Introduction

The formation of novel amino acids is of considerable importance both in terms of enabling the synthesis of naturally occurring peptides and in providing a means by which to modulate peptide properties (Zhang et al. 2013; Lu and Freeland 2006; Perdih and Dolenc 2011). The formation of novel aromatic amino acids via palladium catalysed cross-coupling reactions of orthogonally protected iodoalanines has been well documented in the literature. (Jackson et al. 1992, 2004; Rodriguez et al. 2003; Oswald et al. 2008; Wang et al. 2010) In particular this approach has been successfully utilised to prepare a wide range of phenylalanine derivatives (Deboves et al. 2001; Colgin et al. 2011; Ross et al. 2010, 2011). This strategy can also be used to access heteroaromatic amino acids such as pyridyl (Walker et al. 1997; Tabanella and Valancogne 2003), purinyl (Capek et al. 2004), indolyl (Tanaka et al. 2011) and bicyclic isoxazolyl (Clausen et al. 2009). However, examples of either furyl-alanine (1a and 2a) or thienyl-alanine amino acids (1b and 2b) produced via this route are not well documented (Fig. 1). To the best of our knowledge only two examples exist: The synthesis of 2-thienyl-alanine (1b) in a 10 % yield (Jackson et al. 1989) and 4-phosphothiophen-2-yl alanine in a 59 % yield under optimised conditions (Lilley et al. 2014).

Fig. 1
figure 1

Chemical structures of furyl-alanine (1a and 2a) and thienyl-alanine (1b and 2b)

Full size image

5-membered heteroaromatic amino acids such as 1 and 2 are frequently found in peptide natural products. For example, the 2-furyl moiety (1a) is contained in the natural peptide γ-glutamyl-2-furyl-alanine (3) (Hilker et al. 2010) and the 3-furyl moiety (2a) in Rhizonin (4) (Partida-Martinez et al. 2007) (Fig. 2). γ-Glutamyl-2-furyl-alanine (3) is produced by leaf beetle larvae as a chemical defence/feeding deterrent against insects and Rhizonin (4) is hepatotoxic cyclo-peptide produced by the fungal strain Rhizopus microsporus.

Fig. 2
figure 2

Peptide natural products that contain a furyl-alanine residue

Full size image

The chemical synthesis of 3 and 4 relies on having ready access to suitable 5-membered heteroaromatic amino acid building blocks. Herein, we report the preparation of five orthogonally protected 5-membered heteroaromatic amino acids. The key synthetic step in their preparation involves the Negishi-coupling reaction of an orthogonally protected iodoalanine with various halogenated 5-membered heterocycles.

Materials and methods

IR spectra were recorded on a Perkin Elmer Spectrum RX1 fitted with an ATR attachment. 1H NMR spectra were recorded at 400 MHz using a Bruker Avance 400 MHz and 13C NMR spectra at 100 MHz with a Bruker Avance. Chemical shifts are reported in ppm and are referenced to residual solvent peaks; CHCl3 (1H 7.26 ppm, 13C 77.0 ppm). J couplings are measured in Hertz (Hz). All reactions were monitored by T.L.C. using Merck precoated silica gel plates, Column chromatography was performed using silica gel (40–60 mM) using the solvent system indicated. Zinc dust (particle size <10 μM) was purchased from Sigma-Aldrich®. All reported yields refer to the isolated yield and the product purity was estimated to be >95 % by 1H NMR. Mass spectra were collected on a Waters TQD mass spectrometer and accurate mass spectra were collected on a Waters LCT Premier XE mass spectrometer. High pressure liquid chromatography was performed on an analytical Varian LC with a diode array detector. Optical rotations were measured with a Jasco P-1020 polarimeter.

Synthesis of orthogonally protected heteroaromatic amino acids 13–16, 21 general procedures

Reaction conditions A

Acid washed zinc dust (2.43 mmol, 4.00 equiv.) was heated under vacuum at 100 °C for 30 min whilst vigorously stirring. The zinc dust was cooled to 70 °C and placed under a positive pressure of argon, anhydrous DMF (0.5 mL) and I2 (cat ~0.015 g) were added and the light grey suspension stirred for a further 20 min. The reaction mixture was cooled to 50 °C at point which iodoalanine (7 or 19) (0.61 mmol, 1.00 equiv.) was dissolved in DMF (0.5 mL) and added. Stirring under argon was continued for another 20 min followed by addition of the halo-aromatic (8–12) (0.61 mmol, 1.00 equiv.), Tris(dibenzylideneacetone)dipalladium (Pd2(dba)3) (0.02 mmol, 0.03 equiv.) and tri(o-tolyl)phosphine (P(o-tol)3) (0.07 mmol, 0.10 equiv.). The reaction mixture was left to stir under argon at 50 °C for 5 h, then overnight at room temperature. The cooled reaction mixture was purified via column chromatography without workup (SiO2; 80/20 hexane/EtOAc → 100 % EtOAc and if impure a second column was run using SiO2; 100 % CH2Cl2).

Reaction conditions B

This follows the same reaction procedure as above: procedure A, however, 2-dicyclohexylphosphino-2′,6′-dimethoxybiphenyl (SPhos). is used as a replacement phosphine ligand to P(o-tol)3 in the same molar equivalents. The reaction mixture was purified as outlined in Procedure A.

Reaction conditions C

Acid washed zinc dust (2.43 mmol, 4.00 equiv.) was heated under vacuum at 100 °C for 30 min whilst vigorously stirring. The reaction mixture was cooled to 50 °C and placed under a positive pressure of argon, anhydrous DMF (0.5 mL) and I2 (cat. ~0.015 g) were added and the light grey suspension stirred for a further 20 min. The reaction mixture was cooled to room temperature at which point iodoalanine (7 or 19) (0.61 mmol, 1.00 equiv.) was dissolved in DMF (0.5 mL) and added. Stirring under argon was continued for another twenty mins and the halo-aromatic (8–12) (0.61 mmol, 1.00 equiv), Pd2(dba)3 (0.02 mmol, 0.03 equiv.) and SPhos (0.07 mmol, 0.10 equiv.) were added. The reaction mixture was left to stir overnight at room temperature under an argon atmosphere. The reaction mixture was purified as outlined in Procedure A.

Fmoc-β-(2-furyl)-Ala-OtBu (13)

The general procedure for palladium catalysed Negishi cross-coupling reaction conditions A was followed (0.61 mmol of 7). 13 obtained as a yellow crystalline solid, 11 % (0.07 mmol). v max(solid)/cm−1 2928 (w, C–H), 1702 (s bd, C = O); [α] D27 −5.6° (c 1.0, CHCl3); δH (400 MHz; CDCl3) 1.47 (9H, s, CO t2 Bu), 3.20 (2H, d, J 5.2, β-CH2), 4.24 (1H, app. t, Fmoc-CHCH2), 4.38 (2H, m, Fmoc-CHCH 2), 4.55 (1H, dt, J 5.2, 8.0, α-CH), 5.50 (1H, bd, J 8.0, NH), 6.09 (1H, d, J 2.8, furan-CH), 6.31 (1H, m, furan-CH), 7.30–7.34 (2H, m, Fmoc-ArH), 7.32 (1H, m, furan-CH), 7.39–7.43 (2H, m, Fmoc-ArH), 7.60 (2H, bd, J 7.4, Fmoc-ArH), 7.77 (2H, d, J 7.4, Fmoc-ArH); δC (100 MHz; CDCl3) 170.3, 155.7, 150.7, 144.0, 142.1, 141.4, 127.8, 127.2, 125.3, 120.1, 110.5, 108.1, 82.6, 67.2, 53.6, 47.3, 31.2, 28.0; HRMS m/z (ES+) 456.1784 (M+ + Na. C26H27NO5Na requires 456.1787). General procedure for palladium catalysed Negishi cross-coupling reaction conditions B was followed; Yellow crystalline solid, 42 % (0.26 mmol).

Fmoc-β-(3-furyl)-Ala-OtBu (14)

The general procedure for palladium catalysed Negishi cross-coupling reaction conditions B was followed (0.61 mmol of 7). 14 was obtained as a colourless oil, 27 % (0.17 mmol). v max(solid)/cm−1 3342 (w bd, N–H), 2979 (w, C–H), 1706 (s bd, C = O); [α] d28 −6.8° (c 1.0, CHCl3); δH (400 MHz; CDCl3) 1.46 (9H, s, CO t2 Bu), 2.95 (2H, m, β-CH2), 4.43 (1H, m, Fmoc-CHCH2), 4.42 (2H, m, Fmoc-CHCH 2), 4.51 (1H, m, α-CH), 5.35 (1H, bs, NH), 6.24 (1H, m, furan-CH), 7.23 (1H, m, furan-CH), 7.30–7.34 (2H, m, Fmoc-ArH), 7.37 (1H, m, furan-CH), 7.39–7.43 (2H, m, Fmoc-ArH), 7.57–7.60 (2H, m, Fmoc-ArH), 7.78 (2H, m, Fmoc-ArH); δC (100 MHz; CDCl3) 170.6, 155.8, 144.1, 143.1, 141.5, 140.6, 127.9, 127.2, 125.2, 120.2, 119.2, 111.5, 82.6, 67.1, 54.4, 47.4, 31.0, 28.2; HRMS m/z (ES+) 456.1785 (M+ + Na. C26H27NO5Na requires 456.1787).

Fmoc-β-(2-thienyl)-Ala-OtBu (15)

The general procedure for palladium catalysed Negishi cross-coupling reaction conditions B was followed with 2-iodothiophene (0.61 mmol of 7). 15 was obtained as a yellow oil, 26 % (0.16 mmol). v max(solid)/cm−1 3342 (w bd, N–H), 2977 (w, C–H), 1710 (s bd, C = O); [α] d27 15.0° (c 1.0, CHCl3); δH (400 MHz; CDCl3) 1.48 (9H, s, CO t2 Bu), 3.39 (2H, m, β-CH2) 4.26 (1H, t, J 7.2, Fmoc-CHCH2), 4.35 (1H, dd, 7.2, 10.4, Fmoc-CHCH 2), 4.48 (1H, dd, 7.2, 10.4, Fmoc-CHCH 2), 4.58 (1H, m, α-CH), 5.90 (1H, bd, J 8.0, NH), 6.83 (1H, d, J 3.0 thiophene-CH), 6.96 (1H, dd, J 3.0, 5.0, thiophene-CH), 7.20 (1H, d, J 5.0, thiophene-CH), 7.32–7.36 (2H, m, Fmoc-ArH), 7.40–7.44 (2H, m, Fmoc-ArH), 7.61–7.64 (2H, m, Fmoc-ArH), 7.78 (2H, bd, J 7.6, Fmoc-ArH); δC (100 MHz; CDCl3) 170.0, 155.7, 144.0, 141.4, 137.6, 127.8, 127.1, 127.0, 126.9, 125.2, 124.8, 120.1, 83.8, 67.1, 55.0, 47.3, 32.5, 28.1; HRMS m/z (ES+) 472.1562 (M+ + Na. C26H27NO4SNa requires 472.1559).

General procedure for palladium catalysed Negishi cross-coupling reaction conditions B was followed with 2-bromothiophene; Yellow oil, 24 % (0.15 mmol).

General procedure for palladium catalysed Negishi cross-coupling reaction conditions C with 2-iodothiophene starting material was followed; Yellow oil, 22 % (0.13 mmol).

Fmoc-β-(3-thienyl)-Ala-OtBu (16)

The general procedure for palladium catalysed Negishi cross-coupling reaction conditions B was followed (0.61 mmol of 7). 16 was obtained yellow oil, 20 % (0.12 mmol). v max(solid)/cm−1 3342 (w bd, N–H), 2976 (w, C–H), 1707 (s bd, C = O); [α] d28 −18.1° (c 1.0, CHCl3); δH (400 MHz; CDCl3) 1.45 (9H, s, CO2tBu), 3.15 (2H, m, β-CH2) 4.23 (1H, m, Fmoc-CHCH2), 4.30–4.57 (3H, m, Fmoc-CHCH 2, α-CH), 5.35 (1H, bs, NH), 6.92 (1H, d, J 4.0, thiophene-CH), 6.98 (1H, m, thiophene-CH), 7.26 (1H, m, thiophene-CH), 7.29–7.35 (2H, m, Fmoc-ArH), 7.39–7.44 (2H, m, Fmoc-ArH), 7.58–7.61 (2H, m Fmoc-ArH) 7.77 (2H, d, J 7.6, Fmoc-ArH); δC (100 MHz; CDCl3) 170.7, 155.7, 144.0, 141.4, 136.3, 128.7, 127.8, 127.2, 125.8, 125.2, 122.9, 120.1, 82.5, 67.0, 54.8, 47.3, 32.9, 28.0; HRMS m/z (ES+) 472.1563 (M+ + Na. C26H27NO4SNa requires 472.1559).

Boc- β-(2-furyl)-Ala-OBn (20)

The general procedure for palladium catalysed Negishi cross-Coupling reaction conditions B was followed (0.65 mmol of 19). 20 was obtained yellow oil, 86 % (0.56 mmol). [α] d28 −7.0° (c 1.0, CHCl3); δH (400 MHz; CDCl3) 1.43 (9H, s, BocH), 3.14 (2H, m, β-CH2), 4.61 (1H, m, α-CH), 5.46 (2H, m, CH2Ar), 5.95 (1H, d, J 3.2, furan-CH), 6.24 (1H, dd, J 2.0, 3.2, furan-CH), 7.27 (1H, d, J, 2.0 Hz, furan-CH), 7.31–7.40 (5H, m, ArH); δC (100 MHz; CDCl3) 171.4, 155.2, 150.4, 142.2, 135.4, 128.6, 128.5, 110.4, 108.0, 80.1, 67.3, 52.9, 31.0 28.4, 18.6; HRMS m/z (ES+) 368.1473 (M+ + Na. C19H23NO5Na requires 368.1474).

Z-Glu-(-β-(2-Furyl)-Ala-OBn)-OBn (21)

Boc-β-(2-furyl)-Ala-OBn 20 (0.14 mmol, 1.00 equiv.) was stirred in TFA: DCM 1:4 (10 mL) until the Boc deprotection was seen to be complete by TLC. The reaction mixture was then evaporated under reduced pressure to dryness. The resulting yellow powder was dissolved in DCM (15 mL). Z-Glu-OBn (0.14 mmol, 1.00 equiv.), PyBOP® (0.14 mmol, 1.00 equiv.) DIPEA (3.00 equiv.) were added and the reaction mixture was stirred overnight at room temperature. The reaction mixture was evaporated under reduced pressure and purified via column chromatography (SiO2; 95/5 % hexane/EtOAc → 40/60 % hexane/EtOAc) to yield 21; White powder, 78 % (0.11 mmol). δH (400 MHz; CDCl3) 1.90–2.34 (4H, m, β-CH and γ-CH2), 3.18 (2H, m, β-CH2), 4.46 (1H, m, α-CH) 4.87 (1H, m, α-CH), 5.08–5.28 (6H, m, CH2Ar), 5.62 (1H, d, J 8.0, NH), 5.94 (1H, d, J 2.8, furan-CH) 6.19 (1H, t, J 2.8, furan-CH), 6.28 (1H, d, J 7.2, NH), 7.21 (1H, m, furan-CH), 7.27–7.41 (15H, m, ArH); δC (100 MHz; CDCl3) 171.0, 150.3, 142.3, 136.3. 135.3, 128.8, 128.7, 128.6, 128.5, 128.4, 128.3, 128.2, 110.4, 108.2, 67.6, 67.5, 67.2, 53.6, 51.8, 48.4, 32.2, 30.6, 28.4, 18.5; HRMS m/z (ES+) 599.2387 (M+ + H. C34H34N2O8 requires 599.2393).

Results and discussion

In the synthesis of amino acids via the Negishi coupling of orthogonally protected iodoalanines, a Boc/Bn protecting group strategy is most commonly adopted. It was our aim to prepare amino acids that would be compatible with Fmoc solid phase peptide synthesis (SPPS) and as such an Fmoc/tBu protecting group strategy was adopted. It should be noted that there are only a few examples of novel amino acids being prepared via the Negishi coupling of orthogonally protected iodoalanines in which Fmoc has been employed as the protecting group (Eerland and Hedberg 2012).

The known orthogonally protected iodoalanine (7) was prepared on a multi-gram scale in two steps from commercially available Fmoc-l-serine (5) as shown in Scheme 1. The first step in the synthesis involved protection of the carboxylic acid as the tert-butyl ester and this was achieved in good yield using tert-butyl 2,2,2-trichloroacetimidate (TBTA) (Cobb and Vederas 2007). The conversion of the serine alkyl alcohol (β-OH) to an alkyl iodide was readily achieved in excellent yield using I2, PPh3 and imidazole (Arndt et al. 2001).

Scheme 1
scheme 1

Synthesis of orthogonally protected iodoalanine 7

Full size image

With the orthogonally protected iodoalanine (7) in hand, we attempted to carry out Negishi cross-coupling reactions (Scheme 2) with various halogenated 5-membered heteroaromatics (Table 1, Entries 1–5).

Scheme 2
scheme 2

Synthesis of orthogonally protected heteroaromatic amino acids 13–16

Full size image
Table 1 Synthesis of 5-membered heteroaromatic amino acids
Full size table

The initial coupling conditions employed (Table 1, Conditions A) only yielded the furan derivative 13 (Table 1, Entry 1) in low yield (11 %). The expected products for Entries 2–5 (Table 1, 14–16) were visible in the mass spectrum obtained from the crude reaction mixtures but no products could be isolated by column chromatography. The lack of product formation in Entries 2–5 (Table 1) was in part due to the formation of Fmoc-Ala-OtBu. The formation of the unwanted alanine side product in this type of reaction arises from protonation of the organozinc intermediate and is well documented in the literature (Jackson et al. 1998).

To obtain an improved synthesis of 13 and isolable amounts of products 14–16, we looked to alter the reaction conditions. The bulky biaryl ligand, SPhos has literature precedence for improving Pd cross-coupling reactions, as seen by Buchwald’s group in Suzuki–Miyaura reactions (Barder et al. 2005; Martin et al. 2007) and by Jackson’s group in Negishi type cross-coupling reactions (Ross et al. 2010). Specifically of interest to our work, Jackson’s group showed that in the Negishi type cross-coupling of iodoalanines switching from P(o-tolyl)3 to SPhos could significant increase reaction yields (Ross et al. 2010).

In addition to altering the phosphine ligand, temperature, reaction times and solvent were also investigated. The final conditions used in the reaction [Table 1, conditions (B)] gave greatly improved but less than optimal yields. The lower yields achieved are believed to have arisen due to instability of the Fmoc protecting group under the reaction condition as this has been previous documented (Ross et al. 2010). This hypothesis was also investigated further using Boc-iodoalanine-OBn (19) as an alternative staring material in the couplings. Orthogonally protected 19 was accessed from cheap and readily available Boc-l-serine (17) on a multi-gram scale using the process described in Scheme 3.

Scheme 3
scheme 3

Synthesis of orthogonally protected iodoalanine 19

Full size image

Scheme 4 shows the synthesis of Boc-β-(2-furyl)-Ala-OBn (20) and Fmoc-β-(2-furyl)-Ala-OtBu (13) from their respective iodoalanine pre-cursors via a Pd catalysed Negishi cross-coupling. As Table 2 highlights when the orthogonally protecting groups are altered there is a significant increase in the yield of the product obtained. It was also noted that the reaction was much cleaner and that the unwanted side-products such as the alanine derivative were suppressed to a much greater extent.

Scheme 4
scheme 4

Synthesis of orthogonally protected 2-furyl amino acids 13 and 20

Full size image
Table 2 Yields of orthogonally protected 2-furyl amino acids
Full size table

As previously discussed, furyl-alanines (1a and 2a) can be found in peptide natural products (Partida-Martinez et al. 2007; Hilker et al. 2010) but the stability of these amino acids in synthetic procedures has been reported to be poor under acidic conditions (Schulz et al. 2004). To investigate this further, the TFA deprotection of Boc-β-(2-furyl)-Ala (20) and its subsequent incorporation into a di-peptide (21) was undertaken (Scheme 5). In our hands the furan moiety present in 20 proved stable to the acidic conditions employed and the desired di-peptide (21) was prepared in good yield. 21 is an intermediate in the synthesis of the natural product γ-glutamly-2-furyl-alanine (3) and various attempts were made to carry out the global deprotection of 21 to give 3. Under all conditions, hydrogenation of 21 yielded a complex mixture of inseparable products with the desired product 3 being identified only as a potential product in analysis of the crude reaction mixture by mass spectrometry.

Scheme 5
scheme 5

Synthesis of protected di-peptide (21), a precursor to γ-glutamyl-2-furyl-alanine (3)

Full size image

Conclusions

Five orthogonally protected 5-membered heteroaromatic amino acids have been synthesised (13–16, 20) via the Negishi cross-coupling of protected iodoalanines. The SPhos ligand was favoured in the Negishi cross-coupling reaction and Fmoc/OtBu protected furyl-alanines (13 and 14) and thienyl-alanines (15 and 16) were prepared in low to moderate yields (20–42 %). Despite the lower reaction yields, the synthetic process presented herein to access 13–16 has advantages in that it is; (a) shorter than other published routes (Hilker et al. 2010) (b) generates SPPS compatible amino acids and (c) it is amenable to scale up. Application of a Boc/OBn protecting group strategy does lead to an enhanced yield (86 %) in the Negishi cross-coupling reaction but the reaction conditions required to remove the benzyl ester were found to be incompatible with the furan ring. Amino acids 13–16 are now being investigated as key building blocks in the on resin synthesis of 3 and various analogues.