Expanding the Frontier of Linear Drug Design: Cu‐Catalyzed Csp–Csp 3‐Coupling of Electron‐Deficient SF4‐Alkynes with Alkyl Iodides

Abstract Despite the attractive properties of tetrafluorosulfanyl (SF4) compounds in drug discovery, medicinal research on SF4 molecules is hindered by the scarcity of suitable synthetic methodologies. Drawing inspiration from the well‐established Sonogashira cross‐coupling of terminal alkynes under Pd‐catalysis, it is envisioned that SF4‐alkynes can serve as effective coupling partners. To overcome the challenges associated with the electron‐deficient nature of SF4‐alkynes and the lability of the SF4 unit under transition‐metal catalysis, an aryl radical mediated Csp–Csp 3 cross‐coupling reaction is successfully developed under Cu catalysis. This methodology facilitates the coupling of SF4‐alkynes with alkyl iodides, leading to the immediate synthesis of SF4‐attached drug‐like molecules. These findings highlight the potential impact of SF4‐containing molecules in the drug industry, paving the way for further research in this emerging field.


Introduction
Fluorine (F) atoms, with their high electronegativity and lipophilic nature, have proven to be game-changers in modifying the physicochemical properties of organic molecules. [1]This strategic incorporation of F into suitable positions has been instrumental in designing biologically active compounds, as evidenced by the remarkable success of fluoro-agrochemicals and fluoro-pharmaceuticals, which now dominate 70% and 30% of registered drugs, respectively, in the last five years. [2]While carbon-fluorine (C-F) units are prevalent in fluorinated drugs DOI: 10.1002/advs.202306554 on the market, the intriguing prospect of fluorinated drugs featuring an X-F unit, not C-F, remains untapped. [2]onsidering this awakening background, we focused on the potential of tetrafluorosulfanyl (SF 4 )-containing organic molecules, [3,4] which harbor sulfur-fluorine (S-F) units, as promising and unconventional drug candidates for the future market. [5]The combination of strong electron negativity and high lipophilicity induced by the four F atoms with a hypervalent sulfur atom gives rise to unique properties in the SF 4 moiety.Notably, the trans-type SF 4 structure allows two individual substituents to align linearly (Figure 1a), making it an attractive bio-isostere of non-conjugated linear motifs such as cubanes and bicyclopen tanes (BCP), which mimic p-substituted benzenes and alkynes (Figure 1b). [6,7]espite the immense potential of SF 4 -containing molecules, progress in their synthesis has been hindered by the scarcity of suitable methodologies. [3,4]A recent advance was made by discovering that trans-SF 4 -alkynes, especially those with heteroaryl groups, offer enhanced stability to SF 4 molecules, which led to the exploration of the versatility of SF 4 -alkynes in generating various SF 4 -containing molecules (Figure 1c). [4]These compounds, with their highly electron-deficient nature in the alkyne portion, exhibit exceptional reactivity toward electron-rich species.Moreover, SF 4 -alkynes themselves would be an alternative bioisostere of a linear connection with a length of 6.23 Å, which is 0.37 Å longer than the p-benzene system of 5.86 Å.
Drawing inspiration from the well-established Sonogashira coupling (Figure 1d), [8] a widely used cross-coupling reaction of terminal alkynes with aryl or vinyl halides under Pd(0)/Cu(I) dual catalytic conditions or related transition-metal catalysis, we envisioned the potential of employing SF 4 -alkynes as coupling partners.Sonogashira coupling involves the coordination of Cu(I) to the triple bond of the alkyne, leading to activation of the alkyne moiety and subsequent formation of Cu(I) acetylides.Aryl alkynes and silyl alkynes are the most commonly employed.However, the electron-deficient nature of the triple bond in SF 4alkynes presents challenges for their coordination with Cu(I) during the catalytic process.Nevertheless, it was speculated that this property might enable the direct formation of Cu-acetylides without alkyne activation via coordination.Additionally, the use of dual transition metal catalysts was anticipated due to the competitive activation of the S─F bond rather than the alkynes, allowing the breakdown of the critical SF 4 group. [9]erein, we present an aryl radical-mediated Cu-catalyzed cross-coupling reaction that overcomes the hurdles associated with the electron-deficient nature of terminal SF 4 alkynes without decomposition of the SF 4 -moiety (Figure 1e).This approach successfully coupled SF 4 -alkynes with primary, secondary, and tertiary alkyl iodides, further expanding the scope of this transformation.The late-stage functionalization of drug molecules with alkyl iodides using this methodology resulted in instantaneous synthesis of SF 4 -attached drug-like molecules.Moreover, we demonstrated derivatization of the resulting inner SF 4alkynes through 1,3-dipolar cycloaddition with nitrones or click reactions with azides, enabling the linear linkage of two different heterocycles.The ability to incorporate SF 4 moieties into drug-like molecules through Sonogashira-type cross-coupling reactions allows for the exploration of novel chemical spaces and the potential discovery of promising drug candidates.

Results and Discussion
Among the numerous reaction conditions reported in the literature for Sonogashira-and Sonogashira-type cross-coupling reactions, we were intrigued by Liang's aryl radical-mediated methodology [10] for several reasons.First, unlike common Pd-Cu dual-transition-metal systems, this method utilizes only Cu catalysis.Second, the SF 4 moiety remains stable under radical conditions.Therefore, we investigated the use of 2-(ethynyltetrafluoro- 6 -sulfaneyl)pyridine (1a, 1.0 equiv.), and iodocyclohexane (2a, 1.0 equiv.)as model substrates under Liang's conditions ([Cu(CH 3 CN) 4 ]BF 4 (25 mol%), 2-mesityl-1diazonium salt (2.0 equiv.),terpyridine (25 mol%), K 2 CO 3 (3.0equiv.) in DMSO, 50 °C, 30 min), yielding the desired product 3aa in 70% yield (Table 1, run 1).Encouraged by this suc-cess, we conducted further optimization studies.First, several reactions and catalyst variations were explored.Initial trials using Cu(I) as the catalyst (Table 1, runs 2-10) resulted in yields ranging from 42% to 78%.Subsequently, several Cu(II) catalysts were tested to obtain the desired products in moderate yields (Table 1, runs 11-14).After exploring various options, we found that [Cu(CH 3 CN) 4 ]PF 6 was the most efficient catalyst under these reaction conditions (Table 1, run 5).Considering the activation of the triple bond in 1a, we recognized the significance of the bases in the formation of the Cu-acetylide.Hence, we focused on reactions with different bases under [Cu(CH 3 CN) 4 ]PF 6 catalysis.Organic bases ( i Pr 2 NEt and Et 3 N) and inorganic bases (Cs 2 CO 3 and Na 2 CO 3 ) were also investigated (Table 1, runs 15-18).i Pr 2 NEt exhibited better compatibility, presumably because of its enhanced solubility compared with inorganic bases (Table 1, runs 17 and  18).Next, we focused on various diazonium salts.However, variations in the diazonium salts were detrimental to the transformation, resulting in the formation of byproducts such as aryl coupling products (Table S3, Supporting Information).Our attempts to use different solvents, such as DMF and CH 3 CN, led to lower yields (Table 1, runs 19 and 20; Table S4, Supporting Information ).Reactions in THF and DCM failed to furnish the desired product, and DMSO was identified as the optimal solvent, providing the desired product in 84% yield.During the investigations, we observed that in some cases, starting material 1a remained unconsumed and trace amounts of aryl coupling products were formed.To address this, we increased the equivalence of 2a, leading to an improvement in the transformation yield of 3aa by up to 98% (92% isolated yield; Table 1, run 21).Furthermore, reducing the amount of diazonium salt from 2.0 to 1.0 equiv.resulted in a decrease in yield from 98% to 46% ( Initially, we explored the reaction of terminal alkyne 1a with various secondary alkyl iodides 2a-o (Scheme 1 (I)) encompassing cyclic scaffolds 2a-n and acyclic groups 2o.The reaction proceeded smoothly with secondary alkyl iodides 2a-d bearing 6-, 7-, 8-, and 5-membered rings, yielding the corresponding coupling products 3aa-3ad in high yields (85%-92%).Additionally, cyclic ethers 2e-g, cyclic thioether 2h, pyrrolidine 2i, and azetidine 2j were accepted as the coupling partners, affording products 3ae-3aj in high yields.Furthermore, various functional groups, such as 1,3-dioxolane 2k, pyrimidine 2l, indane 2m, and ethoxy 2n, were well tolerated under the reaction conditions, providing the desired products 3ak-3an in high yields.Moreover, the methodology was successfully extended to a coupling reaction with acyclic secondary alkyl iodide 2o, delivering 3ao in 81% yield.Next, we investigated the cross-coupling reaction between terminal SF 4 -alkyne 1a and primary alkyl iodides.A diverse range of primary alkyl iodides 2p-t, containing both cyclic and acyclic alkyl groups, was well tolerated, yielding the desired coupling products 3ap-3at (Scheme 1 (I)) in good to high yields.Notably, aryl-Br 2r, isoindoline 2s, and polyethylene glycol 2t moieties were also compatible with these reaction conditions.Tertiary alkyl iodides with adamantane 2u and 4-methyltetrahydro-2H-pyran 2v were further attempted and reacted smoothly with 3a to furnish the desired C sp -C sp3 coupling products 3au and 3av (Scheme 1 (I)) in moderate yields of 74% and 35% respectively.Furthermore, 4-CF 3 bearing SF 4 -pyridine 1b reacted steadily with secondary and tertiary cyclic alkyl iodides to afford the desired products 3ba and 3bu, respectively, in good yields.
Building on our success with heteroaromatic substrates, we expanded the scope of this methodology to the synthesis of SF 4alkynes 3 (Scheme 1 (III)) bearing diverse perfluorinated alkyl chains.While short fluoroalkyl groups, such as CF 3 and C 2 F 5 , are often found in drug structures, longer perfluoroalkyl chains are also useful moieties for the design of agrochemicals and decoy molecules to deceive P450BM3. [11]The reactions proceeded smoothly with various perfluoroalkyl iodides, providing the corresponding SF 4 -alkyne products 3 (Scheme 1 (III)) in good yields up to 87%.Notably, while the majority of products exhibited high yields, the lower yield of the perfluorodecyl-ethyl product 3aaa (32%) could be attributed to the highly fluorous character of 2aa.
To determine the relevance of this Sonogashira cross-coupling reaction, we performed robustness screening experiments in the presence of various heterocyclic pharmacophores.Interestingly, even in the presence of a range of nitrogen-containing hetero-cycles, such as quinoline, isoquinoline, pyrazine, and pyrazole, the reaction proceeded well, and there was no significant loss in the yields of 3a, which clearly indicated the tolerance of common heterocyclic pharmacophores under the designed reaction conditions (Table 2).
Motivated by this high functional tolerance, we further examined the reaction of 1a with alkyl iodide 2an in the presence or  absence of alkyl bromide 2an-Br (Scheme 2).Interestingly, the reaction proceeded solely with alkyl iodide 2an (Scheme 2a) and not alkyl bromide 2an-Br (Scheme 2b), even in a mixture of 2an and 2an-Br (Scheme 2c).Thus, performing the reaction with 2an and 2an-Br in one pot furnished 3aan in 70% yield, and 2an-Br was recovered in 83% yield.Therefore, these experiments suggest that the C─Br bonds are quite inactive toward the designed methodology and have high selectivity for C─I bonds.Further details of the reaction using alkyl bromides are provided in Supporting Information.
While all SF 4 -alkyne coupling products 3 are attractive as rod-like linear compounds with a length of 6.23 Å, they offer even greater potential for further chemical transformations based on established alkyne chemical reactions.To demonstrate this, we performed downstream modifications of 3aa as shown in Scheme 3. First, alkyne click chemistry of the synthesized compound 3aa with phenyl azide under metal-free and heating conditions for two days resulted in the formation of triazole-SF 4pyridine 4 in a regioselective manner with a yield of 23%.Additionally, 1,3-dipolar cycloaddition of alkyne 3aa to nitrone in the presence of Et 3 N led to the formation of isoxazoline-SF 4-pyridine 5 in 74% yield and regioselectivity.
Mechanistic studies were performed to gain insights into the reaction mechanism.Radical trap experiment with 3.0 equiv. of TEMPO confirmed that the reaction proceeds via a radical pathway (Figure 2a).TEMPO completely suppressed the formation of coupled products, and the formation of TEMPO adducts 6 and 2iodo-1,3,5-trimethylbenzene 7 was detected by GC-MS (see Supporting Information), further supporting the radical mechanism.
Based on the results obtained from our experiments and relevant literatures, [10] we propose a plausible mechanism outlined in Figure 2b.Initially, the terminal hydrogen of SF 4 -alkyne 1 undergoes deprotonation by i Pr 2 NEt, followed by a reaction with the Cu(I) ligand, leading to the formation of LnCu(I)-alkyne complex 8. Subsequently, the aryl diazonium salt oxidizes the LnCu(I)alkyne complex 8 to Cu(II)-alkyne 10, generating the aryl radical 9.The aryl radical 9 then reacts with alkyl iodide 2, resulting in the formation of the alkyl radical 11 and aryl iodide.The alkyl radical 11 undergoes oxidative addition to Cu(II) complex Scheme 3. Synthetic application of PySF 4 -Alkyne cross coupling product 3aa.Reactions are carried out 0.1 mmol scale 3aa. 10 yielding a Cu(III) alkyne complex 12.This Cu(III) complex 12 undergoes reductive elimination, leading to the formation of the coupled product 3, accompanied by the release of the Cu(I) ligand complex.The Cu(I) complex then continues the catalytic cycle, promoting the desired reaction to furnish the product 3.

Conclusion
In conclusion, we have successfully reported a Sonogashira-type cross-coupling reaction between electron-deficient SF 4 -attached alkynes and alkyl iodides, yielding excellent results under mild conditions.The reaction proceeds via a radical pathway facilitated by a diazonium salt and Cu catalysis, demonstrating its versatility by accepting various coupling partners, including primary, secondary, and tertiary alkyl iodides.This protocol enabled the synthesis of more than 50 novel pyridine/pyrimidine-SF 4 -alkyne compounds, some of which featured perfluorinated alkyl groups.Furthermore, we demonstrated late-stage C sp -C sp 3coupling of biologically relevant molecules, rendering this series of SF 4 -compounds attractive bio-isosteres of linear molecules such as p-substituted benzenes, cubanes, and bicyclopentanes (BCP).We anticipate that this approach will significantly advance the field of linear drug design, opening new possibilities for the development of innovative drug candidates.The findings presented herein highlight the potential impact of SF 4 -containing molecules in the pharmaceutical industry and pave the way for further research in this emerging field.

Scheme 1 .
Scheme 1. Substrate scope of 1 and 2. I) Scope of different alkyl iodides.II) Scope of heterocyclic parts of alkynes.III) Scope of perfluorinated alkyl iodides.IV) Late-stage cross-coupling reactions of biologically attractive molecules.V) Functional group tolerance.All reactions were carried out with 0.2 mmol of SF 4 terminal alkyne 1 except for IV and V) 0.1 mmol of 1. [a] Isolated yields are shown, and [b] 19 FNMR yields are shown in parentheses.[c] Gram scale reaction.

Scheme 2 .
Scheme 2. The comparisons of reactivity of alkyl iodide 2an and bromide 2an-Br.

Table 1 ,
run 22).Control experiments were performed to verify the significance of the various factors involved in the transformation.A reaction attempted in the absence of the diazonium salt yielded compromised results (Table 1, run 23).Additional control experiments without a
a) All reaction carried out 0.1 mmol of 1a; b) 19 F yield; c) DMF as solvent; d)

Table 2 .
Tolerance of various heterocyclic pharmacophores under Sonogashira cross-coupling reaction conditions for 1a and 2a.