A high-throughput synthesis of 1,2,4-oxadiazole and 1,2,4-triazole libraries in a continuous flow reactor

Abstract

We report herein a high-throughput methodology for the synthesis of 1,2,4-oxadiazole and 1,2,4-triazole small-molecule libraries using an integrated synthesis and purification platform. The heterocyclization relies first on a low-temperature peptide coupling of a diverse set of carboxylic acids and hydroxyamidines, hydrazonamides, or pyridyl hydrazides followed by a high-temperature cyclization to yield the respective heterocycles in a continuous flow process. The fully integrated synthesis and purification platform enables the rapid generation of chemical libraries, decreasing the drug discovery cycle time.

---

Introduction

As pharmaceutical companies continue to work in a competitive therapeutic landscape, the ability for drug discovery programs to rapidly synthesize diverse, biologically active small molecules in a highly efficient manner is of the utmost importance. For this reason, the development of automated synthesis platforms has been at the forefront of many high-throughput chemistry organizations within drug discovery.^1–3 With the implementation of systems such as these, the drug discovery cycle time can be greatly decreased, and structure activity relationships can be generated faster than by using traditional medicinal chemistry techniques.

Coupled with this technology is the equally important development of novel methodologies that permit the facile and rapid analoging of complex chemical structures. Ideally, increased diversity from chemical libraries is enabled by having a large set of readily available monomers (such as carboxylic acids and amines). Bioisosteres of amides and esters such as 1,2,4-oxadiazoles and 1,2,4-triazoles are important heterocycles within drug discovery, as they are not susceptible to hydrolysis and demonstrate improved ADME properties.⁴ As such, developing a robust, high-throughput method to synthesize libraries of these heterocycles is a valuable asset to a drug discovery program. We report herein a high-throughput methodology for the rapid synthesis of 1,2,4-oxadiazole and 1,2,4-triazole libraries using an integrated synthesis and purification platform in flow.

Results and discussion

Previously, we have reported on our high-throughput chemistry efforts, using both batch chemistry,^5–8 as well as flow chemistry.⁹ Over time, our platforms have evolved into fully automated synthesis-purification platforms designed to reduce the overall turnaround time for library synthesis from 7–10 days to 2–4 days. Specifically, we have demonstrated the use of the Accendo Conjure flow reactor (Fig. 1) as a library synthesis platform called SWIFT (Synthesis With Integrated Flow Technology).^10,11 The system relies on segmented flow, using an immiscible fluorous spacer to create discrete reaction segments in flow.^12–17 Each reaction segment represents a single library element which, upon exiting the flow reactor, is directly injected onto a mass-triggered preparative HPLC for purification. The system is capable of synthesizing and purifying up to six compounds per hour.^18,19

Fig. 1 Schematic of Conjure flow reactor.

Lange and James have previously utilized the Conjure in a two-stage library synthesis, with an initial first-stage amine functionalization using the incubation chamber, followed by a second-step heterocycle formation in the system's main reactor (Fig. 1).²⁰ Using a similar approach, we have chosen to use this two-stage functionality not to add multiple points of diversity, but to synthesize reactive intermediates in situ in the incubation chamber, followed by a high-temperature second step to promote heterocycle formation.

The synthesis of 1,2,4-oxadiazoles in flow has previously been reported by Cosford.²¹ This procedure utilizes an in situ synthesis of hydroxyamidines, followed by the coupling of an acid chloride and a high-temperature cyclization to yield the final product. In our high-throughput chemistry approach, we sought conditions for the heterocycle formation that could be easily translated to core structures similar to hydroxyamidines (such as hydrazonamides and 2-hydrazinopyridines) that could yield complex chemical matter using a diverse set of carboxylic acid monomers.

Initially, the conditions to cyclize intermediate 1 to 1,2,4-oxadiazole 2 were investigated using design of experiments (DoE) in flow (Scheme 1). For the multi-parameter optimization of heterocycle formation, a range of temperatures, residence times and different equivalents of base (DIPEA) were screened (Table 1).²² Analysis of the data indicated that the amount of base used in the cyclization was irrelevant, but increased reaction times gave higher conversions. However, the most important factor in heterocycle formation was temperature. Minimal conversion to heterocycle 2 was observed at 100 °C, while complete conversion was observed at 200 °C (Table 2). While reactions at 200 °C indicated complete conversion by UV, additional peaks were present in the mass spectrum, indicating decomposition was starting to occur. For this reason, reaction temperatures between 150 and 175 °C were determined to be the optimal temperature for cyclization (Fig. 2).

Scheme 1 1,2,4-Oxadiazole formation from intermediate 1.

Table 1 Parameter ranges for Design-of-Experiments (DoE) optimization of 1,2,4-oxadiazole formation

Parameter	Range
Temperature (°C)	100, 150, 200
Residence time (min)	5, 10, 15
Equivalents of DIPEA	0, 1.5, 4

---

Table 2 Results of DoE optimization of heterocycle formation

Residence time (min)	Temperature (°C)	Eq. DIPEA	% prod. (UV)
a Center point, average of 4 runs.
5.0	100	1.5	5.0
5.0	150	0.0	50.0
5.0	150	4.0	53.9
5.0	200	1.5	88.2
10.0	100	0.0	10.4
10.0	100	4.0	9.9
10.0	150	1.5	87.7^a
10.0	200	4.0	89.2
10.0	200	0.0	98.4
15.0	100	1.5	17.5
15.0	150	0.0	92.2
15.0	150	4.0	92.7
15.0	200	1.5	95.7

---

Fig. 2 DoE contour plots for the (a) relationship between reactor temperature and equivalents of DIPEA at 11 minute residence time and (b) relationship between residence time and temperature at 175 °C reactor temperature.

Having optimized the cyclization step, conditions to form intermediate 1 from carboxylic acids and hydroxyamidines were investigated. The use of HATU as the coupling reagent using DMA as the solvent was shown to rapidly form intermediate 1 using a 1 : 1 stoichiometric ratio of HATU and carboxylic acid. Using the incubation chamber and injection loops, we were able to mimic the sequential addition of reagents typically associated with a peptide coupling. The carboxylic acid and HATU were loaded into the injection loop for 5 minutes at 30 °C, and subsequently loaded into the flow reactors injection loop to yield intermediate 1 (Scheme 2). We subsequently attempted to couple the two reaction steps in flow, using the first stage incubation chamber and injection loop in sequence to form intermediate 1, then injecting this reagent stream into the heated reactor to afford heterocycle 2.

Scheme 2 Flow sequence for the synthesis of intermediate 1.

Analysis of the DoE data predicted an optimal temperature between 150 and 175 °C and reaction times longer than 10 minutes to obtain complete conversion. A small number of experiments were run to determine the maximum isolated yield of 2 in these instances (Table 3). While significant reaction tailing during the purification gave lower than predicted isolated yields, data indicated that longer reaction times and elevated temperatures can be detrimental to reaction outcome (Table 3, entry 3 and 4). Reaction conditions of 11 minutes and 175 °C were deemed optimal for the heterocycle formation in the event that more sterically hindered substrates might have a higher activation barrier to cyclization (Table 3, entry 1). Using a first stage incubation of 30 °C for 5 minutes, followed by a second step at 175 °C for 11 min, 1,2,4-oxadiazole 2 was generated in a 57% isolated yield (Scheme 3).²³

Table 3 Isolated yields of 1,2,4-oxadiazole 2 using DoE-suggested reaction conditions

Entry	Residence time (min)	Temperature (°C)	Isolated yield
1	11.0	150	57%
2	11.0	175	57%
3	13.0	150	51%
4	13.0	175	45%

---

Scheme 3 Continuous flow synthesis of 1,2,4-oxadiazole 2.

A small matrix library of hydroxyamidines and carboxylic acids was synthesized to demonstrate the scope of the heterocycle formation. Aliphatic and aromatic hydroxyamidines can be reacted with aliphatic or aromatic carboxylic acids in modest to high yield (Fig. 3). The chemistry is compatible with protecting groups, as no loss of the Boc-protecting group was observed under the high-temperature conditions. The reaction can also be run with modest to high retention of stereochemistry, which is seemingly dependent on the hydroxyamidine starting material. The three-stage setup of the reaction shown in Scheme 3 is key to obtaining high yields, as when all reagents are mixed in the incubation chamber prior to being injected into the high-temperature reactor, the percent yield decreases and more by-product formation is observed. In instances where yields are low, additional optimization can be carried out to optimize the process if needed. For library purposes, however, the reaction conditions used in library synthesis generated sufficient material necessary for primary assay and ADME screening, providing the medicinal chemist with sufficient information to generate SAR and make a go/no-go decision on the final compounds.

Fig. 3 Library production of 1,2,4-oxadiazoles in continuous flow process. ^aIsolated yielded (in mg) and percent yield after purification. ^bPercent yield when carboxylic acid, HATU (1.0 eq.), hydroxyamidine (1.0 eq.) and DIPEA (3.0 eq.) are loaded into first stage incubation chamber prior to being loaded into high temperature reactor.

We envisioned that other heterocycles such as 1,2,4-triazoles could also be synthesized using the same methodology. Starting materials are readily accessed by treating nitriles, thioamides, or 2-chloropyridines with hydrazine. The resulting hydrazonamides and 2-hydrazinopyridines could be utilized as cores in a library with a set of carboxylic acid monomers to generate a diverse set of 1,2,4-triazoles (Fig. 4 and 5, respectively). To synthesize 1,2,4-triazoles from hydrazonamides, reaction conditions identical to those used for the synthesis of 1,2,4-oxadiazoles were used. It was shown that the same 175 °C and 11 minute residence time was sufficient to convert the starting material into 1,2,4-triazoles in modest to high yields. This chemistry is compatible with a series of aromatic and aliphatic carboxylic acid monomers, which show a minimal effect on stereochemistry.²⁴

Fig. 4 Library production of 1,2,4-triazoles in continuous flow process. ^aIsolated yielded (in mg) and percent yield after purification.

Fig. 5 Library production of 1,2,4-triazoles in continuous flow process. ^aIsolated yielded (in mg) and percent yield after purification.

Synthesizing 1,2,4-triazoles from 2-hydrazidopyridine⁶ required similar conditions to form the intermediate acyl hydrazide, however higher temperatures and longer reaction times were required to ensure complete conversion (250 °C and 20 minutes, respectively).²⁵ Aromatic and aliphatic substrates give modest to high yields, while further optimization would be required to increase the yield of substrates with a quaternary center alpha to the carbonyl.²⁶ Unlike many syntheses of 1,2,4-triazolopyridines, this thermal process required no additives to promote the heterocycle formation.^27–30

Conclusions

In conclusion, we have demonstrated the application of novel, high-throughput methodology to make a diverse set of 1,2,4-oxadiazoles and 1,2,4-triazoles. Using a fully integrated synthesis and purification platform, libraries of heterocycles were rapidly generated in modest to high yield, with sufficient enough material to be used in primary assay screens. When the system is run in a fully integrated fashion, compounds are synthesized and purified at a rate of four compounds per hour. Additionally, the capability to use a first stage incubation chamber to synthesize reactive intermediates was demonstrated in library format, opening the door for additional multi-step reactions to be performed in our automated flow reactor.

Experimental

General information

All starting materials were commercially available and used without purification unless noted. 4-Fluorobenzohydrazonamide hydroiodide was synthesized from 4-fluorobenzothioamide.

General purification procedure

Samples were purified by preparative HPLC on a Phenomenex Luna C8(2) 5 μm 100 Å AXIA column (50 mm × 21.2 mm). A gradient of acetonitrile (A) and 0.1% trifluoroacetic acid in water (B) was used, at a flow rate of 30 mL min⁻¹ (0–0.5 min 5% A, 0.5–6.5 min linear gradient 5–100% A, 6.5–8.5 min 100% A, 8.5–9.0 min linear gradient 100–5% A, 9.0–10 min 5% A). A sample volume of 1.0 mL was injected directly from the flow reactor stream to the HPLC system. A custom purification system was used, consisting of the following modules: Gilson 305 and 306 pumps; Gilson 806 Manometric module; Gilson UV/Vis 155 detector; Gilson 506C interface box; Gilson FC204 fraction collector; Agilent G1968D Active Splitter; Thermo MSQ Plus mass spectrometer. The system was controlled through a combination of Thermo Xcalibur 2.0.7 software and a custom application written in-house using Microsoft Visual Basic 6.0.

General procedure for oxadiazole formation

A stock solution of carboxylic and DIPEA (0.60 M and 1.8 in DMA, respectively, 166.7 μL, 0.10 mmol carboxylic acid (1.0 eq.) and 0.3 mmol DIPEA (3.0 eq.)) and HATU (0.50 M in DMA, 200 μL, 0.10 mmol, 1.0 eq.) were aspirated from their respective source vials, mixed through a PFA mixing tube (0.2 mm inner diameter), and loaded into an incubation chamber held at 30 °C for 5 minutes. The reaction segment was mixed with a stock solution of hydroxyamidine (0.60 M, 166.7 μL, 0.10 mmol, 1.0 eq.), loaded into an injection loop for one minute, followed by injection into the flow reactor (Hastelloy coil, 0.75 mm inner diameter, 1.8 mL internal volume) set at 175 °C, and passed through the reactor at 163 μL min⁻¹ (11 minute residence time) pressurized to 1000 psi using a back-pressure regulator. Upon exiting the reactor, the reaction was loaded directly into an injection loop and purified using preparative HPLC/MS to yield the final compound.

General procedure for monocyclic triazole formation

A stock solution of carboxylic and DIPEA (0.60 M and 1.8 in DMA, respectively, 166.7 μL, 0.10 mmol carboxylic acid (1.0 eq.) and 0.3 mmol DIPEA (3.0 eq.)) and HATU (0.50 M in DMA, 200 μL, 0.10 mmol, 1.0 eq.) were aspirated from their respective source vials, mixed through a PFA mixing tube (0.2 mm inner diameter), and loaded into an incubation chamber held at 30 °C for 5 minutes. The reaction segment was mixed with a stock solution of hydrazonamide (0.60 M, 166.7 μL, 0.10 mmol, 1.0 eq.), loaded into an injection loop for one minute, followed by injection into the flow reactor (Hastelloy coil, 0.75 mm inner diameter, 1.8 mL internal volume) set at 175 °C, and passed through the reactor at 163 μL min⁻¹ (11 minute residence time) pressurized to 1000 psi using a back-pressure regulator. Upon exiting the reactor, the reaction was loaded directly into an injection loop and purified using preparative HPLC/MS to yield the final compound.

General procedure for bicyclic triazole formation

A stock solution of carboxylic and DIPEA (0.60 M and 1.8 in DMA, respectively, 166.7 μL, 0.10 mmol carboxylic acid (1.0 eq.) and 0.3 mmol DIPEA (3.0 eq.)) and HATU (0.50 M in DMA, 200 μL, 0.10 mmol, 1.0 eq.) were aspirated from their respective source vials, mixed through a PFA mixing tube (0.2 mm inner diameter), and loaded into an incubation chamber held at 30 °C for 5 minutes. The reaction segment was mixed with a stock solution of 2-hydrazinopyridine (0.60 M, 166.7 μL, 0.10 mmol, 1.0 eq.), loaded into an injection loop for one minute, followed by injection into the flow reactor (Hastelloy coil, 0.75 mm inner diameter, 1.8 mL internal volume) set at 250 °C, and passed through the reactor at 90 μL min⁻¹ (20 minute residence time) pressurized to 1000 psi using a back-pressure regulator. Upon exiting the reactor, the reaction was collected in a fraction collector. Samples were purified using preparative HPLC/MS to yield the final compound.

Notes and references

1. M. Werner, C. Kuratli, R. E. Martin, R. Hochstrasser, D. Wechsler, T. Enderle, A. I. Alanine and H. Vogel, Angew. Chem., Int. Ed., 2014, 53, 1704–1708.
2. S. V. Ley, D. E. Fitzpatrick, R. J. Ingham and R. M. Myers, Angew. Chem., Int. Ed., 2015, 54, 3449–3464.
3. T. Rodrigues, P. Schneider and G. Schneider, Angew. Chem., Int. Ed., 2014, 53, 5750–5758.
4. S. Hordhoff, S. Bulat, S. Cerezo-Galvez, O. Hill, B. Hoffman-Enger, M. Lopez-Canet, C. Rosenbaum, C. Rummey, M. Thiemann, V. G. Matassa, P. J. Edwards and A. Feurer, Bioorg. Med. Chem. Lett., 2009, 19, 6340–6345.
5. Y. Wang, K. Sarris, D. R. Sauer and S. W. Djuric, Tetrahedron Lett., 2007, 48, 5181–5184.
6. Y. Wang, K. Sarris, D. R. Sauer and S. W. Djuric, Tetrahedron Lett., 2007, 48, 2237–2240.
7. Y. Wang, D. R. Sauer and S. W. Djuric, Tetrahedron Lett., 2005, 47, 105–108.
8. N. P. Tu, P. A. Searle and K. Sarris, J. Lab. Autom., 2015 DOI:10.1177/2211068215590580.
9. Select reviews on flow chemistry: D. T. McQuade and P. H. Seeberger, J. Org. Chem., 2013, 78, 6384–6389; J. C. Pastre, D. L. Browne and S. V. Ley, Chem. Soc. Rev., 2013, 42, 8849–8869; J. Wegner, S. Ceylan and A. Kirschning, Chem. Commun., 2011, 47, 4583–4592; B. P. Mason, K. E. Price, J. L. Steinbacher, A. R. Bogdan and D. T. McQuade, Chem. Rev., 2007, 107, 2300–2318; Microreactors in Organic Synthesis and Catalysis, ed. T. Wirth, Wiley-VCH, Weinheim, 2008; V. Hessel, P. Lob and H. Lowe, Curr. Org. Chem., 2005, 9, 765; V. Hessel and H. Lowe, Chem. Eng. Technol., 2005, 28, 267; B. Ahmed-Omer, J. C. Brandt and T. Wirth, Org. Biomol. Chem., 2007, 5, 733.
10. J. E. Hochlowski, P. A. Searle, N. P. Tu, J. Y. Pan, S. G. Spanton and S. W. Djuric, J. Flow Chem., 2011, 2, 56–61.
11. J. D. Sutherland, N. P. Tu, T. A. Nemcek, P. A. Searle, J. E. Hochlowski, S. W. Djuric and J. Y. Pan, J. Lab. Autom., 2014, 19, 176–182.
12. A. R. Bogdan and N. W. Sach, Adv. Synth. Catal., 2009, 351, 849–854.
13. C. M. Thompson, J. L. Poole, J. L. Cross, I. Akritopoulou-Zanze and S. W. Djuric, Molecules, 2011, 16, 9161–9177.
14. P. P. Lange, A. R. Bogdan and K. James, Adv. Synth. Catal., 2012, 354, 2373–2379.
15. A. R. Bogdan and K. James, Chem.–Eur. J., 2010, 16, 14506–14512.
16. A. R. Bogdan and K. James, Org. Lett., 2011, 13, 4060–4063.
17. N. Hawbaker, E. Wittgrove, B. Christensen, N. W. Sach and D. G. Blackmond, Org. Process Res. Dev., 2015 DOI:10.1021/op500360w.
18. The compounds per hour rate is limited by the preparative HPLC method. A standard HPLC method is 10 minutes long.
19. On average, libraries are run using 15–20 mg of core per reaction.
20. P. P. Lange and K. James, ACS Comb. Sci., 2012, 14, 570–578.
21. D. Grant, R. Dahl and N. D. P. Cosford, J. Org. Chem., 2008, 73, 7219–7223.
22. Dimethylacetamide (DMA) was chosen as the reaction solvent due to the high solubility of compounds and due to its compatibility with the integrated purification.
23. Without the first stage incubation (i.e. all reagents are mixed in the injection loop and loaded into the reactor), the heterocyclization does not take place.
24. For examples of 1,2,4-triazoles synthesized from nitriles and hyrazides, see the following: K.-S. Yeung, M. E. Farkas, J. F. Kadow and N. A. Meanwell, Tetrahedron Lett., 2005, 46, 3429–3432.
25. For the library being run at 250 °C, reactions were collected in a fraction collector and purified separately using preparative HPLC-MS. When using the fully integrated synthesis and purification platform, the observation was made that significant off-gassing of the reaction caused a small percentage of the reaction mixture to be diverted to waste upon rotation of an inject valve connected to the preparative HPLC-MS. While this was determined to be <20% of the reaction mixture, the valve was taken out of line to obtain representative yields for the chemistry.
26. By LC-MS, the major product was the uncyclized acylhydrazide, indicating that a longer reaction time or higher temperature would be required to drive the heterocyclization to completion.
27. M. Ciesielski, D. Pufky and M. Doring, Tetrahedron, 2005, 61, 5942–5947.
28. A. K. Sadana, Y. Mirza, K. R. Aneja and O. Prakash, Eur. J. Med. Chem., 2003, 38, 535–536.
29. A. Moulin, J. Martinez and J.-A. Fehrentz, Tetrahedron Lett., 2006, 47, 7591–7594.
30. No safety assessment or DSC was performed on the high-temperature stability of HATU and therefore more thorough studies should be carried out prior to running this cyclization in batch or on large-scale.

---

Footnote

† Electronic supplementary information (ESI) available: See DOI: 10.1039/c5ra18386c

---