Synthesis of Coumarin-Conjugated Oligonucleotides via Knoevenagel Condensation to Prepare an Oligonucleotide Library

Abstract

DNA-encoded libraries (DELs) are attracting attention as a screening tool in the early stages of drug discovery. In the development of DELs, drug candidate compounds are chemically synthesized on barcode DNA. Therefore, it is important to perform the synthesis under mild conditions so as to not damage the DNA. On the other hand, coumarins are gaining increasing research focus not only because they possess excellent fluorescence properties, but also because many medicines contain a coumarin skeleton. Among the various reactions developed for the synthesis of coumarins thus far, Knoevenagel condensation followed by intramolecular cyclization under mild conditions can yield coumarins. In this study, we developed a new synthetic method for preparing a coumarin-conjugated oligonucleotide library via Knoevenagel condensation. The results showed that coumarins substituted at the 5-, 6-, 7-, or 8-positions could be constructed on DNA to afford a total of 26 coumarin-conjugated DNAs. Moreover, this method was compatible with enzymatic ligation, demonstrating its utility in DEL synthesis. The developed strategy for the construction of coumarin scaffolds based on Knoevenagel condensation may contribute to the use of DELs in drug discovery and medicinal chemistry.

Introduction

DNA-encoded libraries (DELs), first proposed by Brenner and Lerner in 1992,¹⁾ have evolved into a valuable platform for hit identification in the early stage of drug discovery.^2,3) DEL technology has facilitated the development of large-scale combinatorial libraries. In a DEL, each library compound is bound to a unique DNA fragment and serves as an amplifiable identification barcode. A DEL enables the rapid and economical DNA-encoded collection of numerous compounds against biological targets through the confluence of molecular biology, combinatorial chemistry, and high-throughput sequencing.⁴⁾ Accordingly, various synthetic methods for preparing DELs have been developed thus far.^5,6)

Coumarin (2-oxo-2H-1-benzopyran) is an important scaffold for various natural products,^7,8) and warfarin,⁹⁾ umbelliferone,¹⁰⁾ hymecromone,¹¹⁾ and toddaculin¹²⁾ are representative drugs that contain coumarin scaffolds. Owing to their excellent optical properties, coumarin derivatives have been employed as fluorescent probes and tracers in biological applications and solar energy collectors.^13–15) The development of synthetic methods to construct and functionalize the coumarin core has attracted much attention lately. For example, coumarin derivatives can be synthesized by Knoevenagel condensation,^16,17) Pechmann condensation,^18,19) the Perkin reaction,²⁰⁾ the Wittig reaction,^21,22) and the Baylis–Hillman reaction.²³⁾ For DEL synthesis, it is important to select reactions that are less likely to damage DNA. Typically, the adenine and guanine nucleobases in DNA can be removed under acidic conditions.²⁴⁾ In addition, DNA is degraded in the presence of some transition-metal catalysts (e.g., Pd(OAc)₂ and Ce(NH₄)₂(NO₃)₆) during DEL synthesis.²⁵⁾ These findings have revealed that Knoevenagel condensation followed by intramolecular cyclization proceeds from malonic esters and salicylaldehydes under weakly basic conditions, and that this reaction may be suitable to prepare a coumarin-conjugated DNA library. Therefore, we considered that a coumarin scaffold can be constructed by using DNA as a substrate on which malonates or salicylaldehyde are immobilized (Method A and Method B in Fig. 1). In this study, the Knoevenagel condensation-based construction of coumarin scaffolds on DNA was carried out, and the substrate scope and limitations of this reaction were investigated.

Fig. 1. Synthetic Strategy for Preparing a Coumarin-Conjugated DNA Library in This Study

Results and Discussion

Initially, DNA2a and DNA2b, which are substrates for Knoevenagel condensation, were synthesized via amide bond formation by the treatment of a T10-mer oligonucleotide possessing an aminohexyl linker at the 5ʹ-end (DNA1) with NHS-esters 1²⁶⁾ and 2²⁷⁾ under basic conditions (Chart 1). The intramolecular cyclization that occurred after Knoevenagel condensation was investigated using DNA2a and salicylaldehyde as model substrates; the results are summarized in Table 1 and the representative HPLC chart of the crude mixture of the Knoevenagel condensation reaction is shown in Supplementary Fig. S2. The desired DNA was isolated by reverse-phase HPLC, and the molecular weight of the product was measured by matrix assisted laser desorption/ionization-time of flight-mass spectrometry (MALDI-TOF-MS) to confirm the synthesis of the target DNA. Using 100 mM salicylaldehyde and 1 mM piperidine, the coumarin-conjugated DNA3a was isolated in 13% yield and the starting material DNA2a was recovered in 57% yield (entry 1). The formation of DNA3a was confirmed by the condensation of coumarin-3-carboxylic acid succinimidyl ester²⁸⁾ with DNA1 to similarly yield DNA3a (Supplementary Chart S1 and Fig. S1 in Supplementary materials). Comparing entries 1 and 2 and entries 3 and 4, respectively, increasing the concentrations of piperidine led to improved yields of the desired DNA3a and decrease the recovery of the substrate DNA2a. The results showed that the concentration of DNA2a increased from 200 µM to 1 mM based on the conditions in entry 4, resulting in a further increase in the yield of the coumarin-conjugated DNA3a to 69% (entry 5). Finally, by changing the concentration of piperidine to 100 mM based on the condition in entry 5, the substrate DNA2a was completely converted into DNA3a (entry 6).

Chart 1. Construction of a Coumarin Scaffold on DNA via Knoevenagel Condensation Followed by Intramolecular Cyclization

Table 1. Optimization of the Knoevenagel Condensation Reaction from DNA2a

Entry	DNA2aa)	Salicylaldehydea	Piperidinea)	EtOH	DNA3ab)	Recovered DNA2ab)
1	20 nmol (200 μΜ)	10 µmol (100 mM)	0.1 µmol (1 mM)	100 µL	13%	57%
2	20 nmol (200 μΜ)	10 µmol (100 mM)	1 µmol (10 mM)	100 µL	19%	52%
3	20 nmol (200 μΜ)	50 µmol (500 mM)	0.1 µmol (1 mM)	100 µL	13%	61%
4	20 nmol (200 μΜ)	50 µmol (500 mM)	1 µmol (10 mM)	100 µL	50%	26%
5	20 nmol (1 mM)	10 µmol (500 mM)	0.2 µmol (10 mM)	20 µL	69%	10%
6	20 nmol (1 mM)	10 µmol (500 mM)	2 µmol (100 mM)	20 µL	87%	0%

a) The values in parentheses indicate concentrations of DNA2a and reagents. b) Isolated yield after HPLC purification.

The progress of Knoevenagel condensation followed by intramolecular cyclization was similarly investigated for DNA2b. However, the desired DNA3b was not obtained at all. A rate-limiting step in piperidine-catalyzed Knoevenagel condensation is the formation of an iminium ion between piperidine and aldehyde.²⁹⁾ This formation is an equilibrium process; therefore, it was considered that high concentrations of both, salicylaldehyde and piperidine, may be required to construct the coumarin scaffold on DNA, and that the resulting few iminium ions would react with DNA2a (Method A in Fig. 1 and Chart 1). However, the construction of the coumarin scaffold did not proceed by Method B (Fig. 1 and Chart 1) because the concentration of DNA2b attached with salicylaldehyde was possibly too low for iminium ion formation due to the low solubility of DNA in organic solvents.

The substrate scope and limitations of salicylaldehydes in Method A (Chart 1) were investigated. For this investigation, DNA4, which contains four nucleobases and has an aminohexyl linker at the 5ʹ-end, was conjugated with NHS-ester 1 to afford DNA5 in 70% yield, following which Knoevenagel condensation using DNA5 was carried out. Various salicylaldehydes with electron-withdrawing and electron-donating groups were used; the results are summarized in Fig. 2. For most substrates, Knoevenagel condensation followed by intramolecular cyclization successfully proceeded to furnish coumarins, yielding a total of 26 types of the desired DNA6. Thus, not only most of the substituents could be introduced at the 5-, 6-, 7-, and 8-positions of the coumarin scaffold on the DNA, but also tricyclic-fused coumarins (DNA6-27 and DNA6-28) could be synthesized. In addition, it was possible to synthesize 7-aminocoumarin derivatives (DNA6-18 and DNA6-29), which are useful as fluorescent materials.^30–32) On the other hand, the treatment of DNA5 with salicylaldehyde substituted with a hydroxy group afforded none of the desired products (DNA6-10, DNA6-16, or DNA6-23). Given the side reactions that may proceed during this reaction, several studies have shown that salicylaldehyde derivatives with a hydroxy group react with electrophiles via the Friedel-Crafts reaction.^33–36) These findings suggest that the hydroxy group-substituted salicylaldehyde might polymerize in the presence of piperidine, thereby reducing the concentration of salicylaldehyde to prevent Knoevenagel condensation on DNA.

Fig. 2. Substrate Scope and Limitations

Lastly, to assess the applicability of the optimized reaction conditions to the synthesis of the DEL, enzymatic ligation reactions were performed to confirm their compatibility with DEL synthesis (Chart 2). For this investigation, a hairpin DNA modified with an amino linker (DNA7, Supplementary Table S1 in Supplementary materials) was used as the starting material, and a model small molecule possessing a coumarin scaffold was synthesized in three steps: 1) condensation with pentafluorophenyl ester 3³⁷⁾ followed by the deprotection of the Fmoc group, 2) condensation with malonate, and 3) construction of the coumarin scaffold. Three types of double-stranded DNA (dsDNA) barcodes (Supplementary Table S1 in Supplementary materials) were used in this experiment: dsDNA1 corresponding to compound 3, dsDNA2 corresponding to malonate, and dsDNA3 corresponding to salicylaldehyde. For the synthesis of coumarin derivatives by Knoevenagel condensation followed by intramolecular cyclization, reagents possessing both, an active methylene group and an alkoxycarbonyl group, can be used instead of malonates.^16,17) Therefore, dsDNA2 was prepared as a barcode for active methylene compounds. Ligation reactions using these dsDNA barcodes successfully proceeded, and the desired DNA11 could be obtained. Thus, in this study, the synthesis of a DEL with a model coumarin derivative was achieved by a method based on Knoevenagel condensation. In some DEL syntheses, DELs are purified by simple manipulations such as ethanol precipitation, in order to easily provide structural diversity of DELs. However, in the model DEL synthesis in this study, since the reaction efficiency is not quantitative, HPLC purification is required after each reaction, and further improvements of our synthetic method are necessary in terms of simplicity of purification. Additionally, structural changes in nucleobases due to reactions with alkylating reagents and electrophiles may directly lead to the loss of coding information in DELs.³⁸⁾ To this end, Ratnayake et al. reported a qPCR-based method to evaluate the DNA amplifiability after DEL synthesis.³⁹⁾ Although it is necessary to confirm the damage of nucleobases in the synthesized DNA11, our synthetic method will nevertheless be applicable for drug discovery applications using DELs.

Chart 2. Application of the Synthesized Model DEL

Conclusion

Using a new synthesis method that contributes to medicinal chemistry, an oligonucleotide library conjugated with coumarin derivatives was successfully developed. A coumarin scaffold could be constructed on DNA by treating malonate-conjugated DNA with salicylaldehyde and piperidine in an organic solvent. Importantly, most substituents of salicylaldehyde were tolerant to this method, and almost all substituents except a hydroxy group could be introduced at the 5-, 6-, 7-, and 8-positions of coumarin. Furthermore, the applicability of this method to DEL synthesis was demonstrated by synthesizing a model hairpin DNA possessing a coumarin derivative. Our synthetic approach for the construction of a coumarin scaffold based on Knoevenagel condensation may be helpful for drug discovery applications using DELs.

Experimental

General

For HPLC, the SHIMADZU CBM-20A, DGU-20A3, LC-20AT, CTO-20A, SPD-20A, and FRC-10A instruments were utilized. For HPLC purification, a Waters XBridge^® Oligonucleotide BEH C18 OBD™ prep column (130 Å, 2.5 µm, 10 × 50 mm) was used. For the HPLC analysis of the purified oligonucleotides, the Waters XBridge^® Shield RP18 (130 Å, 2.5 µm, 4.6 × 50 mm) was used. MALDI-TOF mass spectra of all the new oligonucleotides were recorded on a Bruker Daltonics Autoflex speed TOF mass spectrometer. The yields of the oligonucleotides were calculated by measuring their absorbances at 260 nm on a NanoDrop instrument (DeNovix DS-11). The oligonucleotides conjugated with an amino linker (DNA1, DNA4, DNA7, and barcode dsDNAs) were purchased from GeneDesign Inc.

Amide Bond Formation Using NHS-Ester 1

A 400 mM solution of NHS-ester 1 in N,N-dimethylformamide (DMF) (80 µL, 32 µmol) and N,N-diisopropylethylamine (DIPEA) (5 µL) was added to a 10 mM aqueous DNA solution (DNA1 and DNA4, 20 µL, 200 nmol) in a 0.5 mL tube. The reaction mixture was shaken at 50 °C for 18 h in a block bath shaker (1000 rpm). To precipitate the oligonucleotide, 3 M of sodium acetate solution (10 µL) and EtOH (300 µL) was added to the solution. After centrifugation, the supernatant was removed, and the resulting pellet was dissolved in 0.1 M aqueous triethylammonium acetate (TEAA) buffer (200 µL). The crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer (pH = 7.0) as eluent A and MeCN as eluent B. A linear gradient from 5 to 20% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min, and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized to obtain DNA2a (144 nmol, 72%) and DNA5 (140 nmol, 70%), respectively.

Amide Bond Formation Using NHS-Ester 2

A 100 mM solution of NHS-ester 2 in DMF (80 µL, 8.0 µmol) and DIPEA (5 µL) was added to a 10 mM aqueous DNA solution (DNA1, 20 µL, 200 nmol) in a 0.5 mL tube. The reaction mixture was shaken at 50 °C for 18 h in a block bath shaker (1000 rpm). To precipitate the oligonucleotide, 3 M of sodium acetate solution (10 µL) and EtOH (300 µL) was added to the solution. After centrifugation, the supernatant was removed, and the resulting pellet was dissolved in 0.1 M aqueous TEAA buffer (200 µL). The crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 5 to 20% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized to obtain DNA2b (100 nmol, 50%).

General Procedure for Knoevenagel Condensation

Solutions of 200 mM piperidine in EtOH (10 µL, 2 µmol) and 1 M salicylaldehyde derivatives in EtOH (10 µL, 10 µmol) were added to monoethyl malonate-conjugated DNA (DNA2a and DNA5, 20 nmol) in a 0.5 mL tube. Salicylaldehydes, which are poorly soluble in EtOH, were used after dissolving in DMF. The reaction mixture was shaken at 50 °C for 24 h in a block bath shaker (1000 rpm). To precipitate the oligonucleotide, 3 M of a sodium acetate solution (5 µL) and EtOH (80 µL) was added to the solution. After centrifugation, the supernatant was removed, and the resulting pellet was dissolved in 0.1 M aqueous TEAA buffer (100 µL). The crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 10 to 30% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized. The yields of the products (DNA3a and DNA6-1–29) are summarized in Table 1 and Fig. 2.

Synthesis of DNA8

A 100 mM sodium phosphate buffer (pH 7.4, 40 µL) and a 20 mM solution of compound 3 in DMF (80 µL, 1.6 µmol) were added to DNA7 (400 nmol) in a 0.5 mL tube. After the reaction mixture was shaken at 50 °C for 1 h in a block bath shaker (1000 rpm), compound 3 in DMF (80 µL, 1.6 µmol) was added. The reaction mixture was shaken at 50 °C for 1 h in a block bath shaker (1000 rpm), then diluted with 0.1 M TEAA (300 µL) prior to purification using an NAP-5 column (Cytiva). The obtained crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 10 to 30% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized. The obtained DNA was dissolved in 10% piperidine aq. (500 µL). After the reaction mixture was shaken at 30 °C for 0.5 h in a block bath shaker (1000 rpm), the resulting crude solution was purified by a NAP-5 column (Cytiva) to afford DNA8 (150 nmol, 38% for two steps).

Synthesis of DNA9

A 250 mM aqueous solution containing barcode dsDNA1 (168 µL, 84 nmol), 10× ligation buffer (70 µL), 500 unit/µL T4 ligase (14 µL, 7000 unit), and deionized water (280 µL) was added to DNA8 (70 nmol) in a 0.5 mL tube. After the reaction mixture was incubated at 20 °C for 24 h, the resulting crude solution was purified by a NAP-10 column (Cytiva). The fractions containing the desired product were collected and lyophilized. Deionized water (7 µL), a 400 mM solution of NHS-ester 1 in DMF (28 µL, 11.2 µmol), and DIPEA (1.4 µL) were added to the obtained DNA in a 0.5 mL tube. The reaction mixture was shaken at 50 °C for 18 h in a block bath shaker (1000 rpm). To precipitate the oligonucleotide, 3 M of a sodium acetate solution (3 µL) and EtOH (140 µL) was added to the solution. After centrifugation, the supernatant was removed and the resulting pellet was dissolved in 0.1 M aqueous TEAA buffer (500 µL). The crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 10 to 30% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized to obtain DNA9 (41 nmol, 58% for two steps).

Synthesis of DNA10

A 250 mM aqueous solution of barcode dsDNA2 (96 µL, 24 nmol), 10× ligation buffer (20 µL), 500 unit/µL T4 ligase (4 µL, 2000 unit), and deionized water (80 µL) was added to DNA9 (20 nmol) in a 0.5 mL tube. After the reaction mixture was incubated at 20 °C for 24 h, the resulting crude solution was purified by a NAP-5 column (Cytiva). The fractions containing the desired product were collected and lyophilized. A 1 M solution of 5-fluoro salicylaldehyde in EtOH (10 µL, 10 µmol) and a 200 M solution of piperidine in EtOH (10 µL, 2 µmol) were added to the obtained DNA in a 0.5 mL tube. The reaction mixture was shaken at 50 °C for 24 h in a block bath shaker (1000 rpm). To precipitate the oligonucleotide, 3 M of a sodium acetate solution (2 µL) and EtOH (80 µL) was added to the solution. After centrifugation, the supernatant was removed, and the resulting pellet was dissolved in 0.1 M aqueous TEAA buffer (500 µL). The crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 10 to 30% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized to afford DNA10 (7.2 nmol, 36% for two steps).

Synthesis of DNA11

A 250 mM aqueous solution of barcode dsDNA3 (9.6 µL, 2.4 nmol), 10× ligation buffer (2 µL), 500 unit/µL T4 ligase (0.4 µL, 200 unit), and deionized water (8 µL) was added to DNA10 (2 nmol) in a 0.5 mL tube. After the reaction mixture was incubated at 20 °C for 24 h, the resulting crude solution was purified by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. A linear gradient from 10 to 30% MeCN (over 30 min) was used at 50 °C at a flow rate of 3 mL/min and the process was monitored by UV observation at 260 nm. The fractions containing the desired product were collected and lyophilized to obtain DNA11 (1.9 nmol, 95%).

HPLC Analysis

The purity of the synthesized oligonucleotides was analyzed by reverse-phase HPLC using 0.1 M TEAA buffer as eluent A and MeCN as eluent B. Linear gradients from 5 to 20% MeCN (over 30 min, for DNA2a, DNA2b, DNA5, and DNA8) and from 10% to 30% MeCN (over 30 min, for DNA3a, DNA6-1–29a, and DNA9–11) were used at 50 °C at a flow rate of 1 mL/min, and the process was monitored by UV observation at 260 nm.

MALDI-TOF-MS Measurement

Approximately 1 µL of the matrix solution, which consisted of a saturated solution of 3-hydroxypicolinic acid in 50% MeCN in 0.1% TFA aq. containing 10 mg/mL of diammonium hydrogen citrate, was spotted on a stainless steel MALDI plate. After the plate was dried at room temperature, 1 µL of an aqueous solution of oligonucleotides was spotted over the matrix and dried again at room temperature. The oligonucleotides were detected in the negative ion mode.

Acknowledgments

This work was partially supported by the Japan Agency for Medical Research and Development (AMED) (Grant Numbers JP19am0401003, JP21ae0121022, JP21ae0121023, and JP21ae0121024).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Brenner S., Lerner R. A., Proc. Natl. Acad. Sci. U.S.A., 89, 5381–5383 (1992).
• 2) Gironda-Martínez A., Donckele E. J., Samain F., Neri D., ACS Pharmacol. Transl. Sci., 4, 1265–1279 (2021).
• 3) Huang Y., Li X., ChemBioChem, 22, 2384–2397 (2021).
• 4) Mullard A., Nature (London), 530, 367–369 (2016).
• 5) Shi Y., Wu Y., Yu J., Zhang W., Zhuang C., RSC Adv., 11, 2359–2376 (2021).
• 6) Fitzgerald P. R., Peagel B. M., Chem. Rev., 121, 7155–7177 (2021).
• 7) Estévez-Braun A., González A. G., Nat. Prod. Rep., 14, 465–475 (1997).
• 8) Vazquez-Rodriguez S., Matos M., Borges F., Uriarte E., Santana L., Curr. Top. Med. Chem., 15, 1755–1766 (2015).
• 9) Fasco M. J., Hildebrandt E. F., Suttie J. W., J. Biochem., 257, 11210–11212 (1982).
• 10) Garg S. S., Gupta J., Sharma S., Sahu D., Eur. J. Pharm. Sci., 152, 105424 (2020).
• 11) Nagy N., Kuipers H. F., Frymoyer A. R., Ishak H. D., Bollyky J. B., Wight T. N., Bollyky P. L., Front. Immunol., 6, 123 (2015).
• 12) Kumagai M., Watanabe A., Yoshida I., Mishima T., Nakamura M., Nishikawa K., Morimoto Y., Biol. Pharm. Bull., 41, 132–137 (2018).
• 13) Duval R., Duplais C., Nat. Prod. Rep., 34, 161–193 (2017).
• 14) Tasior M., Kim D., Singha S., Krzeszewski M., Ahn K. H., Gryko D. T., J. Mater. Chem. C, 3, 1421–1446 (2015).
• 15) Cao D., Liu Z., Verwilst P., Koo S., Jangjili P., Kim J. S., Lin W., Chem. Rev., 119, 10403–10519 (2019).
• 16) Bogdał D. J., Chem. Res. Synop., 1998, 468–469 (1998).
• 17) Vekariya R. H., Patel H. D., Synth. Commun., 44, 2756–2788 (2014).
• 18) Li T.-S., Zhang Z.-H., Yang F., Fu C.-G., J. Chem. Res. Synop., 38–39 (1998).
• 19) Khaligh N. G., Catal. Sci. Tech., 2, 1633–1636 (2012).
• 20) Vilar S., Quezada E., Santana L., Uriarte E., Yanez M., Fraiz N., Alcaide C., Cano E., Orallo F., Bioorg. Med. Chem. Lett., 16, 257–261 (2006).
• 21) Valizadeh H., Vaghefi S., Synth. Commun., 39, 1666–1678 (2009).
• 22) Yavari I., Hekmat-Shoar R., Zonouzi A., Tetrahedron Lett., 39, 2391–2392 (1998).
• 23) Kaye P. T., Musa M. A., Synthesis, 2701–2706 (2002).
• 24) Efcavitch J. W., Heiner C., Nucleosides Nucleotides Nucleic Acids, 4, 267 (1985).
• 25) Potowski M., Losch F., Wünnermann E., Dahmen J. K., Chines S., Brunschweiger A., Chem. Sci., 10, 10481–10492 (2019).
• 26) Aresu E., Fioravanti S., Gasbarri S., Pellacani L., Ramadori F., Amino Acids, 44, 977–982 (2013).
• 27) Gerry C. J., Yang Z., Stasi M., Schreiber S. L., Org. Lett., 21, 1325–1330 (2019).
• 28) Zhou L.-S., Yang K.-W., Feng L., Xiao J.-M., Liu C.-C., Zhang Y.-L., Crowder M. W., Bioorg. Med. Chem. Lett., 23, 949–954 (2013).
• 29) Dalessandro E. V., Collin H. P., Guimarães L. G. L., Valle M. S., Pliego J. R. Jr., J. Phys. Chem. B, 121, 5300–5307 (2017).
• 30) Corrie J. E., Munasinghe V. R. N., Rettig W., J. Heterocycl. Chem., 37, 1447–1455 (2000).
• 31) Takechi H., Oda Y., Nishizono N., Oda K., Machida M., Chem. Pharm. Bull., 48, 1702–1710 (2000).
• 32) Chen J., Liu W., Ma J., Xu H., Wu J., Tang X., Fan Z., Wang P., J. Org. Chem., 77, 3475–3482 (2012).
• 33) Saimoto H., Yoshida K., Murakami T., Morimoto M., Sashiwa H., Shigemasa Y., J. Org. Chem., 61, 6768–6769 (1996).
• 34) Bharate S. B., Bhutani K. K., Khan S. I., Tekwani B. L., Jacob M. R., Khan I. A., Singh I. P., Bioorg. Med. Chem., 14, 1750–1760 (2006).
• 35) Saimoto H., Kido Y., Haga Y., Sakamoto K., Kita K., J. Biochem., 153, 267–273 (2013).
• 36) Quesneau V., Roubinet B., Renard P., Romieu A., Tetrahedron Lett., 60, 151279 (2019).
• 37) Choong I., Burdett M., Delano W., Eranson D. A., International Patent application PCT/US02/25936 (2003).
• 38) Gates K. S., Chem. Res. Toxicol., 22, 1747–1760 (2009).
• 39) Ratnayake A. S., Flanagan M. E., Foley T. L., Smith J. D., Johnson J. G., Bellenger J., Montgomery J. I., Paegel B. M., ACS Comb. Sci., 21, 650–655 (2019).