Molecular Diversity by Olefin Cross-Metathesis on Solid Support. Generation of Libraries of Biologically Promising β-Lactam Derivatives.

The application of the reagent-based diversification strategy for generation of libraries of biologically promising β-lactam derivatives is described. Key features are the versatility of the linker used and the cross-metathesis functionalization at the cleavage step. From an immobilized primary library, diversity was expanded by applying different cleavage conditions, leading to a series of cholesterol absorption inhibitor analogues together with interesting hybrid compounds through incorporation of a chalcone moiety.

1. Introduction

The β-lactam skeleton is an active acylating agent with numerous applications in clinical therapy [1]. Besides the well-known use as antibiotics [2], β-lactam ring has inhibitory effects on prostate specific antigen [3,4], thrombin [5], human cytomegalovirus protein [6], human leukocyte elastase [7], cysteine protease [8,9], and human fatty acid amide hydrolase [10], as well as anticancer properties [11,12,13,14,15] and neuroprotective action [16,17]. A new area in β-lactam-based drugs has been established by development of β-lactam derivatives with strong cholesterol absorption inhibitor reducing LDL concentration [18,19,20,21]. An example is the commercial drug ezetimibe, which is one of the most prescribed drugs in the US. Recently, simvastatin has been shown to provide an incremental benefit on reducing cardiovascular events in acute coronary syndrome patients, when co-administered with ezetimibe [22]. Furthermore, β-lactams have also been considered as peptidomimetic species for mimicking certain properties of proteins and, from a synthetic point of view, they are key synthons for the preparation of various heterocyclic compounds of biological importance [23,24,25].

In drug discovery related organic synthesis, one of the tools for a rapid and efficient construction of diversity-based small molecules is parallel solid-phase synthesis. Particularly, solid-phase chemistry has recently aroused interest in metal-catalyzed cross-coupling reactions since undesirable soluble homodimers can be washed away during purification providing chemoselectivity, while immobilization of one of the substrates makes its homodimerization a less favorable process due to site isolation (Scheme 1) [26]. In this regard, we and others have recognized the usefulness of solid-supported olefin cross-metathesis for generation of biologically relevant molecules [27,28,29,30,31], including a comprehensive study to understand the process [32,33]. From the viewpoint of green chemistry, using solid-phase synthetic sequences allows a significant reduction in solvent waste, since purification is performed by phase separation, avoiding chromatographic isolation of products which requires a large consumption of organic solvents [34].

Scheme 1

Metal-catalyzed solid-phase synthesis.

In the search of diversity, a bunch of very ingenious strategies have been developed, among which diversity-oriented synthesis (DOS) is the most widespread [35,36,37]. According to DOS, skeletal diversity can be basically achieved by two main approaches: one known as “substrate-based diversification” where the same reagent generates different structures by reaction with selected substrates; and “reagent-based diversification” in which different products are obtained when the same substrate is subjected to different reaction conditions [38].

Generally, the point of attachment to the resin in solid-phase organic synthesis (SPOS) limits the possibilities of diversity generation. In previous work, we have developed a series of β-lactam compounds linked to the resin at position 1 or 3 of the ring. Only the substituents on the two remaining positions could be combined [27,39]. In order to maximize structural variation from a library of compounds, we have studied a strategy based on a linker, which ensures release by different reaction conditions to obtain diverse products using a DOS reagent-based diversification approach (Scheme 2).

Scheme 2

Reagent-based diversification approach in solid-phase synthesis.

We have shown that high molecular diversity in β-lactam scaffolds can be obtained using a linker cleavable by two orthogonal conditions: one by treatment with 10% TFA, and the other using a set of olefin cross-metathesis reactions. While Seeberger developed olefin linkers for solid-phase synthesis of glycosides [40], we decided to introduce a new twist using them to generate diversity during cleavage step. Thus, a series of cholesterol absorption inhibitor analogues, together with interesting β-lactam-chalcone hybrid compounds, have been generated. Our results are summarized in this extended paper [41].

2. Results and Discussion

Synthesis started with immobilization of 4-pentenoic acid to Wang resin by standard coupling conditions, to afford the corresponding resin 1 which, in turn, was the substrate for a solid-supported olefin cross-metathesis by reaction with 1-(chloromethyl)-4-vinylbenzene in the presence of second generation Grubbs precatalyst, according to the methodology reported by us (Scheme 3) [32]. Then, oxidation of the supported benzyl chloride 2 with DMSO/NaHCO3 at 155 °C [42] yielded the aldehyde 3, which was characterized by 13C gel-phase NMR (Figure 1). Aldehyde 3 was used as substrate for the synthesis of libraries of β-lactam derivatives.

Scheme 3

Synthesis of versatile linker 3.

Figure 1

13C gel-phase NMR of compounds 2 and 3.

2.1. Primary β-Lactam Library

For the solid-phase version of the classical Staudinger reaction between imines and ketenes [43,44,45], we first developed the synthesis of a set of immobilized imines 5a–d by treatment of the aldehyde with different amines 4a–d in refluxing benzene using a Dean–Stark trap (Scheme 4).

Scheme 4

Primary immobilized β-lactam library.

For the in situ generation of ketenes, we found that stable acid chlorides such as phenoxyacetyl chloride were ideal, however, when aliphatic carboxylic acids derivatives were required, the best choice was the activation of the free acid by Mukaiyama’s reagent. Thus, immobilized imines 5a–d were treated with either acid chloride or carboxylic acids (6a–d) and Mukaiyama’s reagent to obtain the primary β-lactam library (7aa–dd) (Scheme 4).

The relative configuration of C-3 and C-4 substituents was determined taking the coupling constant between H-3 and H-4 nucleus in 1H-NMR experiments after trifluoroacetic acid (TFA) cleavage into account. Coupling constants greater than 4 Hz were indicative of a cis configuration [46], so β-lactams 8aa and 8ab had a cis configuration while 8bb had a trans configuration (Figure 2 and Table 1).

Figure 2

Comparative 1H-NMR spectra of 3,4-cis- and trans-substituted β-lactams.

Table 1

The secondary “trifluoroacetic acid (TFA)” library of β-lactams.

Entry	Starting Material	3,4 Configuration	R1	R2	Product	Yield (%) a
1	7aa	cis	PhO	Bn	8aa	58
2	7ab	cis	PhO	Ph	8ab	43
3	7bb	trans		Ph	8bb	28
4	7bc	trans		4-FPh	8bc	38
5	7bd	trans		4-MeOPh	8bd	53
6	7cb	trans		Ph	8cb	26
7	7cc	trans		4-FPh	8cc	42
8	7cd	trans		4-MeOPh	8cd	60
9	7db	trans		Ph	8db	55
10	7dc	trans		4-FPh	8dc	52
11	7dd	trans		4-MeOPh	8dd	33

a Overall isolated yield after flash column chromatography [five steps, based on loading of resin 2 (0.40 mmol/g, determined by elemental analysis of chlorine)].

During Staudinger reaction, the diastereoselectivity for cis or trans configuration of the resulting β-lactams was governed by substituents on the starting building blocks (imine and ketene) [47,48]. Ketenes that were activated by heteroatoms had strong preference for formation of cis products. This was the case for β-lactams 8aa (J = 5.0 Hz) and 8ab (J = 4.4 Hz) that have been constructed from ketenes derived from dehydrohalogenation of phenoxyacetyl chloride. On the other hand, β-lactam 8bb, which was synthesized from an alkyl-ketene derivative obtained from 5-phenylvaleric acid, gave a trans configuration (J = 2.2 Hz). The presence of a heteroatom tethered to the α-position of the ketene and tended to form the cis products by conrotatory ring closure of the zwitterionic intermediate, formed through a nucleophilic attack of imine nitrogen on the electrophilic carbonyl carbon of the ketene [47]. Conversely, thermodynamically more stable trans-β-lactam 8bb was generated by isomerization of the zwitterion, when a hydrocarbon chain was tethered to the ketene.

1H-NMR signals corresponding to protons H-3 and H-4 showed a deshielding effect when substituent of position 1 changed from N-benzyl to N-aryl, respectively (compare 8aa and 8ab in Figure 2). Geometric optimization employing a semi-empirical method such as AM1 [49] indicated a radical change in the arrangement of the aromatic ring at position 1 (Figure 3). In the N-aryl β-lactam, H-3 and H-4 were in the same plane of the N-1 aromatic ring, while in N-benzyl β-lactam these protons were near the axis of the ring. Due to anisotropic effect, protons H-3 and H-4 in the N-aryl-β-lactam were in a deshielding area, whereas in N-benzyl-β-lactam those protons were in a shielding environment. The absence of an oxygen atom directly linked to position 3 of the β-lactam ring in 8bb, leading to a drastic shift of the signal to higher fields.

Figure 3

AM1 geometric optimization of N-aryl and N-benzyl β-lactams.

Generation of this primary library was monitored by IR and 13C gel-phase NMR. A representative example of 13C gel-phase NMR of resin 7bd is shown in Figure 4. Signals assigned to the immobilized β-lactam are marked (*). The most representative signals were 55.1 ppm (MeO-), and 60.7/60.2 ppm corresponding to C-3 and C-4 of the β-lactam ring.

Figure 4

13C gel-phase NMR of immobilized β-lactam 7bd.

2.2. Secondary β-Lactam Libraries

Taking advantage of the multiple possibilities of cleavage, a series of secondary libraries was generated. Thus, TFA treatment of the primary library, followed by diazomethane methylation gave an eleven compounds’ secondary “TFA” library (Table 1).

The efficiency of the solid-phase synthetic strategy has proven to be excellent. Yields ranged from 26% to 60% for five synthetic steps. Analyzing the reaction outcome of the obtained trans-β-lactams (entries 3–11), no clear tendency could be observed. While yields increase from R2 = Ph to R2 = 4–FPh and 4–MeOPh, in the cases of R1 = phenylpropyl and vinyl, this tendency was reverted in case of R1 = allyl (Figure 5).

Figure 5

Yields of R1 and R2 β-lactam substituents.

In order to increase diversity during cleavage step, we studied the olefin cross-metathesis on the immobilized β-lactams. Thus, a new secondary library was created when β-lactams 7bb–bd, bearing a 3-phenylpropyl group at C-3 position, reacted with different olefins to afford structural diversification at C4 position (Table 2). This secondary library, called “Ru1”, was present in six different alternative β-lactams (10bba–bdc) obtained by reaction of the olefins in presence of second generation Grubbs precatalyst. In accordance with our previous results in the area of olefin cross-metathesis [32], trans-crotonic acid (entries 1 and 2), a type I olefin [50] and allylbenzene (entries 3 and 4), a type II olefin, performed the cleavage process with high efficiency, while 2-bromostyrene (entries 5 and 6), a type IV olefin, provided a more poor yield.

Table 2

The secondary “Ru1” library of β-lactams.

Entry	Starting Material	R2	R3	Product	Yield (%) a	CM Yield (%) b
1	7bb	–H	MeO2C– c	10bba	25	87
2	7bd	–OMe	MeO2C– c	10bda	47	88
3	7bd	–OMe	Bn	10bdb	49	92
4	7bc	–F	Bn	10bcb	31	81
5	7bc	–F	2-BrPh	10bcc	27	50
6	7bd	–OMe	2-BrPh	10bdc	16	41

a Overall isolated yield after flash column chromatography [four steps, based on loading of resin 2 (0.40 mmol/g, determined by elemental analysis of chlorine)]. b Yield of the cross-metathesis step calculated from the ratio between product yield and the yield of the corresponding β-lactams 8bb–bd (see Table 1). c Product was esterified with diazomethane prior to purification.

Hybrid structures, obtained from combining at least two biologically significant moieties, have emerged as a novel approach in finding new chemical entities [51,52,53]. The molecular hybridization strategy has proven to be helpful in many aspects related to drug discovery, such as overcoming drug-resistance problems or improving active transport mechanisms. In particular, β-lactam-based hybrids have recently acquired importance owing to the fact that many of them exhibit very promising biological activity [54,55,56,57,58,59]. In order to further increase diversity and also obtain biologically interesting β-lactam-based hybrids, metathesis conditions, previously developed for the synthesis of chalcones, were applied for the cleavage step (Table 3) [60]. In this secondary library, called “Ru2”, cleavage conditions were performed by reaction of β-lactams 7bc and 7bd with non-immobilized α,β-unsaturated ketones, being the most efficient precatalyst of Hoveyda-Grubbs carbene ruthenium complex [61,62,63]. Thus, immobilized β-lactams 7bc–bd were treated with substituted vinyl phenyl ketones 9d–f in the presence of Hoveyda-Grubbs precatalyst to yield the soluble compounds 10bcd–bdg which combined two recognized pharmacophoric moieties such as azetidinone and chalcone in one molecule.

Table 3

The secondary “Ru2” hybrid β–lactam library.

Entry	Starting Material	R2	R3	Product	Yield (%) a	CM Yield (%) b
1	7bd	–OMe	–CH3	10bdd	20	38
2	7bd	–OMe	–OCH3	10bde	15	28
3	7bc	–F	–I	10bcf	22	58
4	7bc	–F	–CH2Ph	10bcg	20 c	53

a Overall isolated yield after flash column chromatography [four steps, based on loading of resin 2 (0.40 mmol/g, determined by elemental analysis of chlorine)]. b Yield of the cross-metathesis step calculated from the ratio between product yield and the yield of the corresponding β-lactams 8bc–bd (see Table 1). c Product could not be completely purified because of impurities.

Through the analysis of the 13C-NMR spectra of the components of obtained libraries, a clear pattern could be determined (Figure 6). Signals corresponding to substituents of all positions are slightly affected by remaining substituents. For instance, 4-methoxy-phenyl substituent at position 1 in azetidinones 8bd, 8cd, 8dd, 10bda, 10bdb, 10bdc, 10bdd and 10bde showed similar chemical shifts regardless the remaining substituents of β-lactam. Signals of the three carbons of the β-lactam ring (C-2, C-3 and C-4) could clearly be identified, having only a remarkable and expected low-field chemical shift of C-3 in case of 3-phenoxy-cis-β-lactams (8aa and 8ab). Although there was the possibility that other signals could interfere, the patterns shown in Figure 6 were useful for a rapid detection of β-lactam ring containing compounds obtained by high throughput parallel synthesis.

Figure 6

Comparative chart of the 13C-NMR signals of the library components.

3. Materials and Methods

3.1. General Information

Chemical reagents were purchased from commercial suppliers and used without further purification, unless otherwise noted. Solvents were analytical grade or were purified by standard procedures prior to use. Reactions requiring inert atmosphere were carried out under high-purity dry nitrogen atmosphere. Solvents from these reactions were transferred under high-purity dry nitrogen using syringes. All reactions were monitored by thin layer chromatography (Merck, Darmstadt, Germany) performed on silica gel 60 F254 pre-coated aluminum sheets, visualized by a 254 nm UV lamp, and stained with an ethanolic solution of 4-anisaldehyde. Column flash chromatography (Merck, Darmstadt, Germany) was performed using silica gel 60 (230–400 mesh). The purity criteria were (i) the appearance of a single spot by thin layer chromatography (ii) the presence of the corresponding signals in 1H and 13C-NMR and (iii) the range of melting points in case of solid samples. Molecular modeling was performed with HyperChem v8.03 (Hypercube, Gainesville, FL, USA) using the AM1 method

Solid-phase reactions were carried out in polypropylene cartridges equipped with a frit (Supelco, Bellefonte, PA, USA), unless reflux conditions were required. In that case, standard glassware was used. All solid-phase reaction mixtures were stirred at slowest rate. Compounds 8bb–8dd and 10bba–10bdc have been previously reported [41].

3.2. Instrumental and Physical Data

1H-NMR spectra were recorded in a Bruker Avance spectrometer (Bruker Analytik GmbH, Karlsruhe, Germany) at 300 MHz in CDCl3 with tetramethylsilane (TMS) as internal standard (0 ppm). 13C-NMR spectra were recorded on the same apparatus at 75 MHz with CDCl3 as solvent and reference (76.9 ppm). Chemical shifts (δ) were reported in ppm upfield from TMS and coupling constants (J) were expressed in Hertz. The following abbreviations were used to indicate multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, and bs = broad singlet. For preparation of the samples for 13C gel-phase NMR, 50–80 mg of resin was placed in a standard NMR tube and 0.5 mL of CDCl3 was slowly added in order to obtain a gel, which was homogenized by sonication. Spectra were run according to the literature [64,65].

Infrared spectra were recorded on a Shimadzu FT-IR spectrometer model 8101 (Shimadzu, Tokyo, Japan). The resin samples were measured as dispersions in KBr discs, made by compression of a mixture finely powdered in an agate mortar. Approximately 3 mg sample was used in 100 mg KBr.

3.3. Synthetic Procedures

General procedure for the synthesis of 3-phenoxy-β-lactam derivatives (8aa–ab) and β-lactam-chalcone hybrids (10bdd–bcg): a mixture of 4-pentenoic acid (420 μL, 4.15 mmol, 5.0 eq.) in anhydrous DMF (10 mL) and DCC (830 μL, 4.15 mmol, 5.0 eq.) was stirred for 30 min at room temperature and transferred via cannula to Wang resin (750.0 mg, 1.1 mmol/g, 0.83 mmol), which was previously rinsed with anhydrous DMF (5.0 mL). After adding DMAP (100.0 mg, 0.83 mmol, 1.0 eq.), the reaction mixture was stirred for 16 h at room temperature. Then, resin was washed with DMF (3 × 10 mL), 1% AcOH in AcOEt (3 × 10 mL), AcOEt (3 × 10 mL), MeOH (3 × 10 mL) and CH2Cl2 (3 × 10 mL), and dried in vacuo. The immobilized pentenoate (1) (820 mg, 0.83 mmol) was suspended in anhydrous CH2Cl2 (20 mL) and 4-vinylbenzyl chloride (590 μL, 4.15 mmol, 5.0 eq.) was added via syringe under a nitrogen atmosphere. After that, Grubbs’ second generation pre-catalyst (35.0 mg, 41.5 μmol, 5 mol %) was added and the reaction was refluxed for 20 h. The mixture was then filtered, washed with CH2Cl2 (3 × 10 mL), MeOH (3 × 10 mL), CH2Cl2 (1 × 10 mL), and dried under high vacuum. The procedure was repeated once to ensure complete reaction. In the next step, NaHCO3 (68.5 mg, 2.2 eq.) in DMSO (25.0 mL) was added to the immobilized benzyl chloride (2) (915.0 mg, 0.37 mmol) and the mixture was heated to 155 °C for 6 h. After that, the suspension was filtered, washed with DMSO (3 × 10 mL), CH2Cl2 (3 × 10 mL), MeOH (3 × 10 mL), CH2Cl2 (1 × 10 mL) and dried in vacuo to obtain the Wang resin-linked aldehyde 3. In the next step, the immobilized aldehyde 3 (121.2 mg; 0.13 mmol) was placed in a round-bottom flask and suspended in anhydrous benzene (20 mL) and the corresponding amine (10 equiv.) was added. A Dean-Stark trap was then fitted, filled with molecular sieves 4 Å and the suspension was heated in reflux for 14 h. After that, the resin was filtered, washed with benzene (3 × 4 mL), DCM (3 × 4 mL), MeOH (3 × 4 mL) and CH2Cl2 (1 × 4 mL), and dried in vacuo to obtain the resin-bound imine 5.

For 3-phenoxy-β-lactam derivatives (8aa–ab): to an aliquot of immobilized imine 5a–b (0.11 mmoles), triethylamine (0.30 mL, 2.2 mmoles, 20 eq.) and phenoxyacetyl chloride (0.23 mL, 1.65 mmoles, 15 equiv.) was added at 0 °C. The suspension was stirred for 16 h at room temperature. After filtration, the resin was washed with CH2Cl2 (3 × 4 mL), AcOEt (3 × 4 mL), MeOH (3 × 4 mL), CH2Cl2 (1 × 4 mL), and dried under high vacuum. Then, a 10% solution of TFA in CH2Cl2 (3 mL) was added to the polymer-bound β-lactam 7aa–ab. The reaction mixture was stirred for 50 min at room temperature, filtered, and washed with CH2Cl2 (3 mL). The filtrate was evaporated under reduced pressure. Esterification with diazomethane afforded the crude product that was then purified by column chromatography (hexane-AcOEt).

For β-lactam-chalcone hybrids (10bdd–bcg): 5-phenylvaleric acid (49 mg, 0.28 mmoles, 2.5 equiv.) and triethylamine (90 µL, 0.66 mmoles, 6 equiv.) were dissolved in anhydrous chloroform (3 mL) and added to a suspension of immobilized imine 5b–d (0.11 mmoles) in anhydrous chloroform (1.5 mL) under nitrogen atmosphere. After one minute, 2-chloro-1-methylpyridinium iodide (Mukaiyama’s reagent, 84.3 mg, 0.33 mmoles, 3 equiv.) was added and the suspension was stirred at room temperature for 24 h. Then, the reaction mixture was filtered and the resin was washed successively with CH2Cl2 (3 × 4 mL), AcOEt (3 × 4 mL), MeOH (3 × 4 mL), and CH2Cl2 (1 × 4 mL). Resin 7bc–bd (0.11 mmol) was placed in a 25 mL round-bottom flask, purged with dry nitrogen, suspended in anhydrous toluene (3 mL) and olefin 9d–f (0.55 mmol, 5 eq.) dissolved in anhydrous toluene (3 mL) was added via syringe. After addition of Hoveyda-Grubbs precatalyst (3.4 mg, 5.5 µmol, 5 mol %), the flask was fitted with a reflux condenser with a cannula adapted to allow the elimination of generated ethylene during the reaction. The system was heated to 75 °C for one hour under nitrogen atmosphere. The mixture was filtered and the filtrate was evaporated under reduced pressure to afford the crude product. The solvent was evaporated under reduced pressure and the crude material was purified by flash column chromatography (hexane-AcOEt).

3.4. Analytical Data

Methyl (E)-5-(4-(1-benzyl-4-oxo-3-phenoxyazetidin-2-yl)phenyl)pent-4-enoate: RMN de 1H (CDCl3, 300 MHz): δ 7.37–7.05 (m, 11H), 6.90–6.82 (m, 1H), 6.77–6.68 (m, 2H), 6.39 (d, J = 15.9 Hz, 1H), 6.20 (dt, J1 = 15.9 Hz, J2 = 6.6 Hz, 1H), 5.39 (d, J = 4.4 Hz, 1H), 4.89 (d, J = 14.6 Hz, 1H), 4.72 (d, J = 4.4 Hz, 1H), 3.85 (d, J = 14.6 Hz, 1H), 3.69 (s, 3H), 2.60–2.40 (m, 4H). RMN de 13C (CDCl3, 75 MHz): δ 173.2, 165.4, 156.9, 137.6, 134.6, 131.4, 130.3, 129.1, 129.0, 128.8, 128.7, 128.5, 127.8, 125.9, 121.9, 115.4, 82.1, 61.1, 51.5, 44.0, 33.6, 28.1.

Methyl (E)-5-(4-(4-oxo-3-phenoxy-1-phenylazetidin-2-yl)phenyl)pent-4-enoate: RMN de 1H (CDCl3, 300 MHz): δ 7.40–7.02 (m, 11H), 6.98–6.90 (m, 1H), 6.86–6.75 (m, 2H), 6,37 (d, J = 15.9 Hz, 1H), 6.18 (dt, J1 = 15.9 Hz, J2 = 6.2 Hz, 1H), 5.56 (d, J = 5.0 Hz, 1H), 5.37 (d, J = 5.0 Hz, 1H), 3.68 (s, 3H), 2.55–2.40 (m, 4H). RMN de 13C (CDCl3, 75 MHz): δ 173.2, 163.0, 156.9, 137.7, 136.8, 131.3, 130.3, 129.2, 129.0, 128.2, 126.0, 124.5, 122.1, 117.5, 115.7, 81.2, 61.8, 51.5, 33.6, 28.1.

1-(4-Methoxy-phenyl)-4-[4-(3-oxo-3-p-tolyl-propenyl)-phenyl]-3-(3-phenyl-propyl)-azetidin-2-one: 1H-NMR: 7.92 (d, J = 8.1 Hz, 2H), 7.76 (d, J = 15.9 Hz, 1H), 7.62 (d, J = 8.1 Hz, 2H), 7.51 (d, J = 15.9 Hz, 1H), 7.37 (d, J = 8.1 Hz, 2H), 7.31–7.14 (m, 7H), 6.78 (d, J = 9 Hz, 2H), 4.62 (d, J = 2.1 Hz, 1 H), 3.73 (s, 3H), 3.13–3.07 (m, 1H), 2.66 (t, J = 7 Hz, 2H), 2.43 (s, 3H), 1.99–185 (m, 4H). 13C-NMR: 189.8, 166.7, 156.0, 143.8, 143.3, 141.5, 140.6, 135.5, 135.2, 131.1, 129.3, 129.1, 128.6, 128.4, 126.4, 125.9, 122.5, 118.1, 114.3, 60.8, 60.6, 55.4, 35.7, 28.9, 28.4, 21.6. HRMS calcd. for: C35H33NNaO3+; (M + Na+, m/z): 538.23527; found: 538.23673.

1-(4-Methoxy-phenyl)-4-{4-[3-(4-methoxy-phenyl)-3-oxo-propenyl]-phenyl}-3-(3-phenyl-propyl)-azetidin-2-one: 1H-NMR: 8.02 (d, J = 8.8 Hz, 2H), 7.76 (d, J = 15.7 Hz, 1H), 7.63 (d, J = 8.2 Hz, 2H), 7.52 (d, J = 15.7 Hz, 1H), 7.36 (d, J = 8.2 Hz, 2H) 7.30–7.14 (m, 7H), 6.97 (d, J = 8.8 Hz, 2H), 6.77 (d, J = 9.0 Hz, 2H), 4.62 (d, J = 2.1 Hz, 1H), 3.89 (s, 3H), 3.73 (s, 3H), 3.12–2.99 (m, 1H), 2.68–2.63 (m, 2H), 1.98–1.82 (m, 4H). 13C-NMR: 188.5, 166.8, 163.5, 156.0, 142.9, 141.5, 140.5, 135.3, 131.1, 130.9, 130.8, 129.1, 128.4, 126.4, 125.9, 122.3, 118.1, 114.3, 113.9, 60.8, 60.5, 55.5, 55.4, 35.7, 28.9, 28.4. HRMS calcd. for C35H33NNaO4+ (M + Na+, m/z): 554.23018; found: 554.22853.

1-(4-Fluoro-phenyl)-4-{4-[3-(4-iodo-phenyl)-3-oxo-propenyl]-phenyl}-3-(3-phenyl-propyl) azetidin-2-one: 1H-NMR: 7.87 (d, J = 8.5 Hz, 2H), 7.78 (d, J = 15.6 Hz, 1H), 7.70 (d, J = 8.5 Hz, 2H), 7.63 (d, J = 8.1 Hz, 2H), 7.44 (d, J = 15.6 Hz, 1H), 7.37 (d, J = 8.1 Hz, 2H), 7.28–7.14 (m, 7H), 6.93 (m, 2H), 4.63 (d, J = 2.1 Hz, 1H), 3.10 (m, 1H), 2.67 (m, 2H), 2.0–1.82 (m 4H). 13C-NMR: 189.4, 167.0, 144.2, 141.4, 140.5, 137.9, 137.2, 135.1, 133.8, 129.8, 129.3, 128.4, 128.3, 127.9, 126.8, 126.5, 126.0, 125.6, 122.0, 118.3, 118.2, 116.0, 115.7, 100.8, 60.9, 60.8, 35.6, 28.9, 28.4. HRMS calcd. For C33H27FINNaO2+ (M + Na+, m/z): 638.09627; found: 638.09727.

4. Conclusions

An interesting application of DOS of biologically promising compounds has been developed. A key feature was the application of the reagent-based diversification approach using a linker with different possibilities of cleavage. Starting from an immobilized primary library, diversity was expanded through many alternatives according to the reaction conditions used. A series of cholesterol absorption inhibitor analogues was obtained, as well as interesting hybrid compounds through incorporation of a chalcone group via cross-metathesis. Here, we have demonstrated the reliability of the methodology which can be suitable for generating large libraries of analogue structures.