, m H-3a, 4-Me 2, 3
4-Me 23.4, CH3 0.80,d (6.7) H-4 3, 4, 5 H-3a
5 21.7, CH3 0.72, d (6.5) H-4 3, 4, 4-Me
NH 7.90, d (9.4) H-2 2, 3, Val 1 H-2, Val-2, Tyr-5/9
a

Frequency: 900 MHz for 1H and 226 MHz for 13C

b

HMBC correlations are from the proton(s) stated to the indicated carbon.

The presence of Gly, Val, Gln, Ile, Ser, Tyr, and a modified Leu residue was determined by interpretation of 1H, DEPTQ and 2D COSY, TOCSY, HSQC, HMBC and NOESY spectra (Table 1).

The amino acid sequence of the macrocycle was also confirmed by analysis of the HMBC correlations. However, resonance overlap between the Gln/Ser and Ile/Val carbonyls could not be resolved by standard gradient HMBC. Correlations to these carbons in the HMBC spectrum were at first ambiguous due to insufficient digital resolution (33557 Hz/128 points - 262 Hz/pt) in the 13C dimension, even at the highest field available to us. Digital resolution in the 13C dimension can be increased significantly through selective excitation of a narrow region (20 ppm) of the 13C spectrum using shaped pulses. These so-called “band-selective” or “semi-selective” HMBC experiments incorporate selective pulses on the H or X channel to excite only a subset of all carbon resonances and so focus the 2D correlation spectrum on these alone.18, 19

The “band -selective” HMBC experiment was applied to the carbonyl region of 1 (centered at 172 ppm). Selective excitation with a 1.2 ms custom Sinc pulse yielded digital resolution of 3018/256 points (12 Hz/pt), in about one third of instrument time compared to regular HMBC. The resulting 20-fold increase in digital resolution allowed for unambiguous assignment of the carbonyl carbon signals (Table 1 and Figure 1). The connectivity of the modified Leu to Val was established by HMBC correlation from Leu-NH (δH 7.90) to C-1 of Val (δC 170.31). Correlations between Val-2 (δH 4.01) and C-1 of Ser (δC 172.56) connected these two amino acids. The adjacent position of Ser and Ile was established by HMBC correlation of Ser-NH (δH 7.36) of to Ile-1 (δC 170.15). The Gln residue was determined to be adjacent to Ile by the HMBC correlation from H-2 of Ile (δH 4.17) and C-1 of Gln (δC 172.58). The signal at lowest field (δC 173.68) was assigned as the C-5 carbonyl of the Gln residue, based on the HMBC correlations from Gln protons 3a/b (δH 2.11 and 1.99) and 4a/b (δH 1.85 and 1.76) to C5. Gln was connected to Gly based on the HMBC correlation from Gln-NH (δH 8.55) to Gly-1 (δC 171.28). Correlations from both H-2a (δH 4.32) and NH of Gly (δH 7.78) to C-1 of Tyr (δC 172.93) confirmed the linkage of these two residues The structure of the 1 was completed by the HMBC correlation from H-2 of Tyr (δH 3.53) to the C-1 methine (δC 166.0) of the modified Leu, supporting the imino linkage and closing the heptapeptide macrocycle. The amino acid sequence was also confirmed by NOE correlations between alpha protons and amide protons on the neighboring amino acid residue (Table 1 and Figure 1).

Figure 1.

Figure 1

Key HMBC and NOESY correlations in 1 and 2

The determination of the absolute configuration of the amino acid units in 1 was carried out by chemical degradation. Acid hydrolysis of 1 followed by advanced Marfey’s analysis20, 21 of the resulting hydrolysate showed the presence of six amino acids: Gly, L-Tyr, L-Ser, L-Ile, D-Gln (as D-Glu), and L-Val. Peroxidation of 1 with meta-chloroperbenzoic acid (m-CPBA) and subsequent hydrolysis indicated the presence of an additional amino acid, L-Leu, in the acid hydrolysate (see Experimental Section). Based upon these results, the structure of scytonemide A was concluded as shown.

Scytonemide B was obtained as a yellow amorphous solid. The molecular formula was determined as C39H53N7O12, based on the HRESIMS of [M + H]+ and pseudomolecular ion [M + Na]+ peaks at m/z 812.3825 and 834.3640, respectively. The NMR data set for 2 was collected in a 3 mm Shigemi microcell assembly on a 900 MHz spectrometer equipped with a carbon sensitive cryoprobe. The initial structure elucidation of 2 was carried out on the first 247 μg of sample.22 Even this small amount of sample allowed for rapid collection and assignment of all 13C resonances through HSQC/HMBC experiments. Subsequent direct 13C data collection with a DEPTQ experiment23, 24 was performed with material re-isolated from a larger culture. The peptidic nature of 2 was evident from the 1H NMR spectrum in DMSO-d6 (Table 1), where four amide proton signals were observed. Two of the amide protons were found to be isochronous and remained unresolved at δH 8.50 even at 900 MHz. Two additional broad signals corresponding to three protons were found at δH 5.75 and 5.35. These protons were determined to be a part of the ureido moiety of citrulline (Cit), based on the number of nitrogens implied by the molecular formula. Two additional singlets, located further downfield (above δH 9.0 ppm), were assigned as the OH protons of Tyr and Ser. The exchangeable nature of these protons was confirmed in the HSQC spectra.

Detailed analysis of the 1H NMR spectrum, combined with the 2D COSY, TOCSY, HSQC, HMBC, and NOESY NMR data for 2 (Table 2), revealed the presence of the amino acids Phe, Tyr, Asp, Ser and Cit in the molecule (Figure 1). In the α proton region, a multiplet observed at δH 4.88 did not display correlations to any amide proton. The COSY pattern of this multiplet was sufficiently resolved at 900 MHz to allow mapping of its correlations to the remaining protons in the spin system. This, combined with the HMBC correlations from δH 4.88 to the ester carbonyl resonance at δC 168.3, identified the fragment as 3-hydroxyoctanoic acid (3-HOA). Together, these moieties accounted for 16 (two aromatic rings and eight ester/amide carbonyls) of the 17 degrees of unsaturation required by the molecular formula and the macrocyclic ring represented the remaining degree of saturation. The subsequent DEPTQ spectrum confirmed the number of carbon signals ascertained through indirect methods.

Table 2.

NMR Data for Scytonemide B (2) in DMSO-d6a

unit position δC, mult. δH, mult., J(Hz) COSY HMBCb ROESY
Ser 1 172.6, C -
2 55.3, CH 4.41 dd (7.8, 4.6) H-3a, 3b′, NH 1, 3 NH, Phe NH
3a 62.9, CH2 4.37 dd (11.4, 4.6) H-1, OH 1, 2 Phe NH
3b 3.84 dd (12.3, 3.6) OH, NH
OH - 3.84 d (11.2) H-2, H-2′ H-3b, H-3bTyr
NH - 8.50 d (7.6) H-1 2, 3, 1 HOA H-2, H-3b, Phe NH, H-2 3-HOA
Phe 1 171.25, C -
2 56.5, CH 4.20 ddd (9.1, 4.1, 4.0) H-3a, NH 1, 3, NH H-3b, NH
3a 35.3, CH2 3.02, dd (14.8, 3.4) H-2, H-3b 2, 5 H-2, NH, H-5/9
3b 2.89, dd (14.7, 9.8) H-2, H-3a
4 137.7, C -
5,6,8,9 128.6, CH 7.23, m H-7 3, 7
7 126.4, CH 7.20, dt (7.1, 7.6) H-6, H-8 5, 6, 7, 8
NH - 8.54, d (4.8) H-1 Ser 1 H-2, H-3a, H-3b, Ser NH, Ser H-2, Ser H-3a, Cit NH
Tyr 1 172.0, C -
2 57.2, CH 4.10, ddd (12.0, 7.5, 3.0) H-3a, NH 3, 4, NH H-3a, H-3b, NH
3a 35.4, CH2 2.96, dd (14.1, 12.4) H-2, H-3b 2, 5/9, NH H-3a, H-5/9
3b 2.75, dd (14.2, 3.2) H-2, H-3a NH, Ser OH
4 128.3, C -
5,9 129.7, CH 7.10, d (8.5) H-6, H-8 3 H-2, H-3
6,8 115.0, CH 6.67, d (8.5) H-5, H-9 5/9
7 155.7, C -
OH - 9.24, s 6, 7
NH 8.49, d (7.6) H-2 2, 3, Phe 1
Asp 1 171.48, C -
2 49.4, CH 4.32, ddd (5.7, 2.5, 2.5) NH 1, 3, NH H-3b
3a 39.5, CH2 2.54, H-3b 4, NH H-2, H-3b
3b 2.27, dd (13.7, 6.0) H-3a 2, 4, NH H-2, H-3a, NH
4 173.3, C -
NH - 8.17, d (8.4) H-2 Cit 1 H-2, H-3b
Cit 1 171.34, C -
2 52.7, CH 4.14, ddd (11.0, 7.0, 3.0) H-3b, NH 1, 3 NH
3a 28.5, CH2 1.90, m H-3a, H-4a, H-4b NH NH
3b 1.66, m H-2, H-3b, H-4b NH
4a 1.47, m H-4b, H-3b, H-5
4b 25.8, CH2 1.28, m H-3b, H-3a, H-4b, H-5
5a 39.1, CH2 2.89, m H-4a, NH (guan.) 3
5b 2.83, m H-4′
NH (amino) - 7.58, d (8.6) H-2 Tyr 1
NH (guan.) - 5.76, s H-5
NH2 (guan.) - 5.35, s
6 158.6, C 5
3-HOA 1 168.3, C
2a 40.3, CH2 2.59, dd (13.5, 3.3) H-2b 1, 3 H-3, Ser NH
2b 2.10, dd (13.4, 3.5) H-2a, H-3 H-3
3 71.4 4.88, ddd (8.0, 3.4, 2.0) H-2a, H-2b, H-4 Asp 1 H-2a, H-2b, H-4a. H-4b, H-5a
4a 31.4, CH2 1.54, m H-3, H-4′, H-5 5 H-3
4b 1.44, m H-4, H-5 H-3
5a 31.0, CH2 1.08, m H-4, H-4′, H-5′, H-6 4, 6, 7, 8 H-3
5b 1.14, m H-5
6a 25.1, CH2 1.26, m H-5 8
6b 1.09, m H-5
7/7′ 22.2, CH2 1.23, m H-8 5 H-6a
8 13.9, CH3 0.83, t (7.4) H-7 6, 7 H-7
a

Frequency: 900 MHz for 1H and 226 MHz for 13C.

b

HMBC correlations are from the proton(s) stated to the indicated carbon.

The amino-acid sequence was determined by a combination of HMBC and ROESY data (Table 2). The 13C NMR chemical shifts for the amide bond carbonyls belonging to amino acids Cit (δC 171.34), Phe (δC 171.25), and Asp (δC 171.48) were virtually indiscernible in a standard HMBC spectrum and were resolved by application of a band-selective HMBC experiment. The Tyr-Cit-Asp sequence was established by HMBC correlations from Asp-NH (δH 8.17) to C-1 of Cit (δC 171.34) and from Cit-NH (δH 7.58) to C-1 of Tyr (δC 172.0). A second sequence, Phe-Ser-3-HOA, was identified by the HMBC correlations from Phe-NH (δH 8.54) to C-1 of Ser (δC 172.6) and from Ser-NH (δH 8.50) to C-1 of 3-HOA (δC 168.3). HMBC correlations from Tyr-NH (δH 8.49) to C-1 of Phe (δC 171.25) and from H-3 of 3-HOA (δH 4.88) to C-1 of Asp (δC 171.48) connected these two fragments, thus completing the planar structure of 2. The amino acid sequence was also confirmed by NOE correlations between alpha protons and amide protons on the neighboring amino acid residue (Table 2 and Figure 1).

The determination of the absolute configuration of the amino acid units in 2 was carried out by chemical degradation. Acid hydrolysis of 2 followed by advanced Marfey’s analysis20, 21 of the resulting hydrolysate confirmed the presence of D-Cit and showed the presence of four other amino acids:, L-Asp, L-Ser, L-Tyr, and L-Phe. Attempts to selectively hydrolyze 2 at the site of the ester bond for Mosher’s analysis, were not successful under various reaction conditions and the configuration at C-3 of the 3-HOA moiety was not determined. The final structure of scytonemide B is as shown.

Both isolates were evaluated to determine their potential inhibition of the 20S proteasome. Scytonemide A was found to be a potent inhibitor of the 20S proteasome chymotrypsin catalytic activity (IC50 = 96 nM, with bortezomib as positive control), while scytonemide B was not active (IC50 >50 μM). Scytonemide A possesses a rare imine linkage, which has been only encountered previously in minor constituents from one species of Nostoc.25 The major structural difference between the reported nostocyclopetides A1/A225 and 1 is the replacement of 4-MePro found in the nostocyclopeptides with Val in 1. The possible mechanistic explanation for the action of non-covalent inhibitors of 20S proteasome has been recently explored by a combination of solution NMR and X-ray crystallography.26, 27 Nonpolar amino acid residues bind to the auxiliary hydrophobic binding site of the 20S proteasome, while Tyr and Asn (or Gln as seen in 1) side chains appear to be crucial for optimal fit in the main binding pocket. The presence of an imine in scytonemide A also opens up the possibility of a nucleophilic attack at the N-terminal Thr1OH and formation of a covalent linkage, similar to what has been described for other irreversible proteasome inhibitors.

Scytonemide A, while a potent 20S proteasome inhibitor in vitro, lacks activity in the HT-29 cell line (ED50 >20 μg/mL), even though the HT-29 cell line has been reported to be sensitive to the potent proteasome inhibitor, bortezomib.28 This difference may be explained by a lack of cell wall penetration or metabolic instability of scytonemide A. To evaluate if Scytonemide A penetrates the cell wall, we performed a luminescence based cellular 20S proteasome assay. Scytonemide A showed 80% inhibition at 6.7 μM, a concentration where no cytotoxic effect was observed in the same cell line. This indicates that chemical or metabolic instability in the longer HT-29 cytotoxic assay (3 days vs. 2 hr for cellular 20S proteasome assay) may be the reason for the lack of cytotoxic response. This aspect may be improved by chemical modifications of this compound.

Scytonemide B is a depsipeptide containing D-Cit and a short chain β-hydroxy fatty acid in the macrocycle. Substituted and functionalized 3-hydroxyoctanoic acid derivatives have previously been found in cyanobacterial depsipeptides, such as yanucamide A, lyngbyabellin B and trungapeptines A-C, but not the unsubstituted 3-hydroxyoctanoic acid.2931 3-Hydroxyoctanoic acid have been previously identified as a minor constituent in cyanobacterial mats from Antarctic lake sediments.32 Similarly, the presence of D-Cit has not been previously reported in fresh water cyanobacteria. There are few literature examples of incorporation of L-Cit in the peptide macrocycle, including nostopeptins E-K and the recently reported symplocamide A, from a marine Symploca sp.33, 34 Metabolic selection and incorporation of Cit over Arg, or metabolic conversion of Arg to Cit via a deamination process and subsequent epimerization are associated with non-ribosomal peptide synthase complexes.4

Structurally diverse cyclic peptides 1 and 2 represent the metabolic “fingerprint” of S. hofmanii. It would be of great interest to establish the functional roles that the unique building blocks play in both 1 and 2.

Experimental section

General Experimental Procedures

Optical rotations were measured on a Perkin-Elmer 241 polarimeter. UV spectra were obtained on Varian Cary 50 UV-VIS spectrophotometer equipped with a Xenon flash lamp. IR spectra were obtained on a JASCO FTIR-410 Fourier transform infrared spectrometer. NMR spectra were obtained on a Bruker AV900 MHz NMR spectrometer with a 5 mm ATM CPTCI Z-gradient probe, referenced to the corresponding solvent peaks. Low resolution ESI mass spectra were obtained on a ThermoFinnigan TSQuantum triple quadrupole mass spectrometer, and an Agilent 1946A LC-MSD single quadrupole LC-mass spectrometer. High-resolution ESI mass spectra were obtained on a Thermo Electron LTQ FT ICR mass spectrometer. LC-MS data for the advanced Marfey’s reaction were collected on an Agilent 1100 LC/MSD instrument.

Biological Material

Scytonema hofmannii was initially acquired from the Culture Collection of Algae at the University of Texas at Austin (UTEX 1834). The cyanobacterium was grown in a 2.8 L Fernbach flask containing 1 L of inorganic media (Z-45) without aeration.35 Cultures were illuminated with fluorescent lamps at 2 klx with an 18/6 hour light/dark cycle. The temperature of the culture room was maintained at 20 °C. After 6–8 weeks, the biomass of cyanobacteria was harvested by centrifugation and then freeze-dried.

Extraction and Isolation

The combined biomass from two 1 L cultures was extracted by repeated maceration with CH2Cl2:MeOH (1:1) to yield 270.5 mg of anextract, subsequently fractionated on silica gel using a gradient with increasing polarity [hexane-EtOAc (4:1), EtOAc-2-PrOH (5:2), 2-PrOH, MeOH] to afford eight fractions. Fraction 8 (4 mg) was subjected to reversed-phase HPLC (Alltima C8, 5 μm, 250×10 mm, 3 mL/min) with a solvent gradient of 5% MeCN/95% H2O to 100% MeCN over 35 min to afford scytonemide B (2) (tR = 14.2 min, 0.247 mg).

A second harvest from eight 1 L of cultures was extracted by repeated maceration with CH2Cl2-MeOH (1:1) to yield 257 mg of organic extract. The extract was fractionated using silica gel with the same method as described above. Fraction 8 (102.3 mg) was subjected to reversed-phase HPLC (Alltima C8, 5 μm, 250×10 mm, 3 mL/min) with a solvent gradient of H2O-MeCN (95:5) to 100% MeCN over 35 minutes to afford scytonemide A (1, tR = 19.2 min, 1.2 mg), and scytonemide B (2, tR = 14.2 min, 0.9 mg).

Scytonemide A (1)

[α]25D -15 (c 0.0013, MeOH); UV λmax (log ε) 201 (3.45), 224 (3.42), 277 (2.70) nm; IR (neat) νmax 3734, 3296, 2932, 2360, 1650, 1540, 1418, 1257, 1025, 800, 746 cm−1; 1H NMR and 13C NMR data shown in Table 1; ESIMS m/z 745 (M+H+, 30), 763 (100), 767 (40), 783 (55); FTMS m/z [M+H]+ 745.4186 (calcd for C36H57N8O9, 745.4165)

Scytonemide B (2)

[α]25D -10 (c 0.0014, CH3OH); UV λmax (log ε) 209 (4.15), 226 (3.87), 277 (3.20), 365 (2.30) nm; IR (neat) νmax 3285, 2925, 1651, 1540, 1516, 1396, 1253, 1026, 827, 745 cm−1; 1H NMR and 13C NMR data shown in Table 2; ESIMS m/z 812 (M+H+, 19), 834 (100); FTMS m/z [M+H]+ 812.38247 (calcd for C39H53N7O12, 812.38250)

Peptide Hydrolysis

Samples of 80 μg each of 1 and 2 were hydrolyzed with 6 N HCl over the course of 18 h at 110 °C. Each hydrolysate was evaporated to dryness under stream of air and re-dissolved in 100 μL of deionized H2O.

Advanced Marfey’s Analysis20, 21 of 1 and 2

To a 10% acetone solution (50 μL) of l-fluoro-2,4-dinitrophenyl-5-alanine amide (L-FDAA) and 1 M NaHCO3 (40 μL), were added 100 μL of the acid hydrolysate of each peptide, and the mixture was kept at 40 °C for 1 h. After cooling to room temperature, 40 μL of 2 N HCl were added and the resulting solution dried completely and then dissolved in 250 μL of MeCN. A 5 μL aliquot of each of FDAA derivative was analyzed by reversed-phase LC-UV and LC-MS in the ESI positive-ion mode. A MeCN gradient (20–50% in H2O) was used to separate the FDAA derivatives of 1 with UV detection at 340 nm on a 4.6×250 mm C18 column. The peaks were identified by co-injection with a D/L-mixture (L and DL for Cit in 2) of the amino acid derivatives. For the FDAA derivatives of 2, a MeCN (containing 0.1% formic acid) gradient (18–55% in H2O) was used on a 2.1×50 mm C18 column. The following ions were found for 1, corresponding to the Marfey’s derivatives: Tyr m/z 434 [M+H]+, Gly m/z 328 [M+H]+, Glu (for Gln) m/z 400 [M+H]+, Ile m/z 384 [M+H]+, Ser m/z 358 [M+H]+, Val m/z 370 [M+H]+. The following ions were found for 2, corresponding to the Marfey’s derivatives: Phe m/z 418 [M+H]+, Ser m/z 358 [M+H]+, Asp m/z 368 [M+H]+, Cit m/z 428 [M+H]+, Tyr m/z 434 [M+H]+. Retention times of the FDAA derivatives of 1 indicated the presence of L-Tyr, D-Glu, L-Ile, L-Ser and L-Val. Retention times of the FDAA derivatives of 2 revealed L-Phe, L Ser, L-Asp, D-Cit and L-Tyr (See Supporting Information).

Peracid Oxidation of 1

A 0.2 mg aliquot of the peptide was oxidized by m-CPBA (0.1 mg) in 0.5 mL of EtOH-CH2Cl2 (1:1) for 6 h. The reaction was quenched with an excess of DMSO, and, after 2 h, the solvent was removed and the oxidized product was subjected to acid hydrolysis and advanced Marfey’s method analysis.20, 21 L-Leu was identified in the acid hydrolysate in addition to the six amino acids found in the acid hydrolysate of the unmodified peptide; Leu m/z 384 [M+H]+.

20S Proteasome Bioassay

Evaluation of 20S proteasome inhibition was performed as reported previously.36

Cytotoxicity Bioassay

Cytotoxicity against HT-29 human cancer cell lines was conducted as reported previously.37

Supplementary Material

1_si_001

Acknowledgments

This research was supported by NIH grants R01 GM0758556 and P01 CA125066. The 900 MHz NMR spectrometer was funded by NIH grant P41 GM68944. We thank C. Pennington and C.A. Crot of the Mass Spectrometry Laboratory at the University of Illinois at Chicago for acquiring low-resolution MS and FT-MS data.

Footnotes

Supporting Information Available: 1D and 2D NMR data and results of Marfey’s analyses of 1 and 2. This material is available free of charge via the Internet at http://pubs.acs.org.

References: and Notes

  • 1.Tan LT. Phytochemistry. 2007;68:954–979. doi: 10.1016/j.phytochem.2007.01.012. [DOI] [PubMed] [Google Scholar]
  • 2.Williams PG. Trends Biotechnol. 2009;27:45–52. doi: 10.1016/j.tibtech.2008.10.005. [DOI] [PubMed] [Google Scholar]
  • 3.van Wagoner RM, Drummond AK, Wright JLL. Biogenetic diversity of cyanobacterial metabolites. In: Laskin AI, Sariaslani S, Gadd GM, editors. Advances in Applied Microbiology. Vol. 61. Elsevier; San Diego: 2007. pp. 89–217. [DOI] [PubMed] [Google Scholar]
  • 4.Welker M, von Dohren H. FEMS Microbiol Rev. 2006;30:530–563. doi: 10.1111/j.1574-6976.2006.00022.x. [DOI] [PubMed] [Google Scholar]
  • 5.Adams J. Nat Rev Cancer. 2004;4:349–360. doi: 10.1038/nrc1361. [DOI] [PubMed] [Google Scholar]
  • 6.Pickart CM. Cell. 2004;116:181–190. doi: 10.1016/s0092-8674(03)01074-2. [DOI] [PubMed] [Google Scholar]
  • 7.Furuno K, Goodman MN, Goldberg AL. J Biol Chem. 1990;265:8550–8557. [PubMed] [Google Scholar]
  • 8.Orlowski M. Biochemistry. 1990;29:10289–10297. doi: 10.1021/bi00497a001. [DOI] [PubMed] [Google Scholar]
  • 9.Holloway SL, Glotzer M, King RW, Murray AW. Cell. 1993;73:1393–1402. doi: 10.1016/0092-8674(93)90364-v. [DOI] [PubMed] [Google Scholar]
  • 10.Dorsey BD, Iqbal M, Chatterjee S, Menta E, Bernardini R, Bernareggi A, Cassara PG, D’Arasmo G, Ferretti E, De Munari S, Oliva A, Pezzoni G, Allievi C, Strepponi I, Ruggeri B, Ator MA, Williams M, Mallamo JP. J Med Chem. 2008;51:1068–1072. doi: 10.1021/jm7010589. [DOI] [PubMed] [Google Scholar]
  • 11.Furet P, Imbach P, Fuerst P, Lang M, Noorani M, Zimmermann J, Garcia-Echeverria C. Bioorg Med Chem Lett. 2002;12:1331–1334. doi: 10.1016/s0960-894x(02)00178-6. [DOI] [PubMed] [Google Scholar]
  • 12.Furet P, Imbach P, Noorani M, Koeppler J, Laumen K, Lang M, Guagnano V, Fuerst P, Roesel J, Zimmermann J, Garcia-Echeverria C. J Med Chem. 2004;47:4840–4813. doi: 10.1021/jm049660v. [DOI] [PubMed] [Google Scholar]
  • 13.Garcia-Echeverria C, Imbach P, France D, Furst P, Lang M, Noorani AM, Scholz D, Zimmermann J, Furet P. Bioorg Med Chem Lett. 2001;11:1317–1319. doi: 10.1016/s0960-894x(01)00205-0. [DOI] [PubMed] [Google Scholar]
  • 14.Iqbal M, Chatterjee S, Kauer JC, Das M, Messina P, Freed B, Biazzo W, Siman R. J Med Chem. 1995;38:2276–2277. doi: 10.1021/jm00013a002. [DOI] [PubMed] [Google Scholar]
  • 15.Delcros JG, Floc’h MB, Prigent C, Arlot-Bonnemains Y. Curr Med Chem. 2003;10:479–503. doi: 10.2174/0929867033368231. [DOI] [PubMed] [Google Scholar]
  • 16.Adams J. Cancer Cell. 2004;5:417–421. doi: 10.1016/s1535-6108(04)00120-5. [DOI] [PubMed] [Google Scholar]
  • 17.Orlowski RZ, Zeger EL. Expert Opin Investig Drugs. 2006;15:117–130. doi: 10.1517/13543784.15.2.117. [DOI] [PubMed] [Google Scholar]
  • 18.Gaillet C, Lequart C, Debeire P, Nuzillard JM. J Magn Reson. 1999;139:454–459. doi: 10.1006/jmre.1999.1808. [DOI] [PubMed] [Google Scholar]
  • 19.Bernassau JM, Nuzillard JM. J Magn Reson. 1994;103:77–81. [Google Scholar]
  • 20.Harada K-i, Fujii K, Mayumi T, Hibino Y, Suzuki M, Ikai Y, Oka H. Tetrahedron Lett. 1995;36:1515–1518. [Google Scholar]
  • 21.Marfey P. Carlsberg Res Commun. 1984;49:591–596. [Google Scholar]
  • 22.Dalisay DS, Molinski TF. J Nat Prod. 2009;72:739–744. doi: 10.1021/np900009b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Bigler P, Kummerle R, Bermel W. Magn Reson Chem. 2007;45:469–472. doi: 10.1002/mrc.1993. [DOI] [PubMed] [Google Scholar]
  • 24.Burger R, Bigler P. J Magn Reson. 1998;135:529–534. doi: 10.1006/jmre.1998.1595. [DOI] [PubMed] [Google Scholar]
  • 25.Golakoti T, Yoshida WY, Chaganty S, Moore RE. J Nat Prod. 2001;64:54–59. doi: 10.1021/np000316k. [DOI] [PubMed] [Google Scholar]
  • 26.Groll M, Koguchi Y, Huber R, Kohno J. J Mol Biol. 2001;311:543–548. doi: 10.1006/jmbi.2001.4869. [DOI] [PubMed] [Google Scholar]
  • 27.Kaiser M, Groll M, Siciliano C, Assfalg-Machleidt I, Weyher E, Kohno J, Milbradt AG, Renner C, Huber R, Moroder L. Chembiochem. 2004;5:1256–1266. doi: 10.1002/cbic.200400096. [DOI] [PubMed] [Google Scholar]
  • 28.Pitts TM, Morrow M, Kaufman SA, Tentler JJ, Eckhardt SG. Mol Cancer Ther. 2009;8:342–349. doi: 10.1158/1535-7163.MCT-08-0534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Sitachitta N, Williamson RT, Gerwick WH. J Nat Prod. 2000;63:197–200. doi: 10.1021/np990466z. [DOI] [PubMed] [Google Scholar]
  • 30.Milligan KE, Marquez BL, Williamson RT, Gerwick WH. J Nat Prod. 2000;63:1440–1443. doi: 10.1021/np000133y. [DOI] [PubMed] [Google Scholar]
  • 31.Bunyajetpong S, Yoshida WY, Sitachitta N, Kaya K. J Nat Prod. 2006;69:1539–1542. doi: 10.1021/np050485a. [DOI] [PubMed] [Google Scholar]
  • 32.Matsumoto GI, Watanuki K, Torii T. Org Geochem. 1988;13:785–90. [Google Scholar]
  • 33.Sun Q, Ishida K, Matsuda H, Murakami M. Tennen Yuki Kagobutsu Toronkai Koen Yoshishu. 1998;40:499–504. [Google Scholar]
  • 34.Linington RG, Edwards DJ, Shuman CF, McPhail KL, Matainaho T, Gerwick WH. J Nat Prod. 2007;71:22–27. doi: 10.1021/np070280x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Falch BS, Konig GM, Wright AD, Sticher O, Angerhofer CK, Pezzuto JM, Bachmann H. Planta Medica. 1995;61:321–328. doi: 10.1055/s-2006-958092. [DOI] [PubMed] [Google Scholar]
  • 36.Chlipala G, Mo S, Carcache de Blanco EJ, Ito A, Bazarek S, Orjala J. Pharm Biol. 2009;47:53–60. doi: 10.1080/13880200802415483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Seo EK, Kim NC, Mi Q, Chai H, Wall ME, Wani MC, Navarro HA, Burgess JP, Graham JG, Cabieses F, Tan GT, Farnsworth NR, Pezzuto JM, Kinghorn AD. J Nat Prod. 2001;64:1483–5. doi: 10.1021/np0103158. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001

RESOURCES