Stereoselective Control of the Cu Activation of β,β-Diboryl Acrylates for Allylic Coupling Protocols with Concomitant Lactonization

Abstract

The key to a successful C–B activation is to discriminate between two geminal boryl moieties that are exposed to the same reaction conditions. Here we describe a stereoselective C–B activation of β,β-diboryl acrylates forming exclusively the (Z)-α-borylalkenyl copper(I) key intermediate, for subsequent allylic alkylation reactions. The new borylated (Z)-skipped dienoates followed a feasible iodo-lactonization sequence for the preparation of borylated lactone cores, which can be used in drug discovery.

Boron-selective chemical transformations allow the modular and rapid construction of molecular diversity and complexity for applications in organic synthesis for biomedical purposes.¹ The two installed geminal pinacolboryl substituents on 1,1-diborylalkenes²⁻⁶ can be stereoselectively differentiated and transformed in a stepwise manner, showing the potential of 1,1-diborylalkenes as versatile intermediates in organic synthesis.⁷⁻¹² The inclusion of carbonyl groups in β,β-diboryl acrylates contributes to the increase in the functional diversity through the C–B discrimination pathway because Suzuki–Miyaura coupling with arylhalides, in the presence of Pd(OAc)₂ and DtBPF [DtBPF = 1,1′-bis(di-tert-butylphosphino)ferrocene], occurred selectively at the boron site trans to the ester groups (Scheme 1a).¹³ However, here we present a complementary stereoselective C–B activation of β,β-diboryl acrylates with Cu(I) salts that exclusively activates the boron site cis to the ester group, promoting nucleophilic C–C bond formation at the more hindered C(sp²)–B position through allylic coupling reactions (Scheme 1b).

Scheme 1. Complementary Stereoselective C–B Activation of β,β-Diborylacrylates in Pd-Catalyzed Cross-Coupling or Cu-Catalyzed Allylic Alkylation Reactions.

The access to (Z)- or (E)-3-aryl 3-pinacolboryl acrylate compounds has been achieved via syn(14,15) or anti(16) catalytic hydroboration of alkynoates, respectively. However, substituents other than aryl groups at C_β have been less explored, particularly those that include allylic substituents, due to the inherent competition with respect to the hydroboration pathway. In that context, the synthesis of borylated (Z)-skipped dienoates is performed for the first time in this work (Scheme 2), taking advantage of the exclusive formation of the (Z)-stereoisomer for subsequent iodo-lactonization to prepare a series of functionalized lactones, with potential interest in medicinal chemistry. In particular, we focused on the preparation of α,β-unsaturated δ-lactone moieties for straightforward access to the essential core of novel kazusamycin A derivatives, for use as potent antitumor agents.¹⁷

Scheme 2. Structural Design of Kazusamycin A Derivatives.

In our initial experiments, we investigated the allylic coupling reaction of ethyl 3,3′-bispinacolboryl propenoate (1a) with 3-bromoprop-1-ene by employing a catalytic amount of copper salt CuCl and ligand PPh₃ in the presence of a variety of bases, in THF. We initially chose LiO^tBu on the basis of its efficiency in the Cu-catalyzed site-selective activation of 1,1-diborylalkenes, containing aryl or vinyl substituents.¹¹ However, only ethyl (E)-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)acrylate (3a) could be identified in the reaction mixture as a consequence of the selective activation on the boron site cis to the ester group, followed by protonation (Table 1, entry 1). A similar reaction outcome was observed when the iPrCuCl complex was used, in the presence of LiO^tBu as the base (Table 1, entry 2). The lack of allylic coupling moved us to consider alternative bases, avoiding the alkoxide groups that might be responsible for B activation through boron“ate” intermediates followed by a protonation step. When the base involved was Cs₂CO₃, we identified a moderate yield of the desired product ethyl (Z)-3-pinacolboryl-2,5-hexadienoate 2a, with a small amount of protodeborated byproduct 3a (Table 1, entry 3). Replacing Cs₂CO₃ with K₂CO₃ favored the exclusive formation of coupled product 2a, although in a moderate yield (Table 1, entry 4). The highest yield with exclusive formation of product 2a was achieved when K₃PO₄ was the base used, at 60 °C, because lower temperatures decreased the yield (Table 1, entries 5 and 6).

Table 1. Optimization of the Reaction Conditions for Stereoselective Cu Activation of β,β-Diborylacrylate 1a toward Nucleophilic Allylic Coupling.

entrya	Cu(I)/ligand	base	T (°C)	NMR yieldb (%)	2a:3a	2a (Z:E)
1	CuCl/PPh3	LiOtBu	60	>90	1:99	–
2	iPrCuCl	LiOtBu	60	>90	1:99	–
3	CuCl/PPh3	Cs2CO3	60	52	89:11	99:1
4	CuCl/PPh3	K2CO3	60	68	99:1	99:1
5	CuCl/PPh3	K3PO4	60	93	99:1	99:1
6	CuCl/PPh3	K3PO4	30	79	99:1	99:1

^aGeneral conditions: 1,1-diborylalkene (0.2 mmol), 3-bromoprop-1-ene (1.5 equiv), Cu salt (10 mol %), PPh₃ (10 mol %), base (2 equiv), THF (4 mL), T, 16 h.

^bYields determined by NMR with naphthalene as the internal standard.

This reaction can be described as a stereoselective C–B activation of β,β-diboryl acrylates toward the synthesis of borylated (Z)-skipped dienoates, which to the best of our knowledge have been prepared for the first time in this work. Our methodology guarantees the control of the stereoselectivity, under the convenient CuCl/PPh₃ catalytic system, and complements the reported protocols based on catalytic allylboration of alkynes for the synthesis of borylated skipped dienes.¹⁸ With the established reaction conditions presented above, we explored the scope of this Cu-catalyzed stereoselective deborylative allylic alkylation reaction, as summarized in Table 2. The reaction was performed well with methyl 3,3′-bispinacolboryl propenoate (1b) and 3-bromoprop-1-ene, demonstrating the compatibility of an alternative ester group (Table 2, entries 1 and 2, respectively). The reaction of 1a with 3-bromo-3,3-difluoroprop-1-ene allowed the formation of perfluorinated (Z)-skipped product 4, suggesting that the C–C coupling might proceed through an S_N2′ mechanism (Table 2, entry 2). An additional substituent at the R₂ position of the allyl halide is tolerated, as shown by the efficient allylic coupling for both 3-bromo-2-methylprop-1-ene and (3-bromoprop-1-en-2-yl)cyclopentane (Table 2, entries 3 and 4, respectively). Interestingly, even sterically hindered aryl and benzyl groups in this position were tolerated with the Cu-catalyzed strereoselective deborylative allylic alkylation reaction (Table 2, entries 5 and 6, respectively).

Table 2. Substrate Scope for the Stereoselective Cu-Catalyzed Allylic Alkylation of β,β-Diborylacrylates 1a and 1b.

^aGeneral conditions: 1,1-diborylalkene (0.2 mmol), allyl halide (1.5 equiv), CuCl (10 mol %), PPh₃ (20 mol %), K₃PO₄ (2 equiv), THF (4 mL), 60 °C, 16 h.

^bYields determined by NMR with naphthalene as the internal standard.

^cIsolated yield.

^dNMR yield of 78%, isolated yield of 44% on a 1 mmol scale.

^eYield of 3a of 6%.

^fYield of 3b of 8%.

^gYield of 3a of 18%.

^hYield of 3a of 8%.

The inclusion of unsaturated functional groups in 2-(bromomethyl)penta-1,4-diene and 2-(bromomethyl)-3-methylbuta-1,3-diene proved to be compatible with C–C bond formation through allylic coupling, providing access to double skipped system 9 and double diene product 10, although the latter in moderate yield (Table 2, entries 7 and 8, respectively). When we studied the allylic coupling of 1a with 2,3-dibromoprop-1-ene, skipped (Z)-diene 11 was exclusively formed as a result of a chemoselective C–Br coupling, together with the protodeborylated byproduct (18%) (Table 2, entry 9). Eventually, the reactivity of 1a with 3-bromo-2-(bromomethyl)prop-1-ene and 3-chloro-2-(chloromethyl)prop-1-ene provided access to polyfunctionalized products 12 and 13, respectively, with the remaining C(sp³)–halide functionality for downstream transformations (Table 2, entries 10 and 11, respectively). Isolated yields are modest due to the instability of the C(sp²)–Bpin fragment under the purification conditions, despite differently treated silica species being used as stationary phases. According to the overall reactivity found, we suggest that the Cu-catalyzed allylic alkylation between β,β-diboryl acrylates and the allyl halides depicted in Table 2 might include an S_N2′ mechanism. This is in agreement with the reported copper-catalyzed S_N2′-selective allylic alkylation reactions involving gem-diborylalkanes¹⁹⁻²⁵ or gem-diborylalkenes.¹¹ Aiming to generalize this S_N2′ allylic coupling, we explored the most challenging Cu-catalyzed coupling between 1a and (E)-1-bromobut-2-ene and observed the formation of only product 14, consistent with the favored γ selectivity, although the yield was only moderate presumably due to the greater steric hindrance (Scheme 3). Similarly, the coupling between 1a and (E)-1,4-dibromobut-2-ene or (E)-1,4-dichlorobut-2-ene generated exclusively γ-selective products 15 and 16, respectively, in moderate yields (Scheme 3). Notably, the cyclic 3-bromocyclohex-1-ene was a suitable electrophile along the Cu-catalyzed allylic alkylation of 1a and 1b, because products 17a and 17b were efficiently synthesized and isolated in 61% and 56% yields, respectively (Scheme 3).

Scheme 3. Control of the γ Selectivity in the Cu-Catalyzed Allylic Alkylation of β,β-Diborylacrylate 1 with γ-Substituted Allyl Halides.

General conditions: 1,1-diborylalkene (0.2 mmol), allyl halide (1.5 equiv), CuCl (10 mol %), PPh₃ (10 mol %), K₃PO₄ (2 equiv), THF (4 mL), 60 °C, 16 h. Yields determined by NMR with naphthalene as the internal standard. Isolated yields in brackets.

Bearing in mind the inherent difficulty of discriminating between the two geminal Bpin–C(sp²)–Bpin bonds in 1,1-diborylalkenes, we can justify the preferred formation of (Z)-α-borylalkenyl copper(I) species, and the subsequent coupling reaction with allyl bromides, by releasing the steric repulsion between the pinacolboryl and the ester group in the cis disposition.¹¹

The ultimate goal of the stereoselective synthesis of borylated (Z)-skipped dienoates is the iodo-lactonization sequence for preparing a series of borylated lactones, with the versatile C(sp²)–B handle for downstream functionalization of the lactone core (Scheme 4). To the best of our knowledge, borylated lactones have been achieved only through Pd-catalyzed cross-coupling pathways²⁶ or electrophilic oxyboration protocols.²⁷ Here, inspired by previous works on iodo-lactonization reactions by Larock²⁸⁻³¹ and Knochel,³² we screened the feasible reaction of borylated 1,4-dienonate 2a with I₂, at room temperature. After 16 h, we observed complete conversion into borylated lactone 18 (Scheme 4), suggesting that the terminal double bond coordinates selectively to the iodine cation generated from I₂ to render an iodonium intermediate followed by intramolecular rearrangement through the ester group. Interestingly, I₂ activation of the double bond seems to be favored versus the competitive iodo-deborylation reaction,³³ showing the remarkable stability of the Bpin group during the lactonization process. With the aim of preparing α,β-unsaturated δ-lactone moieties for straightforward access to the essential core of novel kazusamycin A derivatives,¹⁷ we extended the study of the iodo-lactonization to γ-substituted borylated (Z)-skipped dienoates 14 and 15. Polysubstituted lactones 19 and 20 were efficiently prepared, in the presence of I₂ at room temperature, with a notable preference for the cis stereoisomer in each case (Scheme 4). It is worth noting that fused lactone 21 was synthesized from borylated (Z)-skipped dienoate 17a, as an exclusive cis stereoisomer, being isolated in 96% yield (Scheme 4). The relative configuration of the C–I bond with respect to the fused C–H bonds was unambiguously assigned by one-dimensional NMR NOE experiments and X-ray single-crystal diffraction analysis of bicycle 21, synthesized as a single diastereoisomer (Scheme 4).

Scheme 4. Iodo-lactonization of Borylated (Z)-Skipped Dienoates and X-ray Diffraction of Bicycle 21.

General conditions: 1,4-dienonate (0.2 mmol), I₂ (3 equiv), CH₂Cl₂ (2 mL), rt, 16 h. Yields determined by NMR with naphthalene as the internal standard. Isolated yields in brackets. In the ORTEP drawing, thermal ellipsoids are drawn at the 50% level.

We conducted the global transformation of β,β-diboryl acrylate 1a into lactones 18–21, through one-pot, two-step allylic coupling/lactonization, and the overall yields were similar to those involving the purification of the intermediate borylated 1,4-dienoates. We also studied the iodo-lactonization of borylated 1,4-dienonate 4a with I₂, but the reaction was not completed, even at longer reaction times, probably due to the less nucleophilic nature of the gem-difluoro-substituted terminal alkene. Purification of the corresponding borylated lactone turned out to be operationally difficult. To circumvent this obstacle, we strategically planned the cross-coupling of 4a with PhI, prior to the iodo-lactonization, in the presence of Pd(PPh₃)₄/K₂CO₃, and arylated 1,4-dienonate 22 could be isolated in high yields (Scheme 5). Subsequent iodo-lactonization of 22 with I₂ resulted in the formation of lactone 23, which contains the pending CF₂I group. Current interest in difluoroalkyl iodide motifs is due to its suitability as a surrogate model for the construction of alkyl–CF₂–alkyl bonds via site-selective coupling reactions.^34,35 Next, we explored the iodo-lactonization of borylated (Z)-skipped dienoate 5a with I₂, and the reaction outcome showed the formation of the desired borylated lactone as the main product; however, once again, purification was not successful. However, when we conducted the Pd-catalyzed cross-coupling between 5a and PhI, the resulting (Z)-skipped dienoate 24 was easily isolated in high yields (Scheme 5) and successful iodo-lactonization was achieved in the presence of I₂, generating α,β-unsaturated δ-lactone 25, containing a quaternary carbon, in a synthetically useful isolated yield.

Scheme 5. Pd-Catalyzed Cross-Coupling of Borylated 1,4-Dienoates 4a and 5a, Followed by an Iodo-lactonization Step.

General conditions for cross-coupling: borylated 1,4-dienonate (0.2 mmol), Pd(PPh₃)₄ (5 mol %), PhI (2 equiv), K₂CO₃ (2 equiv), 70 °C, 16 h. General conditions for iodo-lactonization: arylated 1,4-dienonate (0.2 mmol), I₂ (3 equiv), CH₂Cl₂ (2 mL), rt, 16 h. Yields determined by NMR with naphthalene as the internal standard. Isolated yields in brackets.

Alternatively, we explored the Pd-catalyzed cross-coupling reaction between the borylated lactones and PhI to illustrate the ability to functionalize the Bpin moiety on the lactone core in the last step. When borylated lactone 19 [as a mixture of stereoisomers (63:37 cis:trans)] was reacted with PhI in the presence of Pd(PPh₃)₄/K₂CO₃, both stereoisomers evolved to the corresponding coupled products (Scheme 6a), although the less sterically hindered trans stereoisomer suffered from H–I elimination, which generated the corresponding exocyclic alkene that was subsequently isomerized to deliver α-pyrone 27.³⁶ Functionalized cis-lactone 26 was the major product isolated (Scheme 6a), resulting in an appropriate core for the synthesis of kazusamycin A derivatives.¹⁷ The single diastereoisomer 21 was also coupled with PhI, and the resulting lactone 28 was quantitatively formed and isolated in 90% yield (Scheme 6b). Finally, we conducted the Cu-catalyzed borylation of 21 with B₂pin₂ and were delighted to observe the formation of the new β,β-diborylated lactone 29, increasing the size of the polyfunctionalized platform through the gem-diboron moiety, becoming the first lactone with β,β-diborylated motifs (Scheme 6b).

Scheme 6. Pd-Catalyzed Cross-Coupling of Borylated Lactones 19 and 21.

General conditions for cross-coupling: borylated lactone (0.2 mmol), Pd(PPh₃)₄ (5 mol %), PhI (2 equiv), K₂CO₃ (2 equiv), THF, 70 °C, 16 h. General conditions for β-borylation: borylated lactone (0.2 mmol), [Cu(MeCN)₄]PF₆ (5 mol %), PⁿBu₃ (10 mol %), B₂pin₂ (1.2 equiv), K₂CO₃ (10 mol %), THF, 30 °C, 16 h. Yields determined by NMR with naphthalene as the internal standard. Isolated yields in brackets.

In summary, we have disclosed a stereoselective C–B activation of β,β-diborylacrylates forming exclusively the (Z)-α-borylalkenyl copper(I) key intermediate for subsequent selective allylic alkylation reactions. The new borylated (Z)-skipped dienoates followed the iodo-lactonization sequence to deliver polysubstituted borylated lactone cores that might have potential in drug discovery. Subsequent Pd-catalyzed cross-coupling reactions and Cu-catalyzed β-borylation sequences illustrated the potential of the polyfunctionalization platforms.

Data Availability Statement

The data underlying this study are available in the published article and its Supporting Information.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.3c03640.

• Experimental procedures, product characterization, and spectra (PDF)

The authors declare the following competing financial interest(s): M.M. is a Sanofi employee and may hold shares and/or stock options in the company. E.F. and M.P. have nothing to disclose.