Bioconjugate Chem. 2006, 17, 261−266
261
Synthesis and in Vitro Evaluation of Biotinylated RG108: A High Affinity Compound for Studying Binding Interactions with Human DNA Methyltransferases Esther Schirrmacher,† Carmen Beck,‡ Bodo Brueckner,† Frank Schmitges,† Pawel Siedlecki,§ Peter Bartenstein,# Frank Lyko,† and Ralf Schirrmacher#,* Division of Epigenetics, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 580, 69120 Heidelberg, Germany, Division of Nuclear Medicine, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Institute of Biochemistry and Biophysics, Polish Academy of Science, Pawinskiego 5a, 02-106 Warsaw, Poland, and Department of Nuclear Medicine, Johannes Gutenberg University Mainz, Langenbeckstrasse 1, 55131 Mainz, Germany. Received October 11, 2005; Revised Manuscript Received January 3, 2006
Small-molecule inhibitors of DNA methyltransferases such as RG108 represent promising candidates for cancer drug development. We report the synthesis and in vitro analysis of a biotinylated RG108 conjugate, 2-(1,3dioxo-1,3-dihydro-isoindol-2-yl)-3-(5-{3-[5-(2-oxo-hexahydro-thieno[3,4-d]imidazol-4-yl)pentanoylamino]propoxy}-1H-indol-3-yl)propionic acid (bio-RG108), for the evaluation of interactions with DNA methyltransferase enzymes. The structural design of the chemically modified inhibitor was aided by molecular modeling, which suggested the possibility for extensive chemical modifications at the 5-position of the tryptophan moiety in RG108. The inhibitory activity of the corresponding derivative was confirmed in a cell-free biochemical assay, where bio-RG108 showed an undiminished inhibition of DNA methyltransferase activity (IC50 ) 40 nM). Bio-RG108 therefore represents a suitable bioconjugate for the elucidation of inhibitory mechanisms and for the affinity purification of RG108-associated proteins.
INTRODUCTION The elucidation of protein interactions is a fundamental aspect in our understanding of biological processes. One approach to discover the identity of interacting proteins is the use of chemical cross-linking, a method capable of irreversibly capturing proteins by covalently fixing a biotin affinity handle to the compound interacting with the protein. The binding of biotin to (strept)avidin has therefore found widespread application in life sciences, which is primarily due to its extraordinarily strong noncovalent interaction (1-5) with a binding constant of 1.3 × 10-15 M. Bioanalytical applications derived from biotin/ (strept)avidin interaction have been used in enzymatic reactions, radiolabeling techniques, and fluorescence assays (6-8). DNA methyltransferases catalyze the covalent addition of methyl groups to the carbon-5 position of cytosine residues in the context of genomic DNA (9). The resulting DNA methylation patterns play an important role in the regulation of gene expression and local hypermethylation has been frequently found to be associated with gene silencing (10). One of the most prominent examples for this phenomenon is the hypermethylation of tumor suppressor genes during tumorigenesis (11), which has established DNA methyltransferases as promising targets for oncology drug development (12). The first DNA methyltransferase inhibitor 5-azacytidine has been synthesized 40 years ago (13), has shown clinical ef* To whom correspondence should be addressed. Tel: +496131176736; fax: +496131172386; E-mail: schirrmacher@ nuklear.klinik.uni-mainz.de. † Deutsches Krebsforschungszentrum. ‡ University of Heidelberg. § Polish Academy of Science. # Johannes Gutenberg University Mainz.
fectiveness in phase III trials (14), and has recently gained Food and Drug Administration approval as an anticancer drug. 5-Azacytidine and related compounds inhibit DNA methyltransferases by covalent enzyme trapping, which has been associated with their significant toxicity (15). This characteristic has complicated the clinical application of azanucleoside inhibitors and invigorated the search for novel compounds. More recent analyses have also suggested that noncovalent “active site blockers” of DNA methyltransferases might be characterized by higher specificity and less toxicity (16). This is exemplified by the molecular characteristics of RG108, the first rationally designed small-molecule inhibitor of DNA methyltransferases (17). More specifically, RG108 has been shown to be welltolerated by human cancer cell lines and to inhibit DNA methyltransferases in a noncovalent manner (17). However, the molecular interactions between RG108 and the various DNA methyltransferases (and their associated factors) could not be demonstrated so far. A biotin-conjugated derivative of RG108 displaying the same binding characteristics as the original nonconjugated molecule could provide an opportunity to study the interactions between RG108 and its cellular targets in great detail. In addition, such a bioconjugate could also represent a valuable tool for the biochemical purification of DNA methyltransferase-associated protein factors.
EXPERIMENTAL SECTION Chemicals and Reagents. Biotin, HBTU, and biotin nitrophenyl ester were purchased from Novabiochem (Merck). 5-Hydroxy-DL-tryptophan, 4-nitrobenzylbromide, triphenylphosphine, tert-butyl-3-hydroxypropyl carbamate, diethyl azodicarboxylate, 3-hydroxypropylcarbamic acid allyl ester, tetrakis(triphenylphosphine)palladium(0), 3-aminopropanol, 1,3-dimethoxybenzene, 5,5-dimethyl-1,3-cyclohexadione, and tetraeth-
10.1021/bc050300b CCC: $33.50 © 2006 American Chemical Society Published on Web 02/11/2006
262 Bioconjugate Chem., Vol. 17, No. 2, 2006
ylammonium fluoride from Aldrich were used as received. Cesium carbonate was purchased from Alfa Aesar. N-Ethyldiisopropylamine, acetonitrile, DMF, methylene chloride, methanol, ethanol, ethyl acetate, cyclohexane, n-hexane, toluene, and chloroform were obtained from Merck and were used without further purification. Instruments. 1H NMR and 13C NMR were recorded on a 400-MHz FT-NMR spectrometer (DRX 400, Bruker Analytik GmbH). FD-mass spectra and ESI mass spectra were obtained using a MAT90-spectrometer (Finnigan) and Navigator (ThermoQuest), respectively. 3-(5-Hydroxy-1H-indol-3-yl)-2-(1,3-dioxoisoindolin-2-yl)propanoic Acid (2). 5-Hydroxy-DL-tryptophan (1) (440.4 mg, 2 mmol) and sodium carbonate (1.38 g, 10 mmol) were dissolved in water (15 mL). Phthalic acid mono succinimidyl ester (554 mg, 2 mmol) dissolved in acetonitrile (25 mL) was added, and the reaction mixture was stirred for 4 h until thinlayer chromatography showed the absence of starting material. The mixture was diluted with ethyl acetate (100 mL), and 1 N hydrochloric acid was added until pH reached 2. The organic phase was separated and washed twice with 1 N hydrochloric acid (50 mL) and water, respectively (50 mL). The organic phase was dried over sodium sulfate and evaporated to yield 660 mg (95%) of compound 2. 1H NMR: 13.10 (sbr, 1H); 10.42 (s, 1H); 8.56 (s, 1H); 7.82 (s, 4H); 7.15 (d, 1H, J ) 8.85); 6.84 (d, 1H, J ) 2.04 Hz); 6.79 (d, 1H, J ) 2.04 Hz); 6.52 (dd, 1H, J ) 2.04 Hz, J ) 8.52 Hz); 5.1 (m, 1H); 3.5 (m, 2H). 13C NMR: 171.25; 167.82; 150.52; 135.12; 131.56; 130.97; 128.17; 124.01; 123.85; 112.05; 111.75; 109.20; 102.04; 52.98; 24.53. FD mass spectroscopy: 350.2 ([M]+, 100%). Elem. Anal. (%) calcd: C (65.14), H (4.03), N (8.06). Found: C (65.38), H (4.17), N (7.98). 4-Nitrobenzyl 3-(5-Hydroxy-1H-indol-3-yl)-2-(1,3-dioxoisoindolin-2-yl)propanoate (3). 2 (0.5 g, 1.4 mmol) was dissolved in ethanol (6.9 mL) and water (0.69 mL). Cesium carbonate solution (20%, 0.95 mL) was added, the mixture was stirred for 5 min at room temperature, and the solvents were removed in vacuo. For azeotropic drying, ethanol (3 mL) was added and the solvent again removed in vacuo. The dry residue was dissolved in dimethylformamide (6.9 mL), and 4-nitrobenzyl bromide (400 mg, 1.8 mmol) was added. The reaction mixture was stirred for 6 h. After filtration, the solvent was removed and the residue was purified via column chromatography (ethyl acetate/cyclohexane 50:50) to obtain 415 mg (60%) of 3. 1H NMR (DMSO-d6): 10.5 (s, 1H); 8.5 (s, 1H); 8.2 (d, 2H, J ) 8.86 Hz); 7.8 (s, 4H); 7.6 (d, 2H, J ) 8.86 Hz); 7.1 (d, 1H, J ) 8.52 Hz); 6.9 (d, 1H, J ) 2.38 Hz); 6.8 (d, 1H, J ) 2.38 Hz); 6.5 (dd, 1H, J ) 2.04 Hz, J ) 8.52 Hz); 5.4 (s, 2H); 5.3 (m, 1H); 3.5 (m, 2H). 13C NMR (DMSO-d6): 168.98;167.31; 150.63; 147.40; 143.72; 135.21; 131.02; 130.84; 128.74; 127.89; 124.25; 123.76; 112.02; 111.71; 108.76; 108.36; 102.19; 65.83; 52.67; 39.79; 39.58; 24.53; 24.41; 23.88. FD mass spectroscopy: 485.1 ([M]+, 100%). Elem. anal.(%) calcd: C (64.33), H (3.95), N (8.66). Found: C (64.25), H (4.00), N (8.56). 3-[5-(3-tert-Butoxycarbonylaminopropoxy)-1H-indol-3-yl]2-(1,3-dioxoisoindolin-2-yl)propionic Acid 4-Nitrobenzyl Ester (5). 3 (300 mg, 0.62 mmol) was dissolved in methylene chloride (10.8 mL), and triphenylphosphine (0.243 g, 0.93 mmol) and tert-butyl-3-hydroxypropylcarbamate (0.215 g, 1.2 mmol) were added. A solution of diethyl azodicarboxylate (0.145 mL) in methylene chloride (4 mL) was added dropwise over a period of 1h. The reaction mixture was stirred for 2 h at room temperature. After evaporation of the solvent under reduced pressure and workup with column chromatography (ethyl acetate/n-hexane 3.5:4), 258 mg (65%) of 5 was obtained. 1H NMR (DMSO-d ): 10.5 (s, 1H); 8.2 (d, 2H, J ) 8.51 Hz); 6 7.82 (s, 4H); 7.6 (d, 2H, J ) 8.51 Hz); 7.15 (d, 2H, J ) 8.85
Schirrmacher et al.
Hz); 7.05 (d, 1H, J ) 2.04 Hz); 6.95 (d, 1H, J ) 2.04); 6.87 (m, 1H); 6.63 (dd, 1H, J ) 2.04 Hz, J ) 8.85 Hz); 5.37 (s, 2H); 5.35 (t, 1H, J ) 5.45); 3.8 (m, 2H), 3.48 (m, 2H); 3.15 (m, 2H); 1.8 (m, 2H); 1.4 (s, 9H). 13C NMR (DMSO-d6): 168.96; 167.25; 155.91; 152.54; 147.39; 143.66; 135.20; 131.46; 131.01; 128.66; 127.51; 124.61; 123.76; 112.26; 111.90; 109.09; 101.18; 65.77; 58.78; 52.70; 37.45; 33.02; 29.72; 28.52; 24.52. FD mass spectroscopy: 642.0 ([M]+, 100%). Elem. anal. (%) calcd: C (63.54), H (5.33), N (8.72). Found: C (63.52), H (5.22), N (8.78). 3-[5-(3-Alloxycarbonylaminopropoxy)-1H-indol-3-yl]-2(1,3-dioxoisoindolin-2-yl)propionic Acid 4-Nitrobenzyl Ester (6). 3 (300 mg, 0.62 mmol) was dissolved in methylene chloride (10.8 mL), and triphenylphosphine (0.243 g, 0.93 mmol) and 3-hydroxypropylcarbamic acid allyl ester (0.191 g, 1.2 mmol) were added. A solution of diethyl azodicarboxylate (0.145 mL) in methylene chloride (4 mL) was added dropwise over a period of 1 h. The reaction mixture was stirred for 2 h at room temperature. After evaporation of the solvent under reduced pressure and workup with column chromatography (ethyl acetate/n-hexane 3.5:4), 250 mg (65%) of 6 was obtained. 1H NMR (DMSO-d6): 10.60 (s, 1H); 8.98 (s, 1H); 8.20 (d, 2H, J ) 8.85 Hz); 7.82 (s, 4H); 7.60 (d, 2H, J ) 8.85 Hz); 7.13 (d, 2H, J ) 8.85 Hz); 7.02 (d, 1H, J ) 2.38 Hz); 6.97 (d, 1H, J ) 2.38 Hz); 6.63 (dd, 1H, J ) 2.04 Hz, J ) 8.51 Hz); 5.85-5.96 (m, 1H); 5.38 (s, 2H); 5.35 (t, 1H, J ) 5.45 Hz); 5.25-5.30 (m, 1H); 5.15-5.18 (m, 1H); 4.48 (d, 2H, J ) 5.45 Hz); 3.763.90 (m, 2H); 3.49-3.6 (m, 2H); 3.13-3.18 (m, 2H); 1.801.86 (m, 2H). 13C NMR (DMSO-d6): 168.97; 167.26; 156.82; 156.24; 152.50; 147.39; 143.66; 135.21; 134.11; 131.48; 131.01; 128.66; 127.52; 123.70; 117.14; 112.26; 111.91; 109.11; 101.24; 65.77; 64.45; 60.71; 52.72; 37.79; 29.63; 24.51. FD mass spectroscopy: 626.2 ([M]+, 100%). Elem. anal. (%) calcd: C (63.25), H (4.83), N (8.94). Found: C (63.28), H (4.85), N (8.92). 2-(1,3-Dioxoisoindolin-2-yl)-3-(5-{3-[5-(2-oxohexahydrothieno[3,4-d]imidazol-4-yl)pentanoylamino]propoxy}-1H-indol-3-yl)propionic Acid 4-Nitrobenzyl Ester (7) (via Bocdeprotection). 5 (1 g, 1.5 mmol) was dissolved in trifluoroacetic acid (40 mL) containing 1,3-dimethoxybenzene (0.55 mL). After stirring for 30 min, the trifluoroacetic acid was removed under reduced pressure. Last traces were coevaporated with 2 mL of toluene. The residue was dissolved in 10 mL of dimethylformamide, and 548 mg (1.5 mmol) biotin nitrophenyl ester was added. After stirring for 10 h, the solvent was removed under vacuum and the residue was purified via column chromatography (chloroform/methanol 9:1) to yield 179 mg (15%) of 7. 2-(1,3-Dioxoisoindolin-2-yl)-3-(5-{3-[5-(2-oxohexahydrothieno[3,4-d]imidazol-4-yl)pentanoylamino]propoxy}-1H-indol-3-yl)propionic Acid 4-Nitrobenzyl Ester (7) (via Allocdeprotection). 6 (430 mg, 0.68 mmol) was dissolved in tetrahydrofuran (5.3 mL) under argon atmosphere. Tetrakis(triphenylphosphine)palladium(0) (77 mg, 0.066 mmol) and 5,5dimethyl-1,3-cyclohexadione (0.71 g, 5 mmol) were added, and the reaction mixture was stirred for 30 min at room temperature. The solvent was removed under reduced pressure, and the residue was extracted with diethyl ether (40 mL). The slurry was filtrated and extracted four times with 20 mL of 0.5 N hydrochloric acid. The aqueous phase was adjusted to pH 10 with sodium carbonate and extracted with ethyl acetate (4 × 50 mL). After removal of the solvent, the residue was dissolved in dimethylformamide (5 mL) and biotin nitrophenyl ester (274 mg, 0.75 mmol) was added. After stirring for 10 h, the solvent was removed under vacuum and the residue was purified via column chromatography (chloroform/methanol 9:1) to yield 158 mg (30%) of compound 7. 1H NMR (DMSO-d6): 10.60 (s, 1H); 8.21 (d, 2H, J ) 8.85 Hz); 7.83 (s, 4H); 7.60 (d, 2H, J ) 8.85
Communications
Hz); 7.13 (d, 1H; J ) 8.85 Hz); 7.01 (d, 1H, J ) 2.38 Hz); 6.95 (d, 1H, J ) 2.38 Hz); 6.63 (dd, 1H, J ) 2.38 Hz, J ) 8.51 Hz); 6.61 (s, 1H); 6.34 (s, 1H); 5.38 (s, 2H); 5.35 (t, 1H, J ) 5.10 Hz); 4.30-4.25 (m, 1H); 4.11-4.08 (m, 1H); 3.92-3.76 (m, 2H); 3.61-3.48 (m, 2H); 3.22-3.17 (m, 2H); 3.09-3.04 (m, 1H); 2.81-2.76 (m, 1H); 2.56 (d, 1H, J ) 12.60 Hz); 2.102.06 (m, 2H); 1.84-1.78 (m, 2H); 1.58-1.41 (m, 4H); 1.361.23 (m, 2H). 13C NMR (DMSO-d6): 172.27; 168.97; 167.26; 162.94; 155.65; 152.49; 147.39; 145.25; 143.66; 135.23; 131.47; 131.01; 128.66; 127.54; 125.53; 123.78; 123.46; 112.26; 111.86; 109.08; 101.30; 65.96; 65.77; 61.28; 59.43; 55.67; 52.69; 35.93; 35.49; 33.45; 29.44; 28.48; 28.29; 25.58; 24.48. FD mass spectroscopy: 768.5 ([M]+, 100%). Elem. anal. (%) calcd: C (60.93), H (5.24), N (10.93). Found: C (61.10), H (5.11), N (11.61). The purity of compound 7 was additionally confirmed by HPLC: column: LiChrosorb RP select B 250 × 4 mm, solvent: 0 min: 70/30 water/acetonitrile; 30 min: 100% acetonitrile (flow: 0.7 mL/min, tR (7): 18.1 min) 2-(1,3-Dioxoisoindolin-2-yl)-3-(5-{3-[5-(2-oxohexahydrothieno[3,4-d]imidazol-4-yl)pentanoylamino]propoxy}-1H-indol-3-yl)propionic Acid (8) (bio-RG108). 7 (65 mg, 0.084 mmol) was dissolved in dimethylformamide (2.3 mL). Tetraethylammonium fluoride (31.5 mg, 0.21 mmol) was added, and the reaction mixture was stirred for 30 min. The solvent was removed under vacuum, and the residue was taken up in ethyl acetate (3 mL) and washed with water (twice with 1.5 mL). The solvent was removed under reduced pressure, and 10 mL of a mixture of acetonitrile/diethyl ether (50:50) was added. The slurry was ultrasonicated for 5 min, and the resulting precipitate was filtered and washed with diethyl ether. The precipitate was dissolved in ethanol (2 mL), and 5 mL water was added. The resulting mixture was lyophilized overnight to yield compound 37 mg (70%) of compound 8. 1H NMR (DMSO-d6): 10.56 (s, 1H); 7.86 (t, 1H, J ) 6.80 Hz); 7.82 (s, 4H); 7.12 (d, 1H, J ) 8.85 Hz); 6.98 (d, 1H, J ) 2.38 Hz); 6.92 (d, 1H, J ) 2.38 Hz); 6.65 (dd, 1H, J ) 8.85 Hz, J ) 2.04 Hz); 6.41 (s, 1H); 6.34 (s, 1H); 5.08 (t, 1H, J ) 8.50 Hz); 4.26-4.33 (m, 1H); 3.74-3.92 (m, 2H); 3.5 (d, 2H, J ) 6.8 Hz); 3.14-3.23 (m, 2H); 2.83-2.95 (m, 1H); 2.75-2.82 (m, 1H); 2.54 (d, 1H, J ) 12.60 Hz); 2.04-2.13 (m, 2H); 1.17-1.27 (m, 2H); 1.26-1.66 (m, 6H). 13C NMR: 172.27; 170.62; 167.45; 162.95; 161.46; 152.47; 135.10; 131.46; 131.16; 130.97; 127.53; 124.01; 123.58; 112.26; 109.75; 101.10; 70.20; 65.93; 61.28; 59.44; 56.05; 55.07; 52.89; 35,91; 35.49; 29.43; 28.30; 25.58; 24.38. ESI mass spectroscopy: 656.2 ([M + Na]+, 100%). The purity of compound 8 was confirmed using two independent HPLC methods: First method: column: LiChrosorb RP select B 250 × 4 mm; solvent: 0 min: 100% water; 30 min: 100% acetonitrile; flow: 0.7 mL/min; tR (8): 20.1 min, Second method: column: LiChrosorb RP select B 250 × 4 mm; solvent: 50:50 methanol/water; flow: 0.5 mL/min; tR (8): 7.5 min 5-(Hexahydro-2-oxo-1H-thieno[3,4-d]imidazol-4-yl)-N-(3hydroxypropyl)pentamide (9). Biotin (1 g, 4.09 mmol) and HBTU (1.36 g, 3.6 mmol) were suspended in DMF (50 mL), and DIPEA (0.70 g, 5.4 mmol) was added with stirring. The clear solution was transferred into a dropping funnel and added dropwise to a solution of 3-aminopropanol (0.52 g, 7 mmol) in DMF (75 mL). The solvent was removed in vacuo, and the crude product was dissolved in a minimum amount of methanol (4-7 mL). Diethyl ether (20 mL) was added, and the solution was kept at -20 °C for 12 h. The white crystalline precipitate was filtered by suction and washed with cold diethyl ether to yield 1.1 g (89%) of the product. 1H NMR (DMSO-d6): 7.73 (t, 1H, J ) 5.45 Hz); 6.43 (s, 1H); 6.36 (s, 1H); 4.40 (sbr, 1H); 4.31 (dd, 1H, J ) 5.10 Hz, J ) 7.49 Hz); 4.12 (dt, 1H, J ) 1.70 Hz, J ) 6.13 Hz); 3.38 (t, 2H, J ) 6.13 Hz); 3.03-3.13 (m, 3H);
Bioconjugate Chem., Vol. 17, No. 2, 2006 263
Figure 1. Molecular modeling and structure of 5-substituted RG108. A, RG108 docked into the active site of human DNMT1. Green and orange amino acids form the entry site to the catalytic pocket. The 5-position of the tryptophan moiety (arrow) appeared to protrude from the catalytic pocket and was therefore considered to be particularly amenable to functionally neutral chemical modification. B, chemical structure of original RG108 and 5-substituted RG108.
2.80 (dd, 1H, J ) 5.10 Hz, J ) 12.60 Hz); 2.57 (d, 1H, J ) 12.60 Hz); 2.04 (t, 1H, J ) 7.49 Hz); 1.40-1.57 (m, 6H); 1.201.36 (m, 2H). 13C NMR (DMSO-d6): 172.28; 162.99; 61.30; 59.46; 58.67; 55.66; 35.86; 35.45; 32.74; 28.46; 28.27; 25.57. FD mass spectroscopy: 302.2 ([M + H]+, 100%). Elem. anal. (%) calcd: C (51.80); H (7.69); N (13.94). Found: C (52.11); H (7.56); N (14.61) In Vitro Methylation Assay. A 798 bp promoter fragment (-423/+375 relative to the initiation codon) from the human p16Ink4a gene was used as unmethylated substrate DNA. Four hundred nanograms of substrate DNA, 2 units of M.SssI methylase (New England Biolabs, Frankfurt, Germany), and the cofactor S-adenosyl-L-methionine (final concentration: 80 µM; New England Biolabs) were added to a final reaction volume of 50 µL. Test compounds were added to final concentrations of 10, 30, 100, 300, and 500 µM, and the reactions were incubated for 2 h at 37 °C. After incubation the reactions were purified using the QIAquick PCR Purification kit (Qiagen, Hilden, Germany). Three hundred nanograms of purified DNA was digested for 3 h at 60 °C with 30 units of BstUI (New England Biolabs) and analyzed on 3% Tris-borate EDTA agarose gels.
RESULTS AND DISCUSSION Previous results have strongly suggested that RG108 binds deep in the active site of DNA methyltransferases (16), which rendered the identification of chemically modifyable positions
264 Bioconjugate Chem., Vol. 17, No. 2, 2006
Schirrmacher et al.
Scheme 1
critically important. We therefore used molecular modeling and a previously established three-dimensional model of the human DNMT1 catalytic domain (18) for an analysis of the molecular interactions between RG108 and the DNA methyltransferase active site. Docking of RG108 revealed that the 5-position of the tryptophan moiety protruded from the active site pocket, suggesting that modifications at this position were unlikely to interfere with the various interactions between enzyme and inhibitor (Figure 1). This analysis also indicated that more extensive modifications could be accommodated by the addition of an aminoalkyl spacer (data not shown). Last, the synthesis of 5-substituted RG108 derivatives also appeared favorable in light of the fact that 5-hydroxytryptophan could be used as a convenient starting point for chemical syntheses which are amenable to various chemical alterations, such as Mitsunobu coupling of alcohols. Biotinylated RG108 (bio-RG108) was synthesized starting from racemic 5-hydroxy-DL-tryptophan 1 (Scheme 1) which was phthaloylated using methyl 2-[(succinimido)oxycarbonyl]benzoate (19) to yield 5-hydroxy-RG108 (2) in almost quantitative yields. Racemic mixtures of RG108 have previously been shown to be effective in various assays (16), and therefore racemic
compound 1 was used. The phthaloylation of 1 was slightly modified because the original procedure for the synthesis of RG108 (17) gave yields of 50-60% only. Potassium carbonate was used instead of sodium carbonate for the deprotonation of 1 and 1 N HCl instead of 2 N HCl for the final workup yielding 2 in high yields and high chemical purity, making further purification unnecessary. For further derivatization of 2 at the 5-position of tryptophan, it is essential to protect the carboxy moiety by esterification. Lescrinier et al. reported the successful synthesis of N-Treoc-protected 5-hydroxytryptophan p-nitrobenzyl ester for incorporation of 5-hydroxytryptophan in oligopeptides via conversion of the carboxylic acid into its cesium salt by titration to pH 7 with Cs2CO3 (20). This method was applied to the phthaloylated amino acid 2, and the resulting Cs-salt was subsequently reacted with p-NO2-benzyl bromide in DMF giving compound 3 in 65% yield after column chromatography. Our first attempt to conjugate biotin to 3 was via Mitsunobu coupling of biotinol, which was obtained by reduction of biotinSU ester with NaBH4 (21). Unfortunately, the coupling reaction was unsuccessful due to the limited solubility of biotinol in solvents commonly applied in Mitsunobu reactions (CH2Cl2, CHCl3, THF, benzene, ethyl acetate) and even in DMF or
Communications
Bioconjugate Chem., Vol. 17, No. 2, 2006 265
Figure 2. Inhibition of purified recombinant DNA methyltransferase activity by bio-RG108. Incubation of an unmethylated DNA fragment with purified Sssl methylase results in the generation of a methylated fragment (M) that is protected against restriction cleavage by BstUI. Inhibition of DNMT activity generates unmethylated (U) fragments that are products of BstUI restriction. Numbers indicate inhibitor concentrations (in µM). An biotin derivative lacking the RG108 moiety (9) was analyzed in control experiments and did not show any detectable inhibitory activity.
HMPA. One further attempt with N-biotinylated 3-aminopropan1-ol 9 was also found to be unsuitable, but N-Boc- and N-Allocprotected 3-aminopropan-1-ol in CH2Cl2 could be coupled in moderate yields of 55-60% to yield compounds 5 and 6. The deprotection of 5 with TFA at 0 °C and the subsequent workup to isolate the primary amine was unsuccessful due to decomposition on silica gel. We therefore decided to use the crude amine for direct reaction with commercially available biotinNp ester in DMF and were able to isolate p-NO2-benzyl ester protected 7 in 15% yield. Because of these unsatisfying results, we synthesized the Alloc-protected compound 6 which was deprotected with Pd(0) and dimedon in THF. The instability of the amine upon contact with silica gel was confirmed, and so the crude amine was further reacted with biotin-Np ester to yield p-NO2-benzyl ester protected 7 in 30% yield. Despite the presence of acidic dimedon, the main product (65%) was an enamine built from excess dimedon and the primary amine. To obtain the final bio-RG108 8, the p-NO2-benzyl ester was deprotected with 2.5 equiv of tetraethylammonium fluoride (TEAF) in DMF in 70% yield. The yields dramatically decreased when using more than 2.5 equiv which was proved by HPLC monitoring of the crude reaction mixture using 2-10 equiv of TEAF. Starting from commercially available 5-hydroxytryptophan and biotin nitrophenyl ester, final compound bio-RG108 could be synthesized in an overall chemical yield of 4% (via compound 5) or 8% (via compound 6). The purity of all compounds was determined by 1H and 13C NMR, mass spectroscopy, elemental analysis, and/or HPLC. Due to the major structural alterations, we needed to analyze the inhibitory activity of bio-RG108 8. To this end, we used a cell-free in vitro assay with purified recombinant CpG methylase (17). The results showed that bio-RG108 was able to efficiently inhibit DNA methylation (Figure 2). After normalization to 1 nM enzyme concentration, the IC50 concentration of bio-RG108 was determined to be 40 nM. This is similar to the IC50 concentration of unmodified RG108 (115 nM, cf. ref 17) with the improved potency likely due to the better solubility of bioRG108. Importantly, no inhibitory effect could be observed for a biotin derivative that lacks the RG108 moiety (Figure 2, biotin control), which further confirmed the specificity of bio-RG108. In conclusion, we prepared a biotinylated derivative of the novel mechanism-independent DNA methyltransferase inhibitor RG108 for the evaluation of binding interactions using a biotin/ (strept)avidin binding assay. Despite major derivatization, the
novel compound bio-RG108 exhibited a high inhibitory activity with purified recombinant DNA methyltransferase, thus confirming the results from our molecular modeling approach. BioRG108 therefore represents a valuable probe for the selective immobilization of human DNA methyltransferases and further investigation of proteins interacting with RG108. Additionally, proving the 5-position of RG108 to be most suitable for chemical alterations provides an opportunity to synthesize various bioconjugates such as carbohydrate and peptide conjugates for specific tumor-targeting. A great variety of suitable linkers, e.g. hydroxyalkylcarboxylic acid esters, N-protected aminoalkyl alcohols, etc., can be conveniently attached to the 5-position of 5-hydroxy-RG108, applying mild Mitsunobu conditions and used for further derivatization.
ACKNOWLEDGMENT The authors would like to thank B. Mathiasch, N. Haunold, and A. Vierengel for the recording of NMR and mass spectra. This work was supported by a grant from Cancer Research Technologies (to F.L).
LITERATURE CITED (1) Wilchek, M., and Bayer, E. A. (1990) Methods of Enzymology, Vol. 184, Academic Press, San Diego. (2) Weizmann, Y., Patolsky, F., Katz, E., and Willner, I. (2003) Amplified DNA Sensing and Immunosensing by the Rotation of Functional Sensing and Immunosensing by Rotation of Functional Magnetic Particles. J. Am. Chem. Soc. 125, 452-3454. (3) Rivera, V. R., Merrill, G. A., White, J. A., and Poli, M. A. (2003) An Enzymatic Electrochemiluminescence Assay for the Lethal Factor of Anthrax. Anal. Biochem. 321, 125-130. (4) Richter, J., Adler, M., and Niemeyer, C. M. (2003) Monte Carlo Simulation of the Assembly of bis-Biotinylated DNA and Streptavidin. ChemPhysChem. 4, 79-83. (5) Caswell, K. K., Wilson, J. N., Bunz, U. H. F., and Murphy, C. J. (2003) Preferential End-to-End Assembly of Gold Nanorods by Biotin-Streptavidin connectors. J. Am. Chem. Soc. 125, 1391413915. (6) Hermanson, G. T. (1996) Bioconjugate Techniques, Academic Press, San Diego. (7) Schetters, H. (1999) Avidin and Streptavidin in Clinical Diagnostics. Biomol. Eng. 16, 73-78. (8) Hadour, N. N., Gondran, C., and Cosnier, S. (2004) A New Biotinylated tris Bipyridyl Iron (II) Complex as Redox Biotin-Bridge
266 Bioconjugate Chem., Vol. 17, No. 2, 2006 for the Construction of Supramolecular Biosensing Architectures. Chem. Commun. 3, 324-325. (9) Goll, M. G., and Bestor, T. H. (2005) Eukaryotic Cytosine Methyltransferases. Annu. ReV. Biochem. 74, 481-514. (10) Bird, A. (2002) DNA methylation patterns and epigenetic memory. Genes DeV. 16, 6-21. (11) Jones, P. A., and Baylin, S. B. (2002) The fundamental role of epigenetic events in cancer. Nat. ReV. Genet. 3, 415-428. (12) Lyko, F., and Brown, R. (2005) DNA methyltransferase inhibitors and the development of epigenetic cancer therapies. J. Natl. Cancer Inst. 97, 1498-1506. (13) Sorm, F., Piskala, A., Cihak, A., and Vesely, J. (1964) 5-Azacytidine, a new, highly effective cancerostatic. Experientia 20, 202203. (14) Silvermann, L. R., Deamos, E. P., and Peterson, B. L. et al. (2002) Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J. Clin. Oncol. 20, 2429-2440. (15) Juttermann, R., Li, E., and Jaeniscch, R. (1994) Toxicity of 5-aza2′-deoxycytidine to mammalian cells is mediated primarily by covalent trapping of DNA methyltransferase rather than DNA demethylation. Proc. Natl. Acad. Sci. U.S.A. 91, 11797-11801. (16) Siedlecki, P., Boy, R. G., Musch, T., Brueckner, B., Suhai, S., Lyko, F., and Zielenkiewicz, P. (2006) Discovery of two novel small-molecule inhibitors of DNA methylation. J. Med. Chem. 49, 678-683.
Schirrmacher et al. (17) Brueckner, B., Garcia-Boy, R., Siedlecki, P., Musch, T., Kliem, H. C., Zielenkiewicz, P., Suhai, S, Wiessler, M., and Lyko, F. (2005) Epigenetic reactivation of tumor suppressor genes by a novel smallmolecule inhibitor of human DNA methyltransferases. Cancer Res. 65, 6305-6311. (18) Siedlecki, P., Garcia Boy, R., Comagic, S., Schirrmacher, R., Wiessler, M., Zielenkiewicz, P., Suhai, S., and Lyko, F. (2003) Establishment and functional validation of a structural homology model for human DNA methyltransferase 1. Biochem. Biophys. Res. Commun. 306, 558-563. (19) Casimir, J. R., Guichard, G., and Briand, J. P. (2002) Methyl 2-((succinimidooxy)carbonyl)benzoate (MSB): a new, efficient reagent for N-phthaloylation of amino acid and peptide derivatives. J. Org. Chem. 67, 3764-3768. (20) Lescrinier, T., Busson, R., Rozenski, J., Janssen, G., Van Aerschot, A., and Herdewijn, P. (1996) Incorporation of 5-Hydroxytryptophan in Oligopeptides. Tetrahedron 52, 6965-6972. (21) Islam, I., Ng, K., Chong, K. T., McQuade, T. J., Hui, J. O., Wilkinson, K. F., Rush, B. D., Ruwart, M. J., Borchardt, R. T., and Fisher, J. F. (1994) Evaluation of a Vitamin-Cloaking Strategy for Oligopeptide Therapeutics: Biotinylated HIV-1 Protease Inhibitors. J. Med. Chem. 37, 293-304. BC050300B
