Triazinyl-Amino Acids, New Building Blocks for Pseudopeptides

Abstract

The synthesis of a new heterocyclic building block for unnatural peptide analogues is described, and examples of its application are demonstrated. The 1,3,5-triazine nucleus is readily derivatized to possess an amino and a carboxy terminus, and the third site can be used to incorporate a wide variety of functional groups into an oligomer. This third site can also serve as the point of attachment to a solid-phase resin, allowing convenient construction of macrocyclic pseudopeptides that should be useful as clefts for molecular recognition studies or models of β-strand conformation.

Key words

unnatural amino acids - peptides - macrocycles - peptidomimetics - triazines

References

• 2a Kemp DS. Allen TJ. Oslick SL. Boyd JG. J. Am. Chem. Soc.  1996,  118:  4240 
• 2b Nowick JS. Acc. Chem. Res.  1999,  32:  287 
• 2c Janetka JW. Raman P. Satyshur K. Flentke GR. Rich DH. J. Am. Chem. Soc.  1997,  119:  441 
• 2d Feng Y. Pattarawarapan M. Wang Z. Burgess K. J. Org. Chem.  1999,  64:  9175 
• A few recent leading references:
• 3a Hruby VJ. Balse PM. Curr. Med. Chem.  2000,  7:  945 
• 3b Abell AD. Gardiner J. J. Org. Chem.  1999,  64:  9668 
• 3c Chakraborty TK. Ghosh S. Jayaprakash S. Sharma JARP. Ravikanth V. Diwan PV. Nagaraj R. Kunwar AC. J. Org. Chem.  2000,  65:  6441 
• 3d Geyer A. Moser F. Eur. J. Org. Chem.  2000,  1113 
• 3e Goschke R. Cohen NC. Wood JM. Maibaum J. Bioorg. Med. Chem. Lett.  1997,  7:  2735 
• 3f Rosin CD. Belew RK. Morris GM. Olson AJ. Goodsell DS. Proc. Natl. Acad. Sci. U.S.A.  1999,  96:  1369 
• 4a Kubik S. Goddard R. J. Org. Chem.  1999,  64:  9475 
• 4b Rasmussen PH. Rebek J. Tetrahedron Lett.  1999,  40:  3511 
• 4c Miyake H. Kojima Y. Coord. Chem. Rev.  1996,  148:  301 
• 5a Hamuro Y. Geib SJ. Hamilton AD. J. Am. Chem. Soc.  1997,  119:  10587 
• 5b Zhu J. Parra RD. Zeng H. Skrzypczak-Jankun E. Zeng XC. Gong B. J. Am. Chem. Soc.  2000,  122:  4219 
• 6 Kline T. Sieber-McMaster E. Lau WF. Natarajan S. Bioorg. Med. Chem. Lett.  1997,  7:  1017 
• 7a Yu Q. Baroni TE. Liable-Sands L. Yap GPA. Rheingold AL. Borovik AS. Chem. Commun.  1999,  1467 
• 7b Ranganathan D. Haridas V. Karle IL. J. Am. Chem. Soc.  1998,  120:  2695 
• 7c Berl V. Huc I. Khoury RG. Krische MJ. Lehn J.-M. Nature (London)  2000,  407:  720 
• 8 Ranganathan D. Haridas V. Nagaraj R. Karle IL. J. Org. Chem.  2000,  65:  4415 
• 9 Gellman SH. Acc. Chem. Res.  1998,  31:  173 
• 10 Quirke JME. In Comprehensive Heterocyclic Chemistry   Vol. 3:  Bolton AJ. McKillop A. Pergamon; Oxford: 1984.  p.457 
• 11 Staatz I. Granzer UH. Roth HJ. Liebigs Ann. Chem.  1989,  51 
• 12a Bauer W, Geisenberger J, Menzel H, and Pedrazzi R. inventors; Eur. Patent  1036823.  ; Chem. Abstr. 2000, 133, 253884
• 12b Sano H, Sato N, and Murata J. inventors; Japanese Patent  07150088.  ; Chem. Abstr. 1995, 123, 343739
• 12c Vignali G, Guizzardi F, and Zagnoni G. inventors; Eur. Patent  627428.  ; Chem. Abstr. 1994, 122, 160691
• 12d Teraoka H.; Japanese Patent 11320921, 1999; Chem. Abstr. 1999, 132, 7575
• 13a Scharn D. Wenschuh H. Reineke U. Schneider-Mergener J. Germeroth L. J. Comb. Chem.  2000,  2:  361 
• 13b Falorni M. Giacomelli G. Mameli L. Porcheddu A. Tetrahedron Lett.  1998,  39:  7607 
• 13c Gustafson GR. Baldino CM. O’Donnell M.-ME. Sheldon A. Tarsa RJ. Verni CJ. Coffen DL. Tetrahedron  1998,  54:  4051 
• 16a Solid-phase syntheses were performed using standard Fmoc/Boc reaction conditions on polystyrene beads derivatized with the Wang linker, see: Synthetic Peptides: A User’s Guide   Grant GA. W. H. Freeman; New York: 1992. 
• 16b

  Amino acids were protected at their N-termini with the Fmoc group and at side chains as the N-Boc derivative for Lys and the t-butyl ether for Tyr; couplings used diisopropylcarbodiimide (DIC) and 1-hydroxybenzotriazole (HOBt) in DMF; Fmoc was removed with 20% piperidine in DMF; the final peptide was cleaved from the resin with trifluoroacetic acid.

• 18a The allyl group was removed by shaking the resin under N₂ for 3 h with 150-200 mg Pd(PPh₃)₄ in 3 mL 37:2:1 CHCl₃:HOAc:N-methyl morpholine: Kates SA. Daniels SB. Sole NA. Barany G. Albericio F. In Peptides: Chemistry, Structure, Biology; Proceedings of the 13th American Peptide Symposium   Hodges RS. Smith JA. ESCOM; Leiden: 1994.  p.113 
• 18b

  The completeness of the deprotection was monitored by removing small portions of the resin, cleaving the peptide, and injecting the product on HPLC to observe complete loss of starting material (protected analogue). Two Pd treatments were required for molecule 4.

1

Current address: Organix, Inc. 240 Salem St., Woburn, MA 01801, USA.

14

General Procedure for Synthesis of 1: To a solution of cyanuric chloride (0.2 g, 1.08 mmol) in CH₂Cl₂ in an ice bath was added solid glycine t-butyl ester hydrochloride (180 mg, 1.08 mmol) followed by diisopropylethylamine (DIEA, 376 µL, 2.16 mmol). The reaction was magnetically stirred for 30 min, then allowed to warm to r.t. Piperidine, for version 1a, (107 µL, 1.08 mmol) was then added, followed by another equivalent of DIEA. The reaction was heated in an oil bath to 50 °C for 3 h. Solvent was removed on the rotary evaporator, the residue was taken up in EtOAc and washed twice with a solution of citric acid to remove DIEA salts. NMR and TLC analysis of this crude material indicated that it was almost purely the desired 2-chloro-4-(glycine t-butyl ester)-6-piperidyl triazine, so it was used without further purification. It was dissolved in dioxane, heated to 70 °C, and ethylenediamine (147 µL, 2.2 mmol) and DIEA (188 µL, 1.08 mmol) were added. An excess of ethylenediamine is used to avoid formation of the triazine-ethylenediamine-triazine dimer. The temperature of the bath was increased to 100 °C, and a small amount (approximately 1 mL) of water was added to dissolve a cloudy precipitate that had formed. This temperature was maintained for 4 h. After cooling the reaction mixture to r.t., Fmoc-chloride (1.3 g, 5 mmol) and DIEA (0.87 mL, 5 mmol) were added, and the resulting mixture was stirred at r.t. for 3 h. The solvents were removed on the rotary evaporator. The residue was taken up in EtOAc and washed with dilute aq citric acid. The EtOAc was removed in vacuo and the crude material was purified by flash chromatography on silica gel with 1:1 CH₂Cl₂-EtOAc. The desired Fmoc-1a-t-butyl ester had an R_f of 0.2 in this solvent and 0.5 in pure EtOAc; its yield was 68%. ¹H NMR (400 MHz, d ₆-acetone): δ = 7.84 (2 H, d, J = 7.4 Hz), 7.66 (2 H, d, J = 7.0 Hz), 7.40 (2 H, t, J = 7.2 Hz), 7.31 (2 H, t, J = 7.2 Hz), 6.76 (1 H, br s), 6.32-5.92 (2 H, br m), 4.31 (2 H, d, J = 6.8 Hz), 4.21 (1 H, t, J = 7.1Hz), 3.93 (2 H, s), 3.70 (4 H, br s), 3.50 (2 H, m), 3.35 (2 H, q, J = 5.5 Hz), 1.60 (6 H, s), 1.44 (9 H, s); ¹³C NMR (100 MHz, d ₆-acetone): δ = 170.9, 167.6, 167.3, 165.7, 157.4, 145.2, 142.1, 128.5, 128.0, 126.2, 120.8, 81.1, 67.0, 48.1, 44.6, 44.2, 42.3, 41.1, 28.3, 26.6, 25.6. Calcd for C₃₁H₃₉N₇O₄·H₂O: C, 62.9; H, 7.0; N, 16.6%. Found: C, 63.2; H, 6.7; N, 16.4%. These Fmoc/t-butyl esters of class 1 could be stored in the freezer for months with no apparent decomposition. At room temperature, however, an insoluble decomposition product did slowly form over the course of months. This product was not observed if the t-butyl group was removed with TFA before storage, so the monomers were usually stored as the free carboxylic acid. The other versions of molecule 1 were prepared by analogous synthetic routes:
1b: Fmoc-EDA-TR(NH(CH₂)₄CO₂H)-GlyOAl: ¹H NMR (400 MHz, CDCl₃): δ = 10.75 (1 H, br s), 7.73 (2 H, d, J = 7.4 Hz), 7.55 (2 H, d, J = 7.0 Hz), 7.34 (2 H, t, J = 7.2 Hz), 7.26 (2 H, t, J = 7.2 Hz), 6.33 (2 H, br s), 6.10-5.95 (2 H, br m), 5.87 (1 H, m), 5.29 (1 H, d, J = 17.5 Hz), 5.21 (1 H, d, J = 10.7 Hz), 4.60 (2 H, d, J = 5.4 Hz), 4.33 (2 H, d, J = 7.0 Hz), 4.17 (1 H, m), 4.07 (2 H, s), 3.44 (2 H, m), 3.33 (4 H, m), 2.28 (2 H, t, J = 7.0 Hz), 1.61 (4 H, m). ¹³C NMR (100 MHz, CDCl₃): δ = 176.6, 170.8, 169.9, 167.2, 166.3, 157.4, 145.1, 142.0, 136.9, 128.5, 128.1, 126.2, 120.6, 115.1, 69.2, 66.8, 57.7, 44.3, 43.5, 42.1, 40.0, 35.6, 31.2, 22.3. Yield 80%.
1c: Fmoc-EDA-TR(3-toluidyl)-LeuOH: ¹H NMR (400 MHz, CDCl₃): δ = 7.75 (2 H, d, J = 7.3 Hz), 7.64 (2 H, d, J = 7.0 Hz), 7.53-7.22 (6 H, m), 7.06 (1 H, m), 6.81 (1 H, d, J = 7.8 Hz), 4.36 (1 H, t, J = 7.0 Hz), 4.29 (2 H, m), 4.16 (1 H, m), 2.20 (3 H, s), 1.83 (1 H, m), 1.75 (2 H, m), 0.94 (6 H, d, J = 6.3 Hz). ¹³C NMR (100 MHz, CDCl₃): δ =176.4, 167.3, 166.9, 166.1, 157.2, 146.5, 145.2, 142.1, 139.0, 129.3, 128.3, 127.7, 126.5, 121.0, 119.3, 116.0, 112.9, 66.7, 61.2, 44.4, 43.2, 39.8, 37.3, 23.4, 22.1, 21.9, 20.8. Yield 72%.
1d: Fmoc-EDA-TR(3-toluidyl)-GlyOH: ¹H NMR (400 MHz, d ₆-DMSO): δ = 7.88 (2 H, d, J = 7.3 Hz), 7.67 (2 H, d, J = 7.2 Hz), 7.5-7.25 (6 H, m), 7.08 (1 H, m), 6.73 (1 H, d, J = 7.8 Hz), 4.27 (2 H, m), 4.12 (1 H, m), 3.91 (2 H, s), 3.18 (2 H, m), 2.24 (3 H, s); 2 H’s from the EDA moiety were obscured by HOD, and all the NH’s were exchanged. ¹³C NMR (100 MHz, d ₆ -DMSO): δ = 177.1, 167.6, 167.1, 157.5, 147.0, 145.3, 142.2, 138.9, 129.2, 128.5, 128.0, 126.5, 121.0, 119.5, 115.8, 113.0, 67.1, 58.8, 45.1, 43.4, 41.0, 21.2. Yield 68%.
1e: Fmoc-EDA-TR(hexylamino)-GlyOH: ¹H NMR (400 MHz, CDCl₃): δ = 10.24 (1 H, br s), 7.69 (2 H, m), 7.53 (2 H, m), 7.32 (4 H, m), 6.47-6.05 (2 H, br m), 5.19 (2 H, m), 4.29 (2 H, m), 4.22-3.91 (3 H, m), 3.62-3.08 (6 H, m), 1.66-1.05 (8 H, m), 0.84 (3 H, t, J = 7.6 Hz). ¹³C NMR (100 MHz, CDCl₃): δ = 175.0, 167.2, 167.0, 166.3, 157.2, 145.1, 142.1, 128.4, 128.0, 126.3, 120.5, 66.8, 58.5, 44.2, 43.6, 42.1, 39.9, 32.0, 31.3, 27.3, 22.7, 13.5. Yield 82%.
1f: Fmoc-EDA-TR(2-naphthylamino)-GlyOH: ¹H NMR (400 MHz, CDCl₃): δ = 10.5 (1 H, br s), 8.16 (1 H, br s), 7.71 (5 H, m), 7.53 (3 H, m), 7.45-7.14 (7 H, m), 6.29-5.70 (3 H, br m), 4.35 (2 H, m), 4.26-3.93 (3 H, m), 3.51 (2 H, m), 3.37 (2 H, m). ¹³C NMR (100 MHz, CDCl₃): δ = 175.2, 167.5, 167.0, 157.2, 145.0, 142.8, 142.1, 132.9, 128.4, 128.0, 127.5, 126.1, 124.9, 121.1, 120.6, 118.9, 109.6, 66.7, 58.4, 44.4, 42.3, 39.9. Yield 65%.

15

Yield of the trimer was 64% based on Fmoc-1a-OH. M/z (ES): Calcd 627.4 Da; found 628.3 Da [M + H⁺].

17

m/z (ES) for the heptamer: Calcd 1345.6 Da; found 1346.5 Da [M + H⁺]. The product had a retention time of 13.9 min on a 3.9 × 150 mm Waters Nova-Pak C₁₈ HPLC column, flow rate 1.0 mL/min, 25 min gradient from 95:5 0.1% TFA in H₂O:CH₃CN to 5:95 0.1% TFA in H₂O:CH₃CN.

19

m/z (MALDI-TOF) for 2: Calcd 791.4 Da; found 792.0 Da. The retention time was 18.4 min using the same conditions as in ref. [17] . The purity of the crude final product was greater than 90% as approximated by integration of peaks observed in the HPLC. The yield was 53% based on the amount of resin used, but yield data are at best only rough approximations for these solid-phase reactions, since unweighed aliquots of the resin are removed for ninhydrin testing at various stages of the synthesis. Macrocycles 2, 3 and 4 were synthesized on the 10 micromole scale, starting with approximately 100-120 mg resin loaded at 0.8 milliequiv/g.

20

m/z (ES) for 3: Calcd 900.5 Da; found 901.5 Da. For 4: Calcd 1293.7 Da; found 1294.9 Da. The retention times were 14.4 and 16.2 min respectively using the same conditions as in ref. [17] . Percent purities were in the 68-75% range for these two molecules; the only side-products observed in the crude preparations were very broad peaks eluting at long retention times, which we suspect may be non-cyclized oligomers of the open-chain precursors of 3 and 4, and which are easily separated from the desired macrocycles using HPLC purification. Future work will explore resins with lower loadings to minimize this problem.