De novo synthesis of alkyne substituted tryptophans as chemical probes for protein profiling studies

Abstract

De novo synthesis of alkynalted tryptophan moieties as chemical probes for protein profilling studies, via an unexpected synthesis of Michael acceptor 12 is reported. Friedel Craft's alkylation of 12 and alkyne substituted indoles gave alkynalated tryptophan moieties, which perform as chemical probe by incorporating into actively translating Escherichia coli cells, whereby the alkyne moiety enables multimodal analyses through click chemistry mediated attachment of reporting groups.

Bioisosteric amide bond replacement and mimicking of peptide secondary structure have been the most popular approach for developing peptidomimetic agents.^{1, 2} However, over the years, incorporation of unnatural α-amino acid derivatives has gained interest.^{3, 4} Non-proteinogenic α-amino acids with restricted conformational mobility have great potential to elucidate biologically active conformations of peptides.^{5, 6} There are several examples ranging from simple α-methylated to various α, β-amino acids.⁷ The Orphan and Tirrell research groups have demonstrated that incorporation of an unnatural amino acid possessing a terminal alkyne facilitates click chemistry of a fluorescent reporter to track active protein synthesis; also known as BONCAT (biorthogonal noncanonical amino acid tagging).^8-11 The use of nonproteinogenic α-amino acids has not only led to characterization of, and modification to, activity, stability, bioavailability and binding specificity of proteins and peptides, but has also enabled studies on protein folding, protein function and signal transduction.^{5, 12-14}

A major focus on synthesis of non-proteinogenic amino acids has emphasized modifications to the α-carbon. However, amino acids with aromatic side chains (histidine, phenylalanine, tyrosine and tryptophan) allow modification on the ring structure, thereby maintaining the functionality of the amino acid. Our group has developed chemical probes for important metabolites, including unnatural amino acids, adding terminal alkynes capable of undergoing click-chemistry with an azide modified reporter group. These probes allow multimodal studies such as imaging and mass spectrometry-based proteomic studies which is the primary focus of research in our group.^15-19 We now sought to develop an alkynylated version of tryptophan to evaluate both incorporation of unnatural amino acids into proteins, and characterize amino acid-protein binding.

Tryptophan residues are known to bind cations through π-interactions. This interaction has led to several approaches on the synthesis of Trp regioisomers where the alanine unit is not attached to the indole C-3, but to C-2, C-4 and C-7 ²⁰. At the onset of our research we identified several reports on the synthesis of functionalized Trp moieties with a variety of substituents on the indole ring, including NO₂, halides, alkyl chains, ethers, and others^21-23 To the best of our knowledge, no report of an alkyne substituted Trp moiety directly attached to C-2 or C4-C7 positions of the indole ring (Fig.1, A) has been reported, although propargylation of hydroxy tryptophan has been recently reported (Fig.1, B).²⁴

Figure 1.

Representative alkynylated Tryptophan moieties.

The retrosynthetic approach depicted in Scheme 1 shows that Pd catalyzed Sonogashira coupling should be an accessible route to incorporate an alkyne group on the indole ring. A similar approach has been successfully accomplished on iodotyrosine^{25, 26} where the amino acid was protected using 9-BBN, and Bonjoch et. al. reported the synthesis of 9-BBN protected hydroxy Trp and Tyr to study their activity against malignant melanoma.²⁷ We decided to use a similar approach in protecting bromo-tryptophan. 5- & 6- bromo-tryptophan were subjected to 9-BBN protection to give compounds 1 & 2, and further, the indole N-H was protected using Boc anhydride to give 3 & 4 in near quantitative yields (Scheme 3).

Scheme 1.

Synthesis of 9-BBN protected bromo-tryptophans.

Scheme 3.

a. Activating C-3 position of indole; b. retrosynthetic approach for de novo tryptophan synthesis; c. synthesis of intermediate 12.

To test the incorporation of an alkyne group on the indole ring, compound 1 was subjected to Sonogashira coupling using Pd (II) catalyst, Cu(I), base and TMS-acetylene. MS analysis was employed to analyse the reaction, which did not show any alkyne incorporated product. We therefore decided to study the effects of Pd catalyst, base, solvent and temperature on all four substrates to enhance the Sonogashira coupling reaction.

None of the conditions listed in Table 1 gave indication of product formation, and in most cases dehalogenation of starting material was observed. Further, bromides 2 and 4 were also subjected to the Molander trifluoroboronates,²⁸ but to no avail. With none of Pd catalysed Sonogashira coupling chemistry giving positive reaction outcomes, we further investigated if 5-hydroxy tryptophan would serve as a better substrate for alkyne incorporation.

Table 1.

Failed attempts on Sonogashira coupling of 9-BBN protected 5/6-bromotryptophan.

Compd	Alkyne	Catalyst	Base	Solvent	T (°C)
1		Pd(PPh3)4 PdCl2(PPh3)2 PdCl2dppf	K2CO3	THF Dioxane DMF	25
2			Cs2CO3		80
3			TEA		100
4			DiPEA		120
2		PdCl2(PPh3)2	Cs2CO3	DMF:H2O	80
4				5:1

Reaction conditions: Standard Sonogashira coupling conditions were followed. Alkyne, Pd catalyst, base and solvent mixed in a reaction vessel, degassed by bubbling N₂ and subjected to heating. Reactions were analysed using LTQ-MS.

As shown in Scheme 2, 5-hydroxy tryptophan was protected with 9-BBN to give 5. We envisioned alkylating the hydroxy position of 5 to give 6 via direct o-alkylation of the hydroxy indole, using Karageorge and Macor's strategy,^{29, 30} however the desired o-alkylation was not achieved with this approach. Zhu, et. al.²⁴ have shown that using standard protecting groups like Boc (for amine) and ester (for acid), propargylation of 5-hydroxy tryptophan is possible. We speculated that the ionic nature of 5, with 9-BBN protecting group, may have caused a decreased solubility in acetone leading to the failure. As a control, 5-hydoxy indole on treatment with propargylbromide in the presence of Cs₂CO₃ in acetone gave o-alkylated product 14 (see table 3, entry 3).³¹ Therefore, we began to investigate the possibility of a de novo tryptophan synthesis from indole and substituted indoles.

Scheme 2.

Attempting OH-propargylation of compound 5.

Table 3.

Synthesis of alkyne substituted tryptophan moieties.

Entry	Starting indolea	Productb	Yield %
1		R = H (11)	63
2			51
3			69
4			73
5			71
6			64

Reaction conditions: 12 (1 eq), corresponding indole (1.5 eq) and AlCl3 (2 eq) were added to a reaction flask containing anhydrous DCM at 0 °C. Reaction was monitored by TLC and MS analysis.

^aSee supporting information for synthesis of indole precursors.

^bCorresponding tryptophan moieties.

^cIsolated yields reported after flash chromatography purification.

It is well known that in the presence of a strong base the C-3 indole position is activated to promote nucleophilic attack via SN₂ reaction. For example, strong bases like n-BuLi³² and EtMgBr³³ have been used to bring about nucleophilic substitution at the indole C-3 position. As shown in Scheme 3a, we subjected indole to freshly prepared EtMgBr to successfully synthesize 3-benzyl indole (7) and 3-propargyl indole (8). With this success, we envisioned that bromide 10 (Scheme 3b), which can be obtained from serine methyl ester, would be an excellent substrate to bring about the synthesis of protected tryptophan 11.

Based on the retrosynthetic approach shown in Scheme 3b, in our quest to synthesize bromide 10, the amino group of methyl serine was protected as benzophenone imine 9 and subjected to Appel reaction using CBr₄ and PPh₃. However, instead of brominating the hydroxy group of 9 to give bromide 10, the reaction yielded formation of intermediate 12 (Scheme 3c), an excellent Michael acceptor. Although a known compound,²² formerly used for the de novo syntheses of substituted tryptophan, synthesis of 12 via Appel reaction has not been reported to the best of our knowledge. We also observed the reaction to be highly concentration-dependent such that a concentrated reaction mixture (deep red-brown in colour) did not proceed in the forward direction. However, with significant dilution (observed as a bright yellow colour) of the starting reactants we observed complete conversion to 12.³⁴ With this observation we further developed the process by performing a step-wise scale up of the synthesis of 12 from ester 9 as shown in Table 2.

Table 2.

Scale up process: step-wise synthesis of 12.

Entry	9 (g)	CBr4 (g)	PPh3 (g)	12 (g)	Yield (%)
1	0.10	0.145	0.185	0.061	65.0
2	0.25	0.365	0.463	0.15	64.0
3	0.50	0.731	0.925	0.32	68.0
4	2.50	3.660	4.630	1.57	67.0
5	10.0	14.631	18.526	5.02	71.0

With significant quantities of compound 12 in hand, we sought to use this key intermediate in our substituted indole synthesis. Bartolucci²³ and Balsamini²¹ have reported an electrophilic aromatic substitution approach using EtAlCl₂ and AlCl₃ as Lewis acids to synthesize substituted tryptophan moieties using compound 12. We envisioned that a similar approach using alkyne-substituted indoles would enable successful incorporation of the desired alkyne on the aromatic ring of tryptophan.

With all necessary elements in hand, we set forth a test reaction by reacting the indole with 12 in the presence of AlCl₃ (Table 3; entry 1) in anhydrous DCM at 0 °C. Under these conditions, alkylation at the C3 position went smoothly in less than 2h as monitored by TLC. Next, 5-ethynyl indole (Table 3; entry 2) was subjected to alkylation conditions to give 5-ethynyl tryptophan, which was otherwise inaccessible with other reaction routes as described earlier. To extend the scope of this methodology we decided to study other terminal alkyne groups on the C5 position of indole. 5-hydroxy indole was subjected to o-alkylation using propargyl bromide and Cs₂CO₃ to give a mixture of O-alkylated (14) and O-,N-alkylated (15) products, which were isolated by column chromatography. These two compounds also gave desired tryptophan moieties in good yields when reacted with 12 in presence of AlCl₃ (Table 3; entries 3 and 4). Table 3 entries 5 and 6 show that long chain terminal alkynes on indole rings are accessible via this approach. Purity of the starting indole, compound 12 and maintaining anhydrous conditions was very important for the success of this methodology. Performing this reaction with impure starting material or non-anhydrous solvent resulted in a competing side reaction where dimerization of compound 12 led to formation of side product 13, ³⁵ which was confirmed by X-ray crystallographic studies CCDC 1516420 (Fig 2).

Figure 2.

X-ray crystal structure of side product 13 (ellipsoid view).

To study the application of alkynylated tryptophans in protein synthesis, 19 was subjected to standard deprotection conditions using HCl and LiOH over a two-step process giving alkyne-substituted tryptophan 24 (Scheme 4). Escherichia coli BL21 cells were grown at 37 °C to an OD₆₀₀ of 1.0 on M9 media supplemented with 22 mM glucose and either 200 μM L-tryptophan or 200 μM 24. To confirm that 24 is incorporated into proteins in actively translating E. coli, the cells were lysed and the fluorescent reporter group, azido-rhodamine, was added to proteins incorporating 24 via click chemistry. Proteins were separated by gel electrophoresis, and analysed for fluorescence (Figure 2).

Scheme 4.

Synthesis of alkyne substituted tryptophan 24.

As shown in figure 3a, proteins isolated from E. coli grown in the presence of compound 24 displays strong fluorescence compared to the no alkyne control and no click chemistry control. Figure 3b shows that after coomassie staining an equal amount of protein was loaded into each lane, thus the fluorescence signal is dependent on incorporation of 24 into proteins, demonstrating that compound 24 is a functional non-proteinogenic amino acid.

Figure 3.

Poly-acrylamide gel analysis of proteins isolated from E. coli cells grown in the presence of L-tryptophan (lane 1 & 3) or compound 24 (lane 2 & 4). a) Fluorescence detection of incorporation of compound 24 via cycloaddition of azido-rhodamine (Lanes 1 & 2 no click chemistry controls, Lanes 3 & 4 click chemistry). b) Coomassie stain of proteins shows that all gel lanes contained equivalent protein loading. L = ECL Rainbow ladder.

In summary, we have synthesized new alkyne substituted tryptophan moieties. Sonogashira couplings on 9-BBN protected 5- & 6-bromotryptophan was not successful, nor was o-alkylation of 9-BBN protected 5-hydroxy tryptophan (5). Bromination of the hydroxy group of 9 via Appel reaction yielded Michael acceptor 12, an excellent substrate for the synthesis of alkyne substituted tryptophans via electrophilic aromatic substitution. We have applied this strategy to synthesize new alkyne-substituted unnatural tryptophan moieties and demonstrated their utility within a biological system. We further envision incorporation of diazirine species into the alkyl chain, enabling applications in protein profiling studies that require a covalent bond formation, such as proteomic analysis of proteins that bind tryptophan (e.g., TrpR).