Synthesis of New Azetidine and Oxetane Amino Acid Derivatives through Aza-Michael Addition of NH-Heterocycles with Methyl 2-(Azetidin- or Oxetan-3-Ylidene)Acetates

In this paper, a simple and efficient synthetic route for the preparation of new heterocyclic amino acid derivatives containing azetidine and oxetane rings was described. The starting (N-Boc-azetidin-3-ylidene)acetate was obtained from (N-Boc)azetidin-3-one by the DBU-catalysed Horner–Wadsworth–Emmons reaction, followed by aza-Michael addition with NH-heterocycles to yield the target functionalised 3-substituted 3-(acetoxymethyl)azetidines. Methyl 2-(oxetan-3-ylidene)acetate was obtained in a similar manner, which was further treated with various (N-Boc-cycloaminyl)amines to yield the target 3-substituted 3-(acetoxymethyl)oxetane compounds. The synthesis and diversification of novel heterocyclic amino acid derivatives were achieved through the Suzuki–Miyaura cross-coupling from the corresponding brominated pyrazole–azetidine hybrid with boronic acids. The structures of the novel heterocyclic compounds were confirmed via 1H-, 13C-, 15N-, and 19F-NMR spectroscopy, as well as HRMS investigations.


Introduction
In heterocyclic chemistry, four-membered saturated heterocycles containing one nitrogen or oxygen atom are known as azetidines and oxetanes, respectively [1,2]. The pharmacophore subunit of azetidine in aza-heterocyclic molecules is used for a wide variety of natural and synthetic products exhibiting a variety of biological activities [3,4]. For example, the azetidine subunit is a structure derived from some alkaloids from marine sources, which show relatively potent cytotoxic activity against tumour cells as well as antibacterial activity [5,6]. An azetidine ring is also present in the molecular structure of the well-known antihypertensive drug azelnidipine, which is a dihydropyridine calcium channel blocker [7,8].
Azetidine carboxylic acids are important scaffolds and building blocks for obtaining various biologically active heterocyclic compounds and peptides [9][10][11]. Specifically, L-azetidine-2-carboxylic acid is found in nature in sugar beets (Beta vulgaris) and is a gametocidal agent [12]. In addition, this amino acid is an inhibitor of collagen synthesis that is antiangiogenic [13]. Azetidine-2-carboxylic acid (I) and its 3-aryl derivatives, which are L-proline analogues, have also been widely used as building blocks to prepare small peptides ( Figure 1) [14,15]. Additionally, both azetidine-3-carboxylic (II) and 3-(4-oxymethylphenyl)azetidine-3-carboxylic (III) acids, which are conformationally constrained analogues of β-proline, were employed for the preparation of endomorphin (Azetidin-3-yl)acetic acid VIII could be used as a structural analogue for 4-aminobutanoic acid (GABA) [9]. (3-Arylazetidin-3-yl)acetates IX and X are used for the preparation of pharmaceutically active agents, including the positive allosteric modulators of GABAA receptors [20]. Chalyk et al. developed a general method for isoxazole-containing building blocks, namely azetidine amino ester XI as a 5-aminopentanoic acid (δ-aminovaleric acid) ester analogue [21]. 5-Aminopentanoic acid is a naturally occurring amino acid and a methylene homologue of GABA [22]. Recently, we developed efficient protocols that provide easy access to highly functional heterocyclic compounds by combining heterocyclic moieties with both carboxylic ester functional groups and cycloaminyl units, such as the δ-amino esters azetidine derivatives XII and XIII [23,24].
The pharmacophoric subunit of oxetane, containing various organic compounds, has been extensively studied in medicinal chemistry [25]. This oxetane ring structure is widespread in natural products and has been found to exhibit a number of biological activities. Oxetin, i.e., 3-amino-2-oxetanecarboxylic acid XIV, was isolated from the broth of Streptomyces and has been shown to possess antibacterial and herbicidal effects ( Figure 2) [26]. (Azetidin-3-yl)acetic acid VIII could be used as a structural analogue for 4-aminobutanoic acid (GABA) [9]. (3-Arylazetidin-3-yl)acetates IX and X are used for the preparation of pharmaceutically active agents, including the positive allosteric modulators of GABA A receptors [20]. Chalyk et al. developed a general method for isoxazole-containing building blocks, namely azetidine amino ester XI as a 5-aminopentanoic acid (δ-aminovaleric acid) ester analogue [21]. 5-Aminopentanoic acid is a naturally occurring amino acid and a methylene homologue of GABA [22]. Recently, we developed efficient protocols that provide easy access to highly functional heterocyclic compounds by combining heterocyclic moieties with both carboxylic ester functional groups and cycloaminyl units, such as the δ-amino esters azetidine derivatives XII and XIII [23,24].
The pharmacophoric subunit of oxetane, containing various organic compounds, has been extensively studied in medicinal chemistry [25]. This oxetane ring structure is widespread in natural products and has been found to exhibit a number of biological activities. Oxetin, i.e., 3-amino-2-oxetanecarboxylic acid XIV, was isolated from the broth of Streptomyces and has been shown to possess antibacterial and herbicidal effects ( Figure 2) [26]. The oxetane subunit is a structure derived from natural or synthetic taxanes clinically used in cancer chemotherapy [27]. Notably, the oxetane nucleoside of antibiotic oxetanocin A, isolated from natural sources, inhibits the replication of the human immun-

Results and Discussion
The strategy for the synthesis of novel heterocyclic amino acids containing azetidin rings is outlined in Scheme 1. The synthetic sequence began with methyl (N-Boc-azetidin 3-ylidene)acetate 3, prepared from azetidin-3-one 2 through the Horner-Wadsworth-Em mons (HWE) reaction. The HWE reaction is one of the most reliable and common syn thetic methods for preparing substituted alkene products from aldehydes and ketone with phosphonate esters [35]. Yang et al. recently developed a simple method for the prep aration of compound 3 from methyl 2-(dimethoxyphosphoryl)acetate 1 with a 60% sus pension of NaH in mineral oil in dry THF, followed by the addition of azetidin-2-one 2 The reaction was then quenched with water, and the resulting aqueous solution was ex tracted with EtOAc and concentrated in vacuo; finally, the residue was purified via flas column chromatography [36]. We carried out a similar synthesis for 3, but this metho differed in that the corresponding residue was purified through two-stage vacuum disti lation in a Büchi oven (kugelrohr) [37] at a reduced pressure of 4 × 10 -3 bar by first distillin the volatile fraction at 90 °C for some time (usually approximately 1 h) and then changing th collection vessel and increasing the temperature to 130 °C to produce the pure product 3 (yiel 72%). This method for the preparation of compound 3 allows the purification of large quant ties and works well while trying to avoid stubborn impurities such as mineral oil. Carreira et al. investigated various properties of oxetanes as substituents, leading to many useful developments, especially in the use of oxetanes as substitutes for carbonyl groups, which is of considerable interest due to their similar dipoles and H-bonding ability [30,31]. Powell et al. reported the preparation of derivatives in which the central C=O amide bond of a tripeptide was replaced by the oxetane nucleus [32]. Several reports are devoted to the synthesis and evaluation of the physicochemical and metabolic properties of δ-amino acid oxetane derivatives, such as compound XVI [33].
This study aimed to develop and synthesise new heterocyclic amino acid derivatives containing azetidine and oxetane rings. Such amino acid compounds offer valuable properties as isosteres, new conformationally restricted amino acids, and building blocks that can be used as potentially biologically active substances and peptides, as well as for the generation of DNA-encoded peptide libraries [34].

Results and Discussion
The strategy for the synthesis of novel heterocyclic amino acids containing azetidine rings is outlined in Scheme 1. The synthetic sequence began with methyl (N-Boc-azetidin-3-ylidene)acetate 3, prepared from azetidin-3-one 2 through the Horner-Wadsworth-Emmons (HWE) reaction. The HWE reaction is one of the most reliable and common synthetic methods for preparing substituted alkene products from aldehydes and ketones with phosphonate esters [35]. Yang et al. recently developed a simple method for the preparation of compound 3 from methyl 2-(dimethoxyphosphoryl)acetate 1 with a 60% suspension of NaH in mineral oil in dry THF, followed by the addition of azetidin-2-one 2. The reaction was then quenched with water, and the resulting aqueous solution was extracted with EtOAc and concentrated in vacuo; finally, the residue was purified via flash column chromatography [36]. We carried out a similar synthesis for 3, but this method differed in that the corresponding residue was purified through two-stage vacuum distillation in a Büchi oven (kugelrohr) [37] at a reduced pressure of 4 × 10 -3 bar by first distilling the volatile fraction at 90 • C for some time (usually approximately 1 h) and then changing the collection vessel and increasing the temperature to 130 • C to produce the pure product 3 (yield 72%). This method for the preparation of compound 3 allows the purification of large quantities and works well while trying to avoid stubborn impurities such as mineral oil. Next, having obtained α,β−unsaturated ester 3, aza-Michael addition was carried out with heterocyclic aliphatic and heterocyclic aromatic amines for the formation of heterocyclic amino acid blocks 4. Aza-Michael addition is a powerful and versatile method for constructing C-N bonds containing various highly functional organic compounds, which has remained an important challenge over the last decade [38,39]. In particular, this synthetic strategy has been applied for the preparation of NH-heterocyclic derivatives, such Scheme 1. Synthesis of methyl (1-Boc-3-cycloaminylazetidin-3-yl)acetates 4a-p.
Methyl (N-Boc-azetidin-3-ylidene)acetate 3 was reacted (Scheme 1) with azetidine and DBU in the solvent acetonitrile at 65 • C for 4 h to obtain 1,3 -biazetidine 4a with a 64% yield. Compound 4a was subjected to a detailed spectral analysis. Absorption bands characteristic of the esters at 1731 (C=O, ester) and 1694 (C=O, Boc) cm −1 were observed on the IR spectrum of compound 4a. In the 1 H-NMR spectrum of compound 4a, four characteristic methylene protons CH 2 -2,4 were observed in the regions of δ 3.69-3.86 and 3.94-4.06 ppm, which appeared significantly broadened due to the conformational dynamics of the 3,3-substituted azetidine moiety in the solvent. The second azetidine ring, containing the symmetric fragment CH 2 CH 2 CH 2 , showed methylene protons CH 2 -2 ,4 appearing as a triplet at δ 3.29 ( 3 J = 7.2 Hz) ppm, while two protons CH 2 -3 appeared as a pentet at δ 2.05 ( 3 J = 7.2 Hz) ppm. The 1 H-15 N HMBC spectrum of 4a showed the characteristic resonances of the nitrogen atoms of the azetidine rings at δ −315.4 (N-1, Boc-azetidine) and −337.8 ppm (N-1 , azetidine), respectively. 3-Hydroxy-1,3 -biazetidine 4b was synthesised by analogy to 4a from 3-hydroxyazetidine by aza-Michael addition with a 62% yield. The key information for structure elucidation was also obtained from the 1 H-15 N HMBC spectrum. As expected, the 15 N chemical shifts of the N-1 Boc-azetidine (δ −315.0 ppm) and N-1 azetidine (δ −350.2 ppm) atoms were highly comparable to those of compound 4a. The observed chemical shifts of the azetidine derivatives were consistent with the data reported in the literature [23,24,58,59].
The reaction of 3 with pyrrolidine under these conditions resulted in compound 4c with a 61% yield, while the obtained 3,3-difluoropyrrolidine led to compound 4d with a 64% yield. Although the basicity of 3,3-difluoropyrrolidine was significantly lower (pKa 7.5) than that of pyrrolidine (pKa 11.3), it did not affect the reaction in any way [60]. 1-(Azetidin-3-yl)piperidine 4e was isolated from reaction 3 with piperidine with a 75% yield. 3-(4-Hydroxypiperidin-1-yl)azetidines 4f and 4g were formed from either 4-hydroxypiperidine or 4-hydroxy-4-phenylpiperidine through aza-Michael addition with 75% and 66% yields, respectively. 1 H-15 N HMBC experiments for the products 4f and 4g confirmed the proposed structures of the isomeric piperidines, as the Boc-azetidine-ring protons H-2 and H-4 showed interactions with the nitrogens N-1 of the piperidine rings at δ -324.2 ppm and -324.8 ppm, respectively. When the 2,3-unsaturated ester 3 was used with morpholine, adduct 4h was obtained with a 73% yield after 4 h. It was observed that increasing the size of the heterocyclic aliphatic amines from a four-to a six-membered ring system did not adversely affect the reaction, and all adducts were obtained in moderate-to-good yields. A similar reaction was carried out with 2,3-dihydro-1H-isoindoline, which generated compound 4i with a 64% yield.
The reaction of 2,3-unsaturated ester 3 with 3-(3-trifluoromethyl)-1H-pyrazole could yield regioisomers 4l and A, but only compound 4l was obtained ( Figure 3). The regiochemistry of compound 4l was confirmed with a NOESY experiment, which exhibited NOEs between the pyrazole proton 5 -H and the azetidine 2(4)-H a protons. In the case of compound A, it would not be possible to have NOEs between the protons of the pyrazole and azetidine moieties. In addition, the 1 H-13 C HMBC spectrum of the molecule 4l showed a long-range correlation between the 5 -H pyrazole proton (δ 7.72 ppm) and the quaternary carbon of azetidine C-3 at δ 57.8 ppm, as well as a three-bond correlation with the quaternary carbon of pyrazole C-3 at δ 143.0 (q, 2 J CF = 38.4 Hz) ppm [65].
In the 1 H-NMR spectrum of the 3-(pyrazol-1-yl)azetidine derivative 4j, the methylene protons from the azetidine moiety appeared as two doublets resonating at δ 4.28 and 4.42 ppm ( 2 JHa,Hb = 9.6 Hz), while the aromatic pyrazole protons showed three signals at δ 6.29 The reaction of 2,3-unsaturated ester 3 with 3-(3-trifluoromethyl)-1H-pyrazole could yield regioisomers 4l and A, but only compound 4l was obtained ( Figure 3). The regiochemistry of compound 4l was confirmed with a NOESY experiment, which exhibited NOEs between the pyrazole proton 5′-H and the azetidine 2(4)-Ha protons. In the case of compound A, it would not be possible to have NOEs between the protons of the pyrazole and azetidine moieties. In addition, the 1 H-13 C HMBC spectrum of the molecule 4l showed a long-range correlation between the 5′-H pyrazole proton (δ 7.72 ppm) and the quaternary carbon of azetidine C-3 at δ 57.8 ppm, as well as a three-bond correlation with the quaternary carbon of pyrazole C-3′ at δ 143.0 (q, 2 JCF = 38.4 Hz) ppm [65]. In principle, the use of indazole as an aza-Michael donor can result in the formation of two additional products, N-1 and N-2 adducts, because of its tautomerism. However, Jiang et al. successfully developed an efficient method for the synthesis of the desired 1substituted 1H-indazole compound with a 52% yield through the direct aza-Michael addition of indazole to an α,β-unsaturated malonate compound using Cs2CO3 as a catalyst [66]. Recently, Yang et al. reported a synthetic approach for synthesising 1-substituted 1Hindazoles via the DBU-catalysed aza-Michael reaction of 1H-indazole with enones. This reaction produced regioselective compounds with good substrate tolerance, mild reaction conditions, and high-to-excellent yields (up to 93%) [45].
We investigated the aza-Michael reaction of indazole with methyl (N-Boc-azetidine-3-ylidene)acetate 3 for possible regioisomers. The reaction was monitored via LC/MS, and the full conversion of the starting materials was observed after 16 h. The reaction of the In principle, the use of indazole as an aza-Michael donor can result in the formation of two additional products, N-1 and N-2 adducts, because of its tautomerism. However, Jiang et al. successfully developed an efficient method for the synthesis of the desired 1-substituted 1H-indazole compound with a 52% yield through the direct aza-Michael addition of indazole to an α,β-unsaturated malonate compound using Cs 2 CO 3 as a catalyst [66]. Recently, Yang et al. reported a synthetic approach for synthesising 1-substituted 1H-indazoles via the DBU-catalysed aza-Michael reaction of 1H-indazole with enones. This reaction produced regioselective compounds with good substrate tolerance, mild reaction conditions, and high-to-excellent yields (up to 93%) [45].
We investigated the aza-Michael reaction of indazole with methyl (N-Boc-azetidine-3-ylidene)acetate 3 for possible regioisomers. The reaction was monitored via LC/MS, and the full conversion of the starting materials was observed after 16 h. The reaction of the starting materials in DBU in the solvent acetonitrile at 65 • C led to regioisomer 4m as the sole product with a moderate 69% isolated yield. The unambiguous formation of 4m was easily deduced from 1 H-15 N HMBC spectral data, as it clearly showed a strong three-bond correlation between the indazole nitrogen N-1 (δ -190.8 ppm) with indazole 3 -H (δ 7.99 ppm) and 7 -H (δ 7.39 ppm) protons and azetidine methylene CH 2 -2,4 (δ 4.76 ppm) protons, correspondingly. The regiochemistry of compound 4m was confirmed with a NOESY experiment, which exhibited NOEs between the pyrazole proton 7 -H and the azetidine 2(4)-H a protons ( Figure 3).
In the present work, the treatment of 1,2,4-triazole with α,β-unsaturated ester 3 was carried out in acetonitrile in the presence of DBU to generate compound 4q as a single product but with a low percentage yield of 46% (Table 1, Entry 1). Therefore, the reaction conditions were optimised. It was found that the target product 4q was not formed in the absence of a catalyst (Entry 2). Some salts such as LiF and LiCl did not affect the reaction (Entries 3,4). However, inorganic bases such as Cs2CO3, KOAc and K3PO4 produced 4q with a moderate yield (Entries 5-7). The highest yield of 4q was obtained in the presence of K2CO3 in acetonitrile (Entry 8). Additionally, the effect of the solvents on the reaction was evaluated (Entries 9-11). Ethanol proved to be less effective on the assessed reaction with a final yield of only 44%. Although 1,4-dioxane achieved a 60% product yield and proved to be efficient, MeCN was still better, reaching a 65% yield for the same reaction.  In the present work, the treatment of 1,2,4-triazole with α,β-unsaturated ester 3 was carried out in acetonitrile in the presence of DBU to generate compound 4q as a single product but with a low percentage yield of 46% (Table 1, Entry 1). Therefore, the reaction conditions were optimised. It was found that the target product 4q was not formed in the absence of a catalyst (Entry 2). Some salts such as LiF and LiCl did not affect the reaction (Entries 3,4). However, inorganic bases such as Cs 2 CO 3 , KOAc and K 3 PO 4 produced 4q with a moderate yield (Entries 5-7). The highest yield of 4q was obtained in the presence of K 2 CO 3 in acetonitrile (Entry 8). Additionally, the effect of the solvents on the reaction was evaluated (Entries 9-11). Ethanol proved to be less effective on the assessed reaction with a final yield of only 44%. Although 1,4-dioxane achieved a 60% product yield and proved to be efficient, MeCN was still better, reaching a 65% yield for the same reaction. The regiochemistry of compound 4q was confirmed with a NOESY experiment, which exhibited NOEs between the azetidine protons 2(4)-H a at δ 4.43 ppm and the 1,2,4-triazole proton 5-H at δ 8.33 ppm. The 1 H-15 N HMBC experiment revealed the corresponding threebond connectivities of azetidine methylene CH 2 -2,4 and acetate methylene CH 2 protons with the 1,2,4-triazole nitrogen N-1 (pyrrole-type) at δ -156.5 ppm (Figure 4a).
One of the most effective methods for the structural diversification of aromatic and heterocyclic building blocks is functionalisation through Pd-catalysed Suzuki-Miyaura cross-coupling reactions [69]. With compound 4k containing a 4-bromopyrazole moiety in hand, we further investigated Pd-catalysed coupling with organoboronic acids (Scheme 3). Several coupling systems were evaluated, namely Pd(dba)2-K3PO4 in DCM [70], The aza-Michael reaction of benzotriazole with methyl acrylate was reported to form N-1 and N-2 adducts in a mixture of benzotriazol-1-yl-propionic and benzotriazol-2-ylpropionic acid methyl esters by using anhydrous potassium phosphate (K 3 PO 4 ) as a catalyst [54], while the 1,4-conjugated aza-Michael addition of benzotriazole to dienones catalysed by potassium acetate (KOAc) yielded only the corresponding N-1 isomer [56]. Recently, Chen et al. reported an efficient, regio-and enantioselective aza-Michael reaction for the synthesis of the N-1 isomers from benzotriazole with α-substituted β-nitroacrylates catalysed by a chiral organocatalyst [49].
One of the most effective methods for the structural diversification of aromatic and heterocyclic building blocks is functionalisation through Pd-catalysed Suzuki-Miyaura cross-coupling reactions [69]. With compound 4k containing a 4-bromopyrazole moiety in hand, we further investigated Pd-catalysed coupling with organoboronic acids (Scheme 3). Several coupling systems were evaluated, namely Pd(dba) 2 -K 3 PO 4 in DCM [70], Pd(OAc) 2 -Cs 2 CO 3 in EtOH/H 2 O [71], Pd(PPh 3 ) 4 -K 2 CO 3 in toluene/MeOH [72], and Pd(PPh 3 ) 4 -K 3 PO 4 in 1,4-dioxane [73]. The best Suzuki-Miyaura cross-coupling result was achieved using Pd(PPh 3 ) 4 as a catalyst, using K 3 PO 4 as a base, and performing the reaction at 100 • C in 1,4-dioxane [24]. Under these conditions, the target product 5a was obtained with an excellent (94%) yield. Compounds 5b-f were obtained from the aforementioned compound 4k with methylphenyl-and methoxyphenyl-boronic acids through Pd-catalysed coupling with the same reagents under analogous conditions as those used for compound 5a. Target products 5b-e were achieved with moderate yields of 70-80%, while the target product 5f was obtained with a low yield (29%). Methoxyphenylboronic acids with a substituent in the ortho position reacted less efficiently than those with a substituent in the meta and/or para position. Target compounds 5g and 5h, containing fluoro and chloro moieties, were obtained with fair yields of 63% and 53%, respectively. Compound 4k was reacted with pyridinyl-and thienylboronic acid, generating products 5i-m with 37-64% yields.
Following the successful completion of the aza-Michael addition reactions resulting in amino acid-like blocks containing an azetidine core, we explored another four-membered heterocycle: oxetane. The target oxetane compounds were synthesised as depicted in Scheme 4. The starting oxetan-3-one 6 was used in the Horner-Wadsworth-Emmons Scheme 3. Synthesis of compounds 5a-n via Suzuki-Miyaura cross-coupling reactions.
Next, we explored the Suzuki-Miyaura cross-coupling reaction with cyclopropylboronic acid (Scheme 3). The cyclopropyl group is an increasingly common structural motif in pharmaceutically active molecules [74]. The synthesis of compound 5n was carried out with cyclopropylboronic acid under the same reaction conditions as described above, using the Pd(PPh 3 ) 4 -K 3 PO 4 system and refluxing the reaction in 1,4-dioxane, but the desired product was not obtained. Wallace et al. demonstrated a synthetic approach where aryl bromides reacted with cyclopropylboronic acid in a Pd(PPh 3 ) 4 -K 3 PO 4 system in toluene [75]. Changing the solvent to toluene had a significant effect on the reaction and, finally, the target product 5n was obtained, but only with a 31% yield; however, the cross-coupling reaction in the P(cHex) 3 -Pd(OAc) 2 -K 3 PO 4 system in toluene afforded compound 5n with a sufficient (64%) yield [76].
The structures of synthesised compounds 5a-n were confirmed using NMR spectroscopic methods. For example, in the 1 H-NMR spectrum of compound 5a, two proton singlets from the pyrazole ring were exhibited at δ 7.81 (5-H) and 7.74 (3-H) ppm, while five protons of the phenyl ring appeared at δ 7.17-7.41 ppm. The azetidine-ring signals of the diastereotopic methylene protons were observed as two doublets at δ 4.24 and 4.40 ppm ( 2 J Ha,Hb = 9.6 Hz). In the 1 H-NMR spectrum of compounds 5b and 5c, the characteristic methyl protons from the methylphenyl moiety appeared in the regions of δ 2.28 and 2.31 ppm, respectively, while the methyl protons from the methoxyphenyl moiety in compounds 5d-5f appeared in the region of δ 3.75-3.84 ppm. The 1 H-NMR spectroscopic data for the methoxy-substituted derivatives 5d-5f showed an evident effect of this group on the chemical shift of the aromatic protons [77]. For example, in the case of compound 5d, the aromatic protons (3-H, 5-H) located at the ortho position to the 4-methoxy group were observed upfield (δ 6.83 ppm, d, J = 8. 8 Hz), whereas the protons located at the meta position (2-H, 6-H) were observed downfield (δ 7.32 ppm, d, J = 8. 8 Hz). 15 N-NMR spectroscopic data for the methoxy-substituted derivative 5k showed a significant effect on the 15 N chemical shift of the pyridin-3-yl moiety (δ -116.0 ppm), which was greatly shifted upfield compared with 5j (δ -69.5 ppm).
The 1 H-NMR spectrum of compound 5n showed characteristic resonances for the cyclopropyl moiety, where the methylene protons appeared as multiplets at δ 0.41-0.45 and 0.74-0.78 ppm, and the methine proton appeared as a multiplet at δ 1.58-1.62 ppm. A comparison between the DEPT-90, DEPT-135, and 13 C-NMR spectra of compound 5n clearly indicated the characteristic signals of the cyclopropyl-ring skeleton carbons, namely the methine carbon C-1 (δ 5.3 ppm) and methylene carbons C-2 and C-3 (δ 7.6 ppm) [78].
Following the successful completion of the aza-Michael addition reactions resulting in amino acid-like blocks containing an azetidine core, we explored another fourmembered heterocycle: oxetane. The target oxetane compounds were synthesised as depicted in Scheme 4. The starting oxetan-3-one 6 was used in the Horner-Wadsworth-Emmons reaction with methyl-2-(dimethoxyphosphoryl)acetate 1 to obtain methyl (oxetan-3-ylidene)acetate 7 with a 73% yield according to a procedure similar to that described in the patent literature [79]. With compound 7 in our hands, compound 8a was easily synthesised from 3-N-Boc-aminoazetidine hydrochloride with a 71% yield. The reaction was carried out at 45 • C for 24 h in acetonitrile in the presence of DBU. reaction with methyl-2-(dimethoxyphosphoryl)acetate 1 to obtain methyl (oxetan-3-ylidene)acetate 7 with a 73% yield according to a procedure similar to that described in the patent literature [79]. With compound 7 in our hands, compound 8a was easily synthesised from 3-N-Boc-aminoazetidine hydrochloride with a 71% yield. The reaction was carried out at 45 °C for 24 h in acetonitrile in the presence of DBU. The structure of compound 8a was confirmed through spectral investigations. The IR spectrum of compound 8a revealed a N-H stretching vibration band at 3314 cm −1 and a C=O stretching vibration band at 1719 cm −1 (CH2C=O and Boc groups). The 1 H-NMR spectrum of compound 8a showed characteristic resonance for the Boc-group methyl protons as a singlet at δ 1.44 ppm, and the ester CH3O protons appeared as a singlet that overlapped with the protons of the azetidine moiety at δ 3.64-3.71 ppm. The signals of the oxetane ring for the diastereotopic protons of both methylene groups (CH2-2,4) were observed as two doublets at δ 4.57 and 4.71 ppm ( 2 JHa,Hb = 7.2 Hz). Both methylene protons (C′H2-2,4) of the azetidine ring showed signals in the form of broadened multiplets in the region of δ 3.11-3.24 and 3.64-3.71 ppm, and the methine proton (C′H-3) appeared as a multiplet in the region of δ 4.10-4.35 ppm. A relatively broad peak of the NH proton was observed at δ 4.98-5.06 ppm. The 1 H-15 N HSQC experiment indicated that the aforementioned proton had one-bond connectivity with the nitrogen NH-Boc at δ -288.5 ppm, while the oxetane-ring protons showed a 1 H-15 N HMBC correlation with azetidine nitrogen at δ -348.4 ppm. The 13 C-NMR spectrum of 8a revealed the characteristic signals of the oxetane-ring skeleton carbons at δ 76.0 (C-2,4) and 61.7 (C-3) ppm, while the azetidinering skeleton carbons resonated at δ 55.7 (C′-2,4) and 40.7 (C′-3) ppm.
In addition, compounds 8b-e were obtained from 2,3-unsaturated ester 7 with saturated chiral cyclic amines-namely, (S)-and (R)-3-(Boc-amino)pyrrolidines, and (S)-and The structure of compound 8a was confirmed through spectral investigations. The IR spectrum of compound 8a revealed a N-H stretching vibration band at 3314 cm −1 and a C=O stretching vibration band at 1719 cm −1 (CH 2 C=O and Boc groups). The 1 H-NMR spectrum of compound 8a showed characteristic resonance for the Boc-group methyl protons as a singlet at δ 1.44 ppm, and the ester CH 3 O protons appeared as a singlet that overlapped with the protons of the azetidine moiety at δ 3.64-3.71 ppm. The signals of the oxetane ring for the diastereotopic protons of both methylene groups (CH 2 -2,4) were observed as two doublets at δ 4.57 and 4.71 ppm ( 2 J Ha,Hb = 7.2 Hz). Both methylene protons (C H 2 -2,4) of the azetidine ring showed signals in the form of broadened multiplets in the region of δ 3.11-3.24 and 3.64-3.71 ppm, and the methine proton (C H-3) appeared as a multiplet in the region of δ 4.10-4.35 ppm. A relatively broad peak of the NH proton was observed at δ 4.98-5.06 ppm. The 1 H-15 N HSQC experiment indicated that the aforementioned proton had one-bond connectivity with the nitrogen NH-Boc at δ -288.5 ppm, while the oxetane-ring protons showed a 1 H-15 N HMBC correlation with azetidine nitrogen at δ -348.4 ppm. The 13 C-NMR spectrum of 8a revealed the characteristic signals of the oxetane-ring skeleton carbons at δ 76.0 (C-2,4) and 61.7 (C-3) ppm, while the azetidine-ring skeleton carbons resonated at δ 55.7 (C -2,4) and 40.7 (C -3) ppm.

General Information
All starting materials were purchased from commercial suppliers and were used as received. Flash column chromatography was performed on Silica Gel 60 Å (Merck KGaA, Darmstadt, Germany). Vacuum distillation was performed in a Büchi Model B580 GKR oven (Büchi Labortechnik AG, Flawil, Switzerland). Thin-layer chromatography was carried out on Silica Gel plates (Merck Kieselgel 60 F254) and visualised by UV light (254 nm) (Merck KGaA, Darmstadt, Germany). Melting points were determined using a Büchi M-565 melting point apparatus (Büchi Labortechnik AG, Flawil, Switzerland) and were uncorrected. The IR spectra were recorded on a Bruker Vertex 70v FT-IR spectrometer (Bruker Optik GmbH, Ettlingen, Germany) using neat samples and are reported in the frequency of absorption (cm -1 ). Mass spectra were obtained using a Shimadzu LCMS-2020 (ESI+) spectrometer (Shimadzu Corporation, Kyoto, Japan). High-resolution mass spectra were measured using a Bruker MicrOTOF-Q III (ESI+) apparatus (Bruker Daltonik GmbH, Bremen, Germany). Accurate measurements were achieved using the internal mass calibration of each sample using sodium formate calibration solution as a standard procedure, with a standard deviation always less than 1 ppm [80]. In addition, all data files were recalibrated with an internal standard of sodium formate injected prior to initial sample elution for each sample. Optical rotation data were recorded on a UniPol L SCHMIDT+HAENSCH polarimeter (concentration of compound (g/100 mL) and were included in calculations automatically (Windaus-Labortechnik GmbH & Co. KG, Clausthal-Zellerfeld, Germany). 1 H-NMR and 13 C-NMR spectra were recorded from CDCl 3 solutions at 25 • C on a Bruker Avance III 400 instrument (400 MHz for 1 H, 100 MHz for 13 C) using a directly detecting BBO probe (Bruker BioSpinInternational AG, Faellanden, Switzerland) and a Bruker Avance III 700 instrument (700 MHz for 1 H, 176 MHz for 13 C) equipped with a 5 mm TCI 1 H-13 C/ 15 N/D z-gradient cryoprobe (Bruker BioSpin GmbH, Rheinstetten, Germany). 19 F-NMR spectra (376.46 MHz, absolute referencing via Ξ ratio) were obtained on a Bruker Avance III 400 instrument with a directly detecting broadband observe probe (BBO). 15 N-NMR spectra were recorded from CDCl 3 solutions at 25 • C on either a Bruker Avance III 400 instrument (40 MHz for 15 N) using a directly detecting BBO probe or on a Bruker Avance III 700 instrument (71 MHz for 15 N). The chemical shifts (δ), expressed in ppm, were relative to tetramethylsilane (TMS). 15 N-NMR spectra were referenced against neat external nitromethane (coaxial capillary). The following abbreviations were used in reporting the NMR data: Az, azetidine; Cpr, cyclopropane; i-Ind, iso-Indoline; Morph, morpholine; Ox, oxetane; Pip, piperidine; Ph, phenyl; Pyr, pyridine; Pyrr, pyrrolidine; Prz, pyrazole; Idz, indazole; Imid, imidazole; Bim, benzimidazole; Ind, indole; Btz, benzotriazole; Trz, triazole; Thio, thiophene.

Synthetic Procedures
Neat methyl 2-(dimethoxyphosphoryl)acetate 1 (13.8 g, 76 mmol) was added to a suspension of NaH (60% dispersion in mineral oil) (3.12 g, 78 mmol) in dry THF (250 mL). After 30 min, a solution of 1-Boc-3-azetidinone 2 (13.0 g, 76 mmol) in dry THF (50 mL) was added, and the resulting mixture was stirred for 1 h. The reaction was quenched by the addition of water (250 mL). The organic layer was separated, and the aqueous one was extracted with ethyl acetate (3 × 150 mL). The combined organic solutions were dried over anhydr. Na 2 SO 4 , and then the solvents were removed under reduced pressure. Purification was conducted at 130 • C and 4 × 10 −3 bar pressure via distillation in vacuo to give 3
3.2.5. Methyl(oxetan-3-ylidene)acetate (7) Neat trimethyl phosphonoacetate 2 (3.79 g, 20.8 mmol) was added to a cooled (0 • C) suspension of NaH (60% dispersion in mineral oil) (0.83 g, 20.8 mmol) in dry THF (70 mL). After 20 min, a solution of 3-oxetanone 6 (1.50 g, 20.8 mmol) in dry THF (20 mL) was added, and the resulting mixture was stirred for 1 h. The reaction was quenched by the addition of water (70 mL). The organic layer was separated, and the aqueous one was extracted with ethyl acetate (3 × 70 mL). The combined organic solutions were dried over anhydr. Na 2 SO 4 , and then the solvents were removed under reduced pressure. The obtained residue was purified by flash chromatography (eluent n-hexane/acetone, v/v, 4:1) to give 7 (1.95 g, 73%) as a white solid, mp 50.8-51. 3.2.6. General Procedure for Compounds 8a-g An appropriate N-heterocyclic compound (11.7 mmol), DBU (1.78 g, 11.7 mmol), and methyl 2-(oxetan-3-ylidene)acetate 7 (1.50 g, 11.7 mmol) were dissolved in acetonitrile (4 mL) and stirred at 45 • C for 24 h. The reaction was quenched by the addition of water (15 mL) and extracted with ethyl acetate (3 × 10 mL). The combined organic solutions were dried over anhydr. Na 2 SO 4 , and then the solvents were removed under reduced pressure. Purification was conducted via flash chromatography. the azetidine building blocks were achieved through palladium-catalysed Suzuki-Miyaura cross-coupling reactions from a corresponding brominated pyrazole scaffold with alkyl and aryl boronic acids. These new heterocyclic compounds could be reliably determined using advanced NMR spectroscopy techniques, in particular by conducting 1 H-1 H NOESY, 1 H-13 C HMBC, and 1 H-15 N HMBC experiments.