Tetrahedron 61 (2005) 883–887 New glycolipid inhibitors of Myt1 kinase Bing-Nan Zhou,a Shoubin Tang,a Randall K. Johnson,b Michael P. Mattern,b John S. Lazo,c Elizabeth R. Sharlow,c Kim Harichd and David G. I. Kingstona,* a Department of Chemistry, M/C 0212, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA b GlaxoSmithKline Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406-0939, USA c Department of Pharmacology, University of Pittsburgh, Pittsburgh, PA, USA d Department of Biochemistry, M/C 0459, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA Received 5 May 2004; revised 9 November 2004; accepted 9 November 2004 Available online 10 December 2004 Abstract—A crude extract of a marine alga showed activity against the enzyme Myt1 kinase. Bioassay-directed fractionation led to the isolation of two bioactive glycoglycerolipids. Lipid 1 was identified as sn-1,2-dipalmityl-3-(N-palmityl-6-deoxy-6-amino-a-D-glucosyl) glycerol and lipid 2 as sn-1-palmityl-2-myristyl-3-(N-stearyl-6-deoxy-6-aminoglucosyl)glycerol. Compounds 1 and 2 had IC50 values of 0.12 and 0.43 mg/mL, respectively, in the Myt1 kinase inhibitory bioassay, and were inactive against Akt and Chk1 kinases. q 2004 Elsevier Ltd. All rights reserved. 1. Introduction The enzyme Myt1 kinase, which is a Thr-14 and Tyr-15 specific cdc2 kinase, has been shown to be an important regulator of cdc2/cyclin B kinase activity. It has been reported that the inhibitory phosphorylation of cdc2 is important for the timing of entry into mitosis, and studies have shown that premature activation of cdc2 leads to mitotic catastrophe and cell death.1–3 Inhibition of Myt1 kinase is predicted to cause premature activation of cdc2, and inhibitors of this enzyme would thus be expected to kill rapidly proliferating cells and abrogate normal cell cycle checkpoints. Such inhibitors would thus be attractive for the treatment of cancer, because they could be used in conjunction with conventional chemotherapies to overcome drug resistance and enhance their cytotoxicity by abrogating cell cycle checkpoints. As a part of our systematic search for potential anti-cancer agents from plants and marine organisms,4–6 the methanol extract of an alga designated UM 2972M was found to show activity in a bioassay for inhibitors of Myt1 kinase, with an IC50 of 4 mg/mL. It was thus selected for fractionation for isolation of its bioactive constituents. The algae comprise a large group of marine organisms, and have been subjected to intensive chemical studies. Many Keywords: Marine; Myt1 kinase; Glycolipid. * Corresponding author. Tel.: C1 540 231 6570; fax: C1 540 231 3255; e-mail: dkingston@vt.edu 0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tet.2004.11.013 bioactive compounds, including the neurotoxic amino acids kainic acid and domoic acid7 and the hypocholesterolemic betain lipids8 have been isolated from them, as well as carotenoids,9 steroids,10 halogenated compounds,11,12 fatty acids,13–17 phenolic compounds,18,19 and terpenoids.20–22 2. Results and discussion The methanol extract (UM 2972 M) from an unknown algal species was subjected to partition between aqueous MeOH and organic solvents, and the aqueous MeOH fraction was then stripped of its MeOH and extracted with n-butanol to give a bioactive n-butanol fraction (IC50Z1 mg/mL). Column chromatography of the n-butanol fraction on Sephadex LH 20 with elution with CH2Cl2–MeOH (3:1), followed by repeated chromatography on RP-18 reversed phase silica gel with MeOH–H2O (9:1) gave the two bioactive compounds 1 and 2. Compounds 1 (1.05 mg, 0.025%) and 2 (1.46 mg, 0.033%) had activities of 0.12 and 0.43 mg/mL, respectively, in the Myt1 kinase bioassay. Compound 1 had a molecular weight of 967, as indicated by a pseudomolecular ion (MCNa)C at m/zZ990 in its MALDI TOF mass spectrum, and a composition of C57H109NO10, as indicated by FAB-HRMS. Three spin– spin systems were identified from its 1H NMR and COSY spectra (Table 1). The first spin system indicated the presence of a 1,2 diacylated glyceryl moiety [dH 4.51 ppm (1H, dd, JZ11.9, 3.1 Hz, Hsn-1a), 4.18 ppm (1H, dd, JZ 12.1, 7.0 Hz, Hsn1a), 5.30 ppm (1H, m, Hsn-2), 4.11 ppm 884 B.-N. Zhou et al. / Tetrahedron 61 (2005) 883–887 Table 1. Partial NMR data of lipids 1 and 2a,b Position sn-1 sn-2 sn-3 G-1 G-2 G-3 G-4 G-5 G-6 CH3– (–CH2–)n a-CH2– b-CH2– –COO– –COO– –COO– a b 1 2 dH and J dC dH and J dC 4.51 (dd, JZ12.0, 3.0) 4.18 (dd, JZ12.0, 6.9) 5.31 (m) 4.10 (dd, JZ10.8, 5.2) 3.57 (dd, JZ10.8, 6.4) 4.76 (d, JZ3.6) 3.40 (dd, JZ3.6, 9.5) 3.63 (dd, JZ9.5, 9.6) 3.08 (dd, JZ9.6, 9.2) 4.07 (ddd, JZ9.2, 9.4, 2.0) 3.34 (dd, JZ14.4, 2.0) 2.91 (dd, JZ14.4, 9.4) 0.89 (3!CH3–, t, JZ6.8) 1.0–1.5 2.33 (3!–CH2–, m) 1.59 (3!–CH2–, m) 64.3 4.50 (dd, JZ12.0, 3.0) 4.18 (dd, JZ12.0, 6.9) 5.30 (m) 4.10 (dd, JZ10.7, 5.3) 3.56 (dd, JZ10.7, 6.3) 4.75 (d, JZ3.7) 3.39 (dd, JZ3.6, 9.6) 3.62 (dd, JZ9.6, 9.7) 3.07 (dd, JZ9.7, 9.3) 4.06 (ddd, JZ9.3, 8.4, 2.0) 3.34 (dd, JZ14.3, 2.0) 2.91 (dd, JZ14.3, 8.4) 0.89 (3!CH3–, t, JZ7.0) 1.0–1.5 2.33 (3!–CH2–, m) 1.60 (3!–CH2–, m) 64.3 71.7 67.1 101.0 73.5 75.0 74.9 69.9 54.2 14.5 35.0, 35.2 26.0 174.4 175.2 175.2 71.7 67.1 100.0 73.5 75.0 74.9 68.9 54.2 14.5 35.0; 35.2 26.0 174.4 175.2 175.1 d values are in ppm and J values are in Hz. Chemical shifts were assigned by DQ COSY, HMQC, and HMBC spectra. (1H, dd, JZ11.8, 5.2 Hz, Hsn-3a), and 3.57 ppm (1H, dd, JZ 11.8, 6.4 Hz, Hsn-3b)]. The cross-peaks in the HMBC spectrum [dH/dC: 5.31 (Hsn-2)/174.4 (COO); 2.33 (a-CH2)/ 174.4 (COO); 4.51 and 4.18 (Hsn-1)/175.2 (COO); 2.33 (a-CH2)/175.2 (COO)] also indicated the presence of acyl groups on the 1 and 2 positions of a glycerol moiety. The second group of spin systems contained signals for three long chain fatty acids [dH 0.89 ppm (9H, t, JZ6.83 Hz, 3! CH3–), 1–1.5 ppm (multi –CH2–) 1.59 ppm (6H, m, 3!bCH2–), 2.33 ppm (6H, m, 3!a-CH2–). The third spin system indicated the presence of a glycosyl moiety [dH 4.76 ppm (1H, d, JZ3.8 Hz, HG-1), 3.40 ppm (1H, dd, JZ 9.5, 3.8 Hz, HG-2), 3.63 ppm (1H, d, JZ9.5, 9.1 Hz, HG-3), 3.08 ppm (1H, dd, JZ9.1, 9.7 Hz, HG-4), 4.07 ppm (1H, m, HG-5), and 2.91 (1H, dd, JZ14.3, 9.3 Hz, HG-6a), and 3.34 ppm (1H, d, JZ14.1, 2.1 Hz, HG-6b)]. The HMBC cross-peak dH 4.76 (HG-1)/dC 67.1 (Csn-3) showed that glycosylation was on the 3 position of the glycerol moiety, and that 1 was a glycoglycerolipid. Its 13C NMR spectrum supported this conclusion. The chemical shifts and the coupling constants of the G-1 to G-4 protons of the glycosyl moiety were very close to those of methyl a-D-glucoside, indicating that these hydroxyl groups were unacylated. The chemical shifts of HG-6 in the glycosyl moiety suggested the presence of an aminoacyl group at this position. The coupling constant of the anomeric proton (JZ3.8 Hz) indicated the a configuration, and since glucose has the D-configuration in most cases in nature compound 1 could be assigned as an a-D-6-desoxy-6-aminoglucoside. When 1 was hydrolyzed under various conditions (1% NaOMe in MeOH, with lipase from Mucor javanicus,23,24 or 0.5% HCl) the only ester that could be detected was methyl palmitate. The polar fraction remaining after acidic hydrolysis was reduced by reaction with NaBH4 and the resulting products acetylated and analyzed by GC–MS. Analysis of the retention times and fragmentation patterns of the two products indicated them to be glycerol triacetate and 6-acetylamino-1,2,3,4,5-penta-acetoxyhexane. This result confirmed that 1 was a glycoglycero lipid. A negative Cotton effect between lmax 200–250 nm indicated that 1 was an sn-1, sn-2 diacylglycosylglycero-lipid.25 The difference between the 13C NMR signals for the two carbonyl carbons of the two glyceryl esters (0.6 ppm) indicated that 1 was an sn-1, sn-2 diacyl ester; had it been an sn-1, sn-3 diacyl ester these signals would have been closer or even overlapped.26,27 Thus, compound 1 was assigned the structure sn-1,2-di-palmityl-3-(N-palmityl-6 0 -desoxy-6 0 amino-a-D-glucosyl)-glycerol. Compound 2 also had a molecular weight 967 and a composition of C57H109NO10, as indicated by its MALDI TOF mass spectrum and FAB-HRMS. Its 1H NMR and COSY spectra (Table 1) were very similar to those of 1, and showed the presence of a diacylglyceryl moiety [dH 4.50 ppm (1H, dd, JZ12.0, 3.0 Hz, Hsn-1a), 4.18 ppm (1H, dd, JZ12.0, 6.9 Hz, Hsn-1a), 5.30 ppm (1H, m, Hsn-2), 4.10 ppm (1H, dd, JZ10.7, 5.3 Hz, Hsn-3a), and 3.56 ppm (1H, dd, JZ10.7, 6.3 Hz, Hsn-3b)]. The HMBC cross-peaks [dH/dC: 5.30 (Hsn-2)/174.4 (–COO–); 2.33(a-CH2–)/174.4 (–COOH); 4.50 and 4.18 (Hsn-1)/175.2 (–COO–); 2.33 (a-CH2–)/175.2 (–COO–)] supported this result. Overlapped signals for three long chain fatty acids [dH 0.89 ppm (9H, t, JZ6.83 Hz, 3!CH3–), 1–1.5 ppm (multi –CH2–), 1.59 ppm (6H, m, 3!b-CH2–), 2.33 ppm (6H, m, 3!a-CH2–) could also be found in its NMR spectrum. NMR signals for a similar glycosyl moiety to that of 1 [dH 4.75 ppm (1H, d, JZ3.7 Hz, HG-1), 3.39 ppm (1H, dd, JZ9.6, 3.7 Hz, HG-2), 3.62 ppm (1H, d, JZ9.6, 9.7 Hz, HG-3), 3.07 ppm (1H, dd, JZ9.3, 9.7 Hz, HG-4), 4.06 ppm 885 B.-N. Zhou et al. / Tetrahedron 61 (2005) 883–887 (1H, m, HG-5), and 2.91 (1H, dd, JZ14.3, 8.4 Hz, HG-6a), and 3.34 ppm (1H, d, JZ14.3, 2.0 Hz, HG-6b) were also observed. All the spectroscopic data for 2 were consistent with its assignment as an a-D-6-desoxy-6amino-glucoside. Hydrolysis of 2 with 0.5% HCl followed by 1% NaOMe in methanol and then by treatment with diazomethane yielded approximately equimolar amounts of methyl myristate, methyl palmitate and methyl stearate as determined by GC– MS. The polar fraction remaining after hydrolysis was reduced with NaBH4 and acetylated, and the resulting product analyzed by GC–MS. Analysis of the retention times and fragmentation patterns of the two products indicated than to be glycerol triacetate and 6-acetylamino1,2,3,4,5-penta-acetoxyhexane. These results indicated that 2 was a glycoglycero lipid with three different acyl groups, so the locations of these groups needed to be determined. When 2 was hydrolyzed with 1% NaOMe in absolute methanol, only methyl palmitate and methyl myristate were detected by GC–MS. Since amides are hydrolyzed more slowly than esters under alkaline conditions, this result indicated that the stearyl moiety acylated the 6-amino position of glucose, and that the myristyl and palmityl groups both acylated the glyceryl moiety. When 2 was hydrolyzed with lipase from M. javanicus23 and the products treated with diazomethane, methyl stearate and methyl palmitate were both isolated. This indicates that the myristyl group is located at the more hindered sn-2 position. Compound 2 was thus assigned as sn-1-palmityl-2myristyl-3-(N-stearyl-6 0 -desoxy-6 0 -amino-a-D-glucosyl)glycerol. Three 6 0 -desoxy-6 0 -amino-glucosylglycerolipids have previously been reported in the literature,28 but the acyl groups in these compounds were different from ours. Compounds 1 and 2 are thus new natural products. As noted earlier, compounds 1 and 2 had IC50 values of 0.12G0.07 and 0.43G0.01 mg/mL, respectively, in the Myt1 kinase bioassay. The bioactivity of compounds of this type as inhibitors of Myt1 kinase is new, although diacylglycerol is well known as an activator of protein kinase C.29 The inhibition against Myt1 kinase appeared to be selective as we observed no inhibition of two other protein kinases, Chk1 and Akt, with concentrations of compounds 1 and 2 as high as 100 mg/mL. Although simple acylglycerols are not normally thought of as ‘druggable’ entities, they can serve as prototypes for compounds with inproved pharmacological properties. Thus, conformationally constrained diacylglycerol-bislactones have been investigated as diacylglycerol analogs,30 and the conversion of an O-glycoside to a C-glycoside boosted the potency of a ceramide O-glycoside by two to three orders of magnitude.31 In addition, it is worth noting that a sea alga has yielded a sulfoquinovosyldiacylglycerol which acts as a potent inhibitor of eukaryotic DNA polymerases and HIV reverse transcriptase type 1.32 3. Experimental 3.1. General experimental procedures CD and NMR spectra were obtained as previously described;33 a 9 Hz optimization was employed for the long-range coupling pathway in HMBC determinations. MALDI spectra were determined on a Kratos Kompact SEQ (Kratos Analytical, Manchester, UK) time of flight mass spectrometer, and FAB mass spectra were obtained on a JEOL JMS-HX-110 instrument. The conditions for GC–MS conditions for fatty acid methyl ester: Hewlett Packard HP5710A Gas Chromatograph with HP-5 column (30 M! 0.32 mM i.d.), oven temperature programmed from 80 to 280 8C at 8 8C/min for 1 and at 4 8C/min for 2, with He carrier gas at 12 psi. A VG7070E-HF mass spectrometer, scanned from 50 to 500 amu at 1.5 s/scan, was used for detection. Sephadex LH-20 (Sigma) was employed for gel permeation chromatography. Column chromatography was carried out on LRP-2 (RP-18) and reversed-phase TLC on MKC18F silica gel 60 A (RP-18) from Whatman. 3.2. Bioassay methods The crude extract, fractions, and pure compounds were assayed for inhibitory activity against Myt1 kinase by a modification of a previously described microtiter-based fluorescence polarization assay.34 Briefly, 10 mg recombinant glutathione-S-transferase (GST) tagged-Myt1 protein was incubated in kinase buffer (50 mM Tris–HCl, pH 7.5, 20 mM MgCl 2 and 1 mM DTT) containing 2.5 nM fluorescein-labeled Cdc2-derived peptide (containing Myt1 phosphorylation target sites Thr-14 and Tyr-15) (Molecular Devices, Sunnyvale, CA), 100 mM ATP, and compounds solubilized in dimethyl sulfoxide. The kinase reaction was incubated for 1 h at room temperature after which an anti-phospho-Cdc2 antibody (1 mg/mL) (Cell Signaling Technology, Beverly, MA) was added and incubated for an additional hour. Fluorescence polarization measurements were taken in black 96 well microtiter plates (Costar, Acton, MA) with a Wallac Victor 1420 multilabel microtiter plate reader (Perkin Elmer, Boston, MA). Duplicate samples were subjected to 25 flashes and experiments were performed 2–3 times. Dimethyl sulfoxide was used as the vehicle control; staurosporine (Sigma, St. Louis, MO) was used as the positive control and it had an IC50 value of 1.9 mM in this assay. The effects of 1 and 2 on Akt and Chk1 activity were also studied with 384 well microtiter-based IMAP (Molecular Devices, Sunnyvale, CA) fluorescence polarization assays. Briefly, compounds 1 or 2 (0.1 to 100 mg/mL) were incubated for 30–60 min at room temperature in the kinase buffer (10 mM Tris–HCl, pH 7.2, 10 mM MgCl2, 0.1% bovine serum albumin, and 0.05% NaN3) with 0.2 U/mL Akt or 0.4 U/mL Chk1, 100 nM fluorescently labeled Akt substrate or Chk1 crosstide substrate peptide, 5 mM ATP in a total volume of 20 mL. IMAP binding solution (15–60 mL) was added to each well and incubated at room temperature for 30 min. Data were collected on an Analyst GT (Molecular Devices) and analyzed using SoftMaxPro software. Triplicate experiments were performed 2–3 times. Staurosporine (Sigma-Aldrich, St. Louis, MO) was used as a 886 B.-N. Zhou et al. / Tetrahedron 61 (2005) 883–887 positive control for Akt (IC50Zw60 nM) and UCN-01 (National Cancer Institute, Bethesda, MD) was the positive control for Chk1 (IC50Zw2 mM). 3.3. Isolation of bioactive compounds The crude algal extract UM 2972M35 (4.04 g) was partitioned between 50% aq MeOH and n-hexane, and the aq MeOH layer was then partitioned with CH2Cl2. The 50% aq MeOH fraction was then evaporated in vacuo to remove MeOH, and the residual H2O fraction was extracted with BuOH. The BuOH fraction (1.56 g) was active in the Myt1 kinase bioassay (IC50Z1.0 mg/mL) and was subjected to column chromatography on Sephadex LH 20 (20 g, eluted with CH2Cl2–MeOH (3:1). A total of 45 fractions (5 mL/ each fraction) was collected and fractions 8–12 (119.0 mg), which showed the characteristic 1H NMR spectra of a glycoglycerolipid, were the most active (IC50Z0.4–1.0 mg/ mL). The active fractions 9–10 (IC50Z0.4 mg/mL) were repeatedly purified by chromatography on an RP-18 column with MeOH–H2O (9:1) as eluant. Compound 1 (1.05 mg, 0.026% based on crude extract) was isolated from fractions 13–28, and compound 2 (1.35 mg, 0.033% based on crude extract) from fraction 11. The inhibitory activities of the isolated compounds were IC50Z0.12 mg/mL for 1 and IC50Z0.43 mg/mL for 2. 3.3.1. Compound 1. MALDI TOF mass spectrum: m/z 990 (MCNa)C. HRFABMS m/z 968.8059 [MCH]C; (calcd for C57H110NO10 968.8130). [a]23 D ZC41.8 (c 0.12, MeOH). 1 H NMR and 13C NMR spectra: see Table 1. 3.3.2. Alkaline hydrolysis of 1. Compound 1 (0.1 mg) in MeOH (0.2 mL) was treated with 1% NaOMe–MeOH solution (0.8 mL) and the solution was stirred at room temperature for 5 h. After partition between n-hexane and MeOH, the n-hexane extract was subjected on GC–MS. Methyl palmitate was detected with retention time 15 0 14 00 and EIMS m/z 270 (M)C, 227, 143, 87, 74, 57, 55. Its retention time and fragmentation peaks were same as the data from a standard sample in the data base. 3.3.3. Enzymatic hydrolysis of 1. Compound 1 (0.1 mg) was mixed with lipase (1 mg) from M. javanicus (Fluka Chemie AG, CH-9471, Switzerland, EEC No. 2326199) in dioxane–H2O (1:1, 1 mL) and the reaction mixture was incubated at 38 8C for 4 h. After removal of the dioxane and H2O, the residue was dissolved in 2 mL MeOH and partitioned with n-hexane. The n-hexane extract was treated with diazomethane and analyzed by GC–MS. Methyl palmitate was detected as described above. 3.3.4. Complete hydrolysis of 1. Compound 1 (0.12 mg) was treated with 0.5% HCl at 65 8C overnight. The reaction mixture was evaporated to dryness in vacuo and the residue mixed with 1% NaOMe in MeOH and incubated at room temperature for 6 h. After partitioning with n-hexane, the n-hexane fraction was evaporated to dryness and treated with diazomethane and analyzed by GC–MS as described above. Methyl palmitate was the only ester detected. The aqueous fraction was evaporated to dryness and treated with NaBH4 in MeOH at room temperature for 4 h with stirring. After evaporation of the solvent, the reaction mixture was acetylated with Ac2O (0.5 mL) and dry pyridine (20 mL) with stirring overnight. After removal of the reagents with N2, the reaction mixture was extracted with CH2Cl2. The CH2Cl2 extract was subjected to GC–MS and glycerol triacetate (retention time: 7 0 20 00 , EIMS m/z 158 (MK60)C, 145, 116, 115, 183, 74, 73, and 61) and 6-acetylamino1,2,3,4,5-penta-acetoxyhexane (retention time: 17 0 47 00 , and EIMS m/z 433 (M)C, 374, 331, 304, 289, 259, 207, and 87) were detected. 3.3.5. Compound 2. MALDI TOF mass spectrum: m/zZ 990 (MCNa)C. HRFABMS m/z 968.8076 [MCH]C; (calcd for C57H110NO10 968.8130). [a]23 D ZC31.1 (c 0.10, MeOH). 1H NMR and 13C NMR spectrum data: see Table 1. 3.3.6. Alkaline hydrolysis of 2. Compound 2 (0.11 mg) in MeOH (0.2 mL) was treated with 1% NaOMe in MeOH (1.0 mL) and the solution was stirred at room temperature for 4 h. After extraction with n-hexane, the n-hexane extract was analyzed by GC–MS. Methyl myristate was detected with retention time 17 0 46 00 and EIMS m/z 242 (MC), 199, 143, 87, 74, and 57 and methyl palmitate with retention time 22 00 58 00 and EIMS m/z 270 (MC), 227, 185, 143, 87, 74, and 57. Their retention times and fragmentation patterns were the same as those of standard samples in the database. 3.3.7. Enzymatic hydrolysis of 2. Compound 2 (0.12 mg) was hydrolyzed with lipase from M. javanicus and the n-hexane extract treated with diazomethane and analyzed by GC–MS as previously described. Two peaks were detected. One of them was identified as methyl palmitate by its retention time (22 0 53 00 ) and EIMS (m/z 270 (MC), 239, 227, 185, 171, 143, 129, 115, 99, 87, 74, 57). The second was identified as methyl stearate by its retention time (27 0 42 00 ) and EIMS (m/z 298 (MC), 267, 255, 213, 199, 185, 157, 143, 129, 97, 87, 74, and 57). The retention times and fragmentation patterns of these compounds were the same as those of standard samples in the database. The MeOH fraction after partition was evaporated to dryness in vacuo. The residue was treated with 1% NaOMe in MeOH at room temperature for 4 h. After extraction with n-hexane, the n-hexane fraction was concentrated, treated with diazomethane, and analyzed by GC–MS. The major peak was identified as methyl myristate by its retention time of 17 0 37 00 and EIMS m/z 242 (MC), 199, 143, 87, 74, 57, and 55. 3.3.8. Complete hydrolysis of 2. Compound 2 (0.10 mg) was treated with 0.5% HCl solution (1 mL) at 65 8C overnight, the reaction mixture evaporated to dryness, and the residue incubated with 1% NaOMe in MeOH (1 mL) at room temperature for 6 h. After partition with n-hexane, the n-hexane fraction was evaporated to dryness and treated with diazomethane and analyzed by GC–MS. The first peak was identified as methyl myristate by its retention time of 17 0 20 00 and EIMS m/z 242 (MC), 199, 143, 87, 74, 57, and 55. The second peak was identified as methyl palmitate by its retention time of 22 00 55 00 and EIMS m/z 270 (MC), 239, 227, 185, 143, 87, 74, 57. The third peak was detected as methyl stearate by its retention time of 27 0 46 00 and EIMS m/z 298 (MC), 267, 255, 213, 199, 143, 87, 74, 57. The aqueous fraction was evaporated to a small volume and reduced with B.-N. Zhou et al. / Tetrahedron 61 (2005) 883–887 NaBH4 at room temperature for 4 h with stirring. After evaporation of the solvent and drying under vacuum overnight, the reaction mixture was acetylated with Ac2O (0.5 mL) and dry pyridine (20 mL) with stirring overnight. After removal of the reagents with N2, the reaction mixture was extracted with CH2Cl2. The CH2Cl2 extract was subjected to GC–MS and showed two intense peaks, which were identified as glycerol triacetate and 6-acetylamino-1,2,3,4,5-penta-acetoxyhexane as previously described. Acknowledgements This work was supported by a National Cooperative Drug Discovery Group award to the University of Virginia (U19 CA 50771, Dr. S. M. Hecht, Principal Investigator), and this support is gratefully acknowledged. We also thank Mr. Tom Glass and Mr. William R. Bebout, Sr., (Virginia Polytechnic Institute and State University) for spectroscopic support References and notes 1. Mueller, P. R.; Coleman, T. R.; Kumagai, A.; Dunphy, W. G. Science 1995, 270, 86–89. 2. Liu, F.; Stanton, J. J.; Wu, Z.; Piwnica-worms, H. Mol. Cell. Biol. 1997, 17, 571–583. 3. Lago, M. A. Myt1 kinase inhibitors. US patent No. US 6,391,894, May 21, 2002. 4. Zhou, B.-N.; Johnson, R. K.; Mattern, M. R.; Fisher, P. W.; Kingston, D. G. I. Org. Lett. 2001, 3, 4047–4049. 5. Zhou, B.-N.; Johnson, R. K.; Mattern, M. R.; Wang, X. Y.; Hecht, S. M.; Beck, H. T.; Kingston, D. G. I. J. Nat. Prod. 2000, 63, 217–221. 6. Zhou, B.-N.; Slebodnick, C.; Johnson, R. K.; Mattern, M. R.; Kingston, D. G. I. Tetrahedron 2000, 56, 5781–5784. 7. Impellizzeri, G.; Mangiafico, S.; Oriente, G.; Piattelli, M.; Sciuto, S.; Fattorusso, E.; Magno, S.; Santacroce, C.; Sica, D. Phytochemistry 1975, 14, 1549–1557. 8. Eichenberger, W.; Boschetti, A. FEBS Lett. 1978, 88, 201–204. 9. Katayama, T.; Yokoyama, H.; Chichester, C. O. Int. J. Biochem. 1970, 1, 438–444. 10. Combaut, G.; Bruneau, Y.; Codomier, L.; Teste, J. J. Nat. Prod. 1979, 42, 150–151. 11. Codomier, L.; Bruneau, Y.; Combaut, G.; Teste, J. C. R. Acad. Sci. Paris Ser. D 1977, 284, 1163–1165. 12. Fenical, W. Tetrahedron Lett. 1974, 15, 4463–4466. 887 13. Paul, V. J.; Fenical, W. Tetrahedron 1984, 40, 2913–2918. 14. Weinheimer, A. J.; Spraggins, R. L. Tetrahedron Lett. 1969, 5185–5188. 15. Radunz, A. Phytochemistry 1967, 6, 399–406. 16. Klenk, E.; Knipprath, W.; Eberhagen, D.; Koof, H. P. Hoppe Seyler’s Z. Physiol. Chem. 1963, 334, 44–59. 17. Kajiwara, T.; Kashibe, M.; Matsui, K.; Hatanaka, A. Phytochemistry 1991, 30, 193–195. 18. Glombitza, K. W.; Zieprath, G. Planta Med. 1989, 55, 171–175. 19. Li, S. M.; l Glombitza, K. W. Bot. Mar. 1991, 34, 455–457. 20. Amico, V.; Oriente, G.; Piattelli, M.; Tringali, C.; Fattorusso, E.; Magno, S.; Mayol, L. J. Chem Soc., Chem. Commun. 1976, 1024–1025. 21. Enoki, N.; Ishida, R.; Urano, S.; Ochi, M.; Tokoroyama, T.; Matsumoto, T. Chem. Lett. 1982, 1837–1840. 22. Wright, A. D.; König, G. M.; Sticher, O. Tetrahedron 1990, 46, 3851–3858. 23. Kobayashi, M.; Hayashi, K.; Kawazoe, K.; Kitagawa, I. Chem. Pharm. Bull. 1992, 40, 1404–1410. 24. Murakami, N.; Morimoto, T.; Imamura, H.; Nagatsu, A.; Sakakibara, J. Tetrahedron 1994, 50, 1993–2002. 25. Michelsen, P. Chem. Scr. 1985, 25, 217–218. 26. Diehl, B. W. K.; Herling, H.; Riedl, I.; Heinz, E. Chem. Phys. Lipids 1995, 77, 147–153. 27. Máñez, S.; Recio, M. d. C.; Gil, I.; Gómez, C.; Giner, R.-M.; Waterman, P. G.; Rı́os, J.-L. J. Nat. Prod. 1999, 62, 601–604. 28. Dai, J. Q.; Zhu, Q. X.; Zhao, C. Y.; Yang, L.; Li, Y. Phytochemistry 2001, 58, 1305–1309. 29. Newton, A. C. Curr. Opin. Cell. Biol. 1997, 9, 161–167. 30. Kang, J.-H.; Kim, S. Y.; Lee, J.; Marquez, V. E.; Lewin, N. E.; Pearce, L. V.; Blumberg, P. M. J. Med. Chem. 2004, 47, 4000–4007. 31. Yang, G.; Schmieg, J.; Tsuji, M.; Franck, R. W. Angew. Chem., Int. Ed. 2004, 43, 3818–3822. 32. Ohta, K.; Mizushina, Y.; Hirata, N.; Takemura, M.; Sugawara, F.; Matsukage, A.; Yoshida, S.; Sakaguchi, K. Chem. Pharm. Bull. 1998, 46, 684–686. 33. Zhou, B.-N.; Mattern, M. R.; Johnson, R. K.; Kingston, D. G. I. Tetrahedron 2001, 57, 9549–9554. 34. Kristiansdottir, K.; Rudolph, J. Anal. Biochem. 2003, 316, 41–49. 35. The algal material was provided by GlaxoSmithKline Pharmaceuticals, but regrettably its taxonomy was not fully determined and cannot now be ascertained. Glyceroglycolipids are distributed among the red, brown, and green algae,36 and so chemotaxonomy cannot be applied to the determination of the type of algal material that was extracted. 36. Khotimchenko, S. V. Chem. Nat. Comp. 2002, 38, 223–229.