Fluorous Phase Approach to α-Hydroxytropolone Synthesis

Abstract

α-Hydroxytropolones (αHTs) are troponoids that demonstrate inhibition against an array of therapeutically significant targets, making them potential drug leads for several human diseases. We have utilized a recently discovered one-pot three-component oxidopyrylium cycloaddition in a solid-supported synthesis of αHTs. Though the procedure is time efficient and generates assay-ready molecules, the system suffers from low yields and an inability to perform reaction modifications on resin-bound intermediates. In order to combat these issues with the solid-phase platform, we incorporated fluorous tags into our synthetic route. Through the implementation of fluorous phase chemistry, we demonstrate a substantial increase in the overall yield of αHTs, as well as an ability to execute metal-catalyzed cross-coupling and amide coupling on fluorous tagged intermediates. We also show that tagged molecules can be separated from non-fluorous impurities, and vice versa, by utilizing fluorous liquid-liquid and solid-phase extractions. Hence, these proof-of-principle investigations describe the viability of a fluorous phase approach to αHT synthesis, and its potential to serve as a combinatorial technique to produce structurally diverse substrates.

Graphical abstract

Introduction

α-Hydroxytropolones (αHTs) are highly oxygenated troponoids with therapeutic potential against a wide variety of biological targets.¹ Their activity is generally associated with the contiguous array of three oxygen atoms, which bind to and inhibit many bridged dinuclear metalloenzymes (Scheme 1A). Our lab has been investigating an oxidopyrylium cycloaddition/ring-opening procedure that has facilitated the synthesis of a structurally diverse αHT library (Scheme 1B).^2,3 With these molecules in hand, we are currently conducting medicinal chemistry studies related to HIV,⁴ hepatitis B,⁵ herpes simplex virus,⁶ aminoglycoside resistance,⁷ and fungal pathogens.⁸

Scheme 1.

(A) The most favorable binding mode of αHTs to dinuclear metalloenzymes. (B) Oxidopyrylium cycloaddition/ring-opening route used to synthesize αHTs.

During our studies, we recently discovered a one-pot three-component oxidopyrylium cycladdition that generates new oxabicyclic compounds through alcohol incorporation (Scheme 2A).^9,10 Additionally, benzyl alcohol-derived oxabicyclic intermediates produced from this reaction can be directly converted to αHTs when subjected to methanesulfonic acid (Scheme 2B). Considering this information, we developed a solid-phase synthesis of αHTs utilizing polystyrene-supported benzyl alcohol (Scheme 2C).¹¹ The method is extremely time efficient (i.e. a batch of 8 molecules can be produced in 1.5 days) and provides compounds with assay-ready purity. However, improvements to the system are necessary for it to be completely viable as a combinatorial technique for αHT synthesis. First, though yields for the process are practical for biological screening, they are substantially low overall (i.e. 5-20% yield for 9 substrates). Secondly, the platform does not permit modification of resin-bound intermediates, as conditions for both Suzuki coupling and trifluoroacetic acid (TFA) deprotection cause unintended cleavage (results not reported).

Scheme 2.

(A) Illustration of three-component oxidopyrylium cycloaddition, highlighting alcohol incorporation. (B) Utilization of three-component oxidopyrylium cycloaddition in αHT synthesis. (C) Overview of solid-phase synthesis of αHTs

In order to address the issues corresponding with solid-phase synthesis, we turned our attention to a fluorous phase approach. Fluorous phase chemistry allows the successful integration of solution-phase reaction conditions and provides the ability to monitor reactions through conventional analytical techniques that cannot be performed on resin-bound molecules.¹² Also, in relation to traditional organic synthesis, fluorous tagged compounds can be purified easily and efficiently by exclusive techniques based on fluorine-fluorine interactions. These methods include fluorous solid phase extraction (FSPE)¹³ and fluorous liquid-liquid extraction (FLLE)¹⁴ that utilize perfluorinated silica and solvents, respectively. The following describes our attempts at synthesizing αHTs by employing fluorous phase synthesis.

Results and Discussion

The integration of fluorous phase chemistry into our synthetic route poses three challenges: 1) verify that fluorous tagged alcohols could successfully be incorporated into oxabicyclic products, 2) assess how well the tagged compounds work with fluorous separation techniques, and 3) demonstrate the capability to remove the tagged segment and form the final product. Our studies began with the preparation of the fluorous tagged benzyl alcohol derivative 10, which can be generated in high yields through a previously reported approach.¹⁵ Generally, the one-pot three-component oxidopyrylium cycloaddition is conducted under sealed conditions with a surplus of alcohol (5 equivalents in relation to monomeric ylide) to drive the incorporation.⁹ Considering the large molecular weight of 10, slight modifications to our originally optimized procedure were required (Scheme 3A). For instance, the quantity of alcohol was reduced (i.e. stoichiometric with monomeric ylide) to avoid excessive use of starting material. Also, incorporation time was increased to accommodate the lower alcohol concentration. We were able to synthesize the fluorous tagged oxabicyclic compound 11a with a 34% yield using these adjustments, which is consistent for aryl-containing products obtained through the previously described method.

Scheme 3.

Newly optimized conditions with fluorous tagged benzyl alcohol 10 for small scale (A) and scale-up (B) procedures.

^aEquivalents are calculated on the basis of monomeric ylide.

However, when larger scale experiments were performed under these sealed conditions, reaction times were significantly increased, yielding little to no incorporation after 6 days. Taking into account our original mechanistic hypothesis for the one-pot three-component oxidopyrylium cycloaddition,⁹ we speculated that the lack of conversion was due to the increased generation of methanol (MeOH) that accompanies the scaled up procedure. Since MeOH is more prevalent and trapped within the reactor, the equilibrium of the alcohol incorporation is driven significantly from the fluorous tagged ylide back to 5a. In order to remove the MeOH byproduct from the reaction, the solvent system was changed to a 2:1 mixture of carbon tetrachloride (CCl₄) and perfluoromethylcyclohexane (C₇F₁₄) and the vessel was exposed to a light stream of argon (Scheme 3B). This solvent combination was chosen not only for its ability to dissolve the quantities of fluorous tagged alcohol necessitated by the reaction but for its higher boiling point (76 °C for both solvents) that supersedes MeOH (64.7 °C). The boiling temperature, along with the argon flow, allows for the evaporation of MeOH without drying out the reaction. Thus, the equilibrium of the alcohol incorporation is shifted towards the product and the reaction time is decreased. With these scale-up alterations, compound 11a was generated in a 38% yield in approximately 2 days, demonstrating a similar yield to its smaller scale counterpart.

Moving forward, we wanted to test if the newly obtained oxabicyclic product could be purified by fluorous phase separation techniques. Since the fluorine content of 11a (<50%) is not acceptable for FLLE,¹⁴ we chose FSPE as our starting point. Compound 11a was purposely contaminated with the non-fluorinated derivative 3a and loaded onto a 4g column of standard tridecafluoro functionalized silica gel (SiliaBond^® Tridecafluoro, Figure 1). Following a typical FSPE procedure,¹³ 3a was extracted with 80% MeOH in water and 11a was retrieved with acetone. Overall, 94% of 11a and 93% of 3a were recovered, proving that FSPE can be used to purify the fluorous tagged oxabicycles.

Figure 1.

Overview of FSPE on a mixture of fluorous 11a and nonfluorous 3a.

After demonstrating that it is possible to incorporate fluorous tagged alcohols into our oxabicyclic products and purify them with fluorous extraction, we directed our attention towards the ring-opening/debenzylation phase of 11a to αHT 4a. In our previous report,⁹ the benzyl-derived oxabicycle 6a underwent both ring-opening and debenzylation when exposed to liquid- and solid-phase methanesulfonic acid. Upon subjecting 11a to these conditions, we only observed the ring-opening product, as fluorous tagged tropolone 12a was isolated in quantitative yields (Scheme 4). The fluorous segment was removed upon returning to our original demethylation procedure (i.e. HBr in acetic acid), and a crude mixture of αHT 4a and fluorous tagged impurities was obtained. Considering the high fluorine content of the benzylated byproducts (>50%), both FSPE and FLLE were successfully used to purify the final product. With regard to the latter purification method, a 50% solution of methoxyperfluorobutane (HFE-(HFE-7100) in C₇F₁₄ proved very effective in separating unwanted contaminants from the αHT (refer to Supp. Info.). Compound 4a was obtained in 85% yield after FSPE and 95% yield after FLLE.

Scheme 4.

Ring-Opening/debenzylation of 11a to corresponding αHT 4a.

We found the result described above very promising since it indicates that the electron deficient nature of the fluorous tag introduces stability to our intermediates in harsh media. The added stabilization would allow us to perform modification at multiple stages of our synthetic route, unlike in the case of solid-phase synthesis. To test this hypothesis, we synthesized a pair of fluorous tagged oxabicycles (11j and 11k) with functional handles to attempt a series of experiments involving metal catalyzed cross coupling and amide coupling.

Initially, various palladium catalyzed Suzuki reactions, including systems with Pd(OAc)₂/JPhos or Pd(PPh)₄, were performed on compound 11j with benzyl and naphthyl boronic acids to complete conversion. The resulting oxabicycles were subjected to ring-opening/debenzylation conditions, and the corresponding αHTs were synthesized and confirmed through NMR analysis (results not shown). Unfortunately, pure final compounds could not be obtained with these procedures due to the presence of phosphine derived byproducts. Such impurities are not very soluble in the nonfluorous mobile phase of FSPE (80% MeOH in water or 70% acetonitrile (ACN) in water), and hence contaminate the fluorous eluent (acetone or MeOH) to the detriment of subsequent reaction purity.

In order to bypass this issue, we moved towards Suzuki conditions that utilize water-soluble SPhos as the activating ligand (Scheme 5).¹⁶ The phosphine byproducts were significantly removed from the intermediate but large amounts of solvent were still necessary during FSPE to achieve adequate purity (refer to Supp. Info.). However, the resulting bisphenyl oxabicyclic intermediate was successfully converted to pure αHT 4j after ring-opening/debenzylation. Studies on making purification for metal-catalyzed cross-coupling reactions on our fluorous tagged oxabicyclic products more efficient are currently ongoing. Nevertheless, this data demonstrates that modification through a fluorous phase approach is more tolerated than our previously reported solid-phase synthesis.

Scheme 5.

Overview of Suzuki cross-coupling route to αHT 4j.

^aEquivalents are calculated on the basis of monomeric ylide.

Moving forward to amide coupling reactions, compound 11k was used as a starting point to generate fluorous tagged intermediates with carboxylic acid functionalities (Scheme 6). Through a simple reaction sequence involving an acid-mediated deprotection followed by a ring-opening, 11k was converted to oxabicycle 11l and tropolone 12l near quantitatively. Each of these carboxylic acid derivatives was reacted with piperidine in the presence of PyBOP and N,N-diisopropylethylamine (DIPEA) to either produce compound 11m in 59% yield or 12m in 84% yield after FSPE and hexane wash. The corresponding αHT was then obtained by two different paths: 1) ring-opening/debenzylation of oxabicycle 11m and 2) debenzylation of tropolone 12m. Both steps were purified by FLLE to separate the fluorous tagged impurities from the final product 4m, demonstrating a divergent synthetic pathway to access an amide-derived αHT with fluorous phase chemistry.

Scheme 6.

Overview of divergent amide coupling on fluorous tagged oxabicycle and tropolone with piperidine to αHT 4m.

^aEquivalents are calculated on the basis of monomeric ylide.

Taking into account that a secondary amine was used during our initial amide modification experiments, we tested if primary amines could also undergo a two path approach to αHTs (Scheme 7). When oxabicycle 11l was subjected to PyBOP coupling conditions with 1-naphthylmethylamine, major decomposition was observed and the anticipated amide could not be obtained after FSPE. On the other hand, tropolone 12l was successfully converted to the corresponding amide-derivative 12n using the PyBOP procedure outlined above and taken further to αHT 4n in high yields for both steps. Thus, even though the oxabicycle route is limited to more sterically hindered secondary amines, the tropolone pathway can be used to achieve the coupling of both types of amines.

Scheme 7.

Overview of divergent amide coupling on fluorous tagged oxabicycle and tropolone with primary amine to αHT 4n.

Conclusion

In summary, we have demonstrated the first fluorous phase synthesis of αHTs utilizing one-pot three-component oxidopyrylium cycloaddition chemistry. In a prior report, we have shown that this reaction could be leveraged towards incorporation of solid-phase resins.¹¹ Though the method was time efficient and yielded assay-ready compounds, our solid-supported synthesis resulted in low yields and an inability to modify intermediates while attached to the resin. By incorporating fluorous tags instead of solid supports, we have significantly improved reaction output and generated intermediates that can undergo modification through metal catalyzed cross coupling and amide coupling. Studies are currently ongoing to fully optimize this platform in order to make it a combinatorial technique for αHTs.

Experimental Section

General Information

All starting materials and reagents were purchased from commercially available sources and used without further purification, with the exception of CH₂Cl₂, which was purified on a solvent purification system prior to reactions.^{17 1}H, ¹⁹F, and ¹³C NMR shifts were measured using the solvent residual peak as the internal standard and reported as follows: chemical shift, multiplicity (s = singlet, bs = broad singlet, d = doublet, t = triplet, dd = doublet of doublets, q = quartet, m = multiplet), coupling constant (Hz), integration. Infrared (IR) spectral bands are characterized as broad (br), strong (s), medium (m), and weak (w). Mass spectra were recorded on a spectrometer by the electrospray ionization (ESI) technique with a time-of-flight (TOF) mass analyzer. Microwave reactions were performed via the Biotage Initiator (external IR temperature sensor). Where noted, reaction products were purified via silica gel chromatography using a Biotage^® Isolera Prime, with Biotage^® SNAP Ultra 10 g or 25 g cartridges, in a solvent system of ethyl acetate (EtOAc) in hexane. Fluorous solid-phase extraction (FSPE) was performed using columns packed with SiliaBond^® Tridecafluoro silica gel.

Synthesis of Fluorous Tagged 3-Benzyloxy-8-Oxabicyclo[3.2.1]octenes: 5-methyl-3-((4-(perfluorooctyl)benzyl)oxy)-6-phenyl-8-oxabicyclo[3.2.1]octa-3,6-dien-2-one (11a)

Method A. Dimer 9 (24 mg, 0.0856 mmol), benzyl alcohol 10 (91 mg, 0.172 mmol, 2 equiv.), and CH₂Cl₂ (0.25 M, 344 μL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 0.5-2 mL), and the reaction mixture was heated to 60 °C in silicon bath oil for 36 hr. Phenyl acetylene (378 μL, 3.44 mmol, 40 equiv.) was added to the sealed tube, and the reaction mixture was subjected to microwave irradiation at 100 °C for 1 hr. The reaction was then immediately purified by column chromatography (silica (10 g), 0% EtOAc/hexane to 20% EtOAc/hexane gradient over 23 column volumes), giving 11a as a yellow oil (43 mg, 34% yield). Method B,. Dimer 9 (200 mg, 0.714 mmol) and benzyl alcohol 10 (751 mg, 1.43 mmol, 2 equiv.) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 10-20 mL) and the vessel was purged with argon (3x). Carbon tetrachloride (2.75 mL) and perfluoromethylcyclohexane (1.38 mL) were added to the sealed reactor, and the reaction mixture was heated to 60 °C in silicon bath oil for 48 hr under argon. Phenyl acetylene (3.14 mL, 28.6 mmol, 40 equiv.) was added to the sealed tube, and the reaction mixture was subjected to microwave irradiation at 100 °C for 2.5 hr. The reaction mixture was evaporated and purified by column chromatography (silica (25 g), 0% EtOAc/hexane to 15% EtOAc/hexane gradient over 17 column volumes), giving 11a as a yellow oil (401.9 mg, 38% yield). FSPE Protocol A,. A Biotage^® SNAP Ultra 10 g cartridge was filled with SiliBond^® Tridecafluoro silica gel (4 g) and preconditioned with 50:50 MeOH/H₂O. Compound 11a (46.4 mg, 0.063 mmol) was contaminated with 3a (13.5 mg, 0.056 mmol), and the mixture was loaded onto the cartridge in CHCl₃. The nonfluorous material was provided by elution with 80:20 MeOH/H₂O (40 mL), while the fluorous component was obtained by elution with acetone (40 mL). The resulting fractions were then concentrated, giving 11a as a yellow oil (43.4 mg, 94% recovery) and 3a as a white solid (12.6 mg, 93% recovery). The cartridge was washed with MeOH (40 mL) and acetone (40 mL) prior to reuse, and it was recycled up to 30 times. The 1H NMR for 3a was consistent with previously reported data.^2a Characterization of 11a. R_f = 0.27 in 15% EtOAc in hexanes. IR (thin film, KBr) 3060 (w), 2981 (w), 2935 (w), 1712 (s), 1606 (s), 1211 (s), 1150 (s) cm^-1. ¹H Hz, 2H), 7.38 – 7.31 (m, 3H), 7.20 – 7.15 (m, 2H), 6.29 (d, J = 2.4 Hz, 1H), 6.24 (s, 1H), 5.02 (d, J = 2.5 Hz, 1H), 4.97 (d, J = 12.7 Hz, 1H), 4.85 (d, J = 12.7 Hz, 1H), 1.64 (s, 3H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.8 (t, J = 9.9 Hz, 3F), -110.7 (t, J = 14.5 Hz, 2F), -121.1 – -121.4 (m, 2F), -121.7 – -122.1 (m, 6F), -122.56 – -122.89 (m, 2F), -126.0 – -126.2 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 189.7 (s), 158.8 (s), 144.6 (s), 140.0 (s), 133.0 (s), 128.9 (s), 128.9 (t, J = 24.4 Hz), 128.8 (s), 127.4 (s), 127.4 (t, J = 6.4 Hz), 126.0 (s), 123.2 (s), 122.0 (s), 86.5 (s), 86.1 (s), 68.8 (s), 22.0 (s). HRMS (ESI TOF) m/z: [M + Na]⁺ calcd for C₂₉H₁₇F₁₇O₃Na 759.0802.

6-(3-bromophenyl)-5-methyl-3-((4-(perfluorooctyl)benzyl)oxy)-8-oxabicyclo[3.2.1]octa-3,6-dien-2-one (11j)

Dimer 9 (200 mg, 0.714 mmol) and benzyl alcohol 10 (751 mg, 1.43 mmol, 2 equiv.) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 10-20 mL), and the vessel was purged with argon (3x). Carbon tetrachloride (2.75 mL) and perfluoromethylcyclohexane (1.38 mL) were added to the sealed reactor, and the reaction mixture was heated to 60 °C in silicon bath oil for 36 hr under argon. 1-Bromo-3-ethynylbenzene (1.72 mL, 14.3 mmol, 20 equiv.) was added to the sealed tube, and the reaction mixture was subjected to microwave irradiation at 100 °C for 3 hr. The reaction mixture was evaporated and purified by column chromatography (silica (25 g), 0% EtOAc/hexane to 15% EtOAc/hexane gradient over 27 column volumes), giving 11j as a yellow oil (366.7 mg, 32% yield). Due to the high cost of 1-bromo-3-ethynylbenzene, fractions containing it were also concentrated (2.029 g, 87% recovery (2.321 g would be 100% theoretical yield of unreacted product)). R_f = 0.29 in 15% EtOAc in hexanes. IR (thin film, KBr) 3064 (w), 2982 (w), 2936 (w), 1713 (s), 1605 (s), 1588 (m), 1210 (s), 1151 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.58 (d, J = 8.3 Hz, 2H), 7.52 – 7.44 (m, 3H), 7.38 – (m, 1H), 7.24 – 7.18 (m, 1H), 7.09 – 7.05 (m, 1H), 6.34 (d, J = 2.4 Hz, 1H), 6.23 (s, 1H), 5.02 (d, J = 2.5 Hz, 1H), 4.96 (d, J = 12.6 Hz, 1H), 4.85 (d, J = 12.6 Hz, 1H), 1.63 (s, 3H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.8 (t, J = 9.5 Hz, 3F), -110.7 (t, J = 14.5 Hz, 2F), -121.2 (bs, 2F), - – -122.1 (m, 6F), -122.6 – -122.9 (m, 2F), -126.0 – -126.3 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 189.3 (s), 157.5 (s), 144.7 (s), 139.9 (s), 135.1 (s), 131.8 (s), 130.3 (s), 129.1 (s), (t, J = 24.3 Hz), 127.4 (s), 127.4 (t, J = 6.5 Hz), 124.8 (s), 124.5 (s), 123.0 (s), 121.5 (s), 86.4 (s), 86.1 (s), 68.8 (s), 22.0 (s). HRMS (ESI TOF) m/z: [M + Na]⁺ calcd for 836.9904; Found 836.9907.

tert-butyl 5-methyl-2-oxo-3-((4-(perfluorooctyl)benzyl)oxy)-8-oxabicyclo[3.2.1]octa-3,6- diene-6-carboxylate (11k)

Dimer 9 (200 mg, 0.714 mmol) and benzyl alcohol 10 (751 mg, 1.43 mmol, 2 equiv.) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 10-20 mL), and the vessel was purged with argon (3x). Carbon tetrachloride (2.75 mL) and perfluoromethylcyclohexane (1.38 mL) were added to the sealed reactor, and the reaction mixture was heated to 60 °C in silicon bath oil for 48 hr under argon. tert-Butyl propiolate (1.96 mL, 14.3 mmol, 20 equiv.) was added to the sealed tube, and the reaction mixture was subjected to microwave irradiation at 100 °C for 1 hr. The reaction mixture was evaporated and purified by column chromatography (silica (25 g), 0% EtOAc/hexane to 10% EtOAc/hexane gradient over 24 column volumes), giving 11k as a white solid (550.5 mg, 51% yield). M.P. = 109-112 °C. R_f = 0.21 in 10% EtOAc in hexanes. IR (thin film, KBr) 2982 (w), 2938 (w), 1703 (s), 1615 (m), 1604 (m), 1197 (s), 1146 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.58 (d, J = 8.3 Hz, 2H), 7.49 (d, J = 8.3 Hz, 2H), 6.98 (d, J = 2.5 Hz, 1H), 6.15 (s, 1H), 5.01 (d, J = 2.5 Hz, 1H), 4.86 (d, J = 12.5 Hz, 1H), 4.79 (d, J = 12.5 Hz, 1H), 1.72 (s, 3H), 1.48 (s, 9H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.7 (t, J = 9.9 Hz, 3F), -110.7 (t, J = 14.5 Hz, 2F), -121.1 –-121.3 (m, 2F), -121.7 – -122.0 (m, 6F), -122.5 – -122.8 (m, 2F), -126.0 – -126.2 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 188.7 (s), 162.2 (s), 151.1 (s), 143.8 (s), 139.9 (s), 137.6 (s), 128.9 (t, J = 24.4 Hz), 127.5 (s), 127.4 (t, J = 6.4 Hz), 121.9 (s), 86.0 (s), 85.6 (s), 82.4 (s), 68.6 (s), 28.2 (s), 21.4 (s). HRMS (ESI-TOF) m/z [M + Na]⁺ calcd for C₂₈H₂₁F₁₇O₅Na 783.1010 Found 783.1011.

5-methyl-2-oxo-3-((4-(perfluorooctyl)benzyl)oxy)-8-oxabicyclo[3.2.1]octa-3,6-diene-6-carboxylic acid (11l)

Oxabicycle 11k (435.8 mg, 0.573 mmol) and 50% trifluoroacetic acid in CH₂Cl₂ (2.2 mL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was subjected to microwave radiation at 70 °C for 2 min. The organic layer was concentrated under reduced pressure and then subjected to azeotropic removal with CH₂Cl₂ (5 × 2 mL), resulting in a yellowish solid. The solid was washed with hexane (5 × 4 mL) and dried under vacuum to give 11l as an off-white solid (396.7 mg, >95% yield). M.P. = 144-146 °C. R_f = 0.25 in 10% MeOH in CH₂Cl₂. IR (thin film, KBr) 2867 (br), 1708 (m), 1676 (s), 1616 (m), 1605 (m), 1199 (s), 1147 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.59 (d, J = 8.3 Hz, 2H), 7.48 (d, J = 8.3 Hz, 2H), 7.31 (d, J = 2.5 Hz, 1H), 6.17 (s, 1H), 5.10 (d, J = 2.5 Hz, 1H), 4.84 (d, J = 12.3 Hz, 1H), 4.79 (d, J = 12.3 Hz, 1H), 1.77 (s, 3H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.9 (t, J = 9.9 Hz, 3F), -110.8 (t, J = 14.4 Hz, 2F), -121.3 (bs, 2F), -121.8 – -122.1 (m, 6F), -122.8 (bs, 2F), -126.0 – -126.7 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 188.1 (s), 167.9 (s), 148.9 (s), 143.9 (s), 142.3 (s), 139.6 (s), 129.1 (t, J = 24.3 Hz), 127.6 (s), 127.4 (t, J = 6.4 Hz), 121.5 (s), 86.3 (s), 85.5 (s), 68.8 (s), 21.3 (s). HRMS (ESI-TOF) m/z: [M + Na]⁺ calcd for C₂₄H₁₃F₁₇O₅Na 727.0384 Found 727.0387.

5-methyl-3-((4-(perfluorooctyl)benzyl)oxy)-6-(piperidine-1-carbonyl)-8-oxabicyclo[3.2.1] octa-3,6-dien-2-one (11m)

A solution of oxabicycle 11l (60 mg, 0.0852 mmol), PyBOP (48.8 mg, 0.0937 mmol, 1.1 equiv.), and DIPEA (32.6 μL, 0.187 mmol, 2.2 equiv.) in CH₂Cl₂ (0.04 M, 2 mL) was stirred at room temperature for 20 min in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL). Piperidine (9.3 μL, 0.0937 mmol, 1.1 equiv.) was then added to the sealed vessel, and the reaction mixture was subjected to microwave radiation at 85 °C for 5 min. The mixture was concentrated under reduced pressure and purified by FSPE as described in Protocol A. The acetone layer was evaporated, producing a yellow solid. The solid was washed with hexane (5 × 2 mL) to give 11m as a white solid (38.7 mg, 59% yield). M.P. = 173-175 °C. R_f = 0.29 in 40% EtOAc in hexanes. IR (thin film, KBr) 3070 (w), 2942 (m), 2862 (m), 1711 (s), 1609 (s), 1212 (s), 1151 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.58 (d, J = 8.3 Hz, 2H), 7.50 (d, J = 8.3 Hz, 2H), 6.35 (s, 1H), 6.29 (d, J = 2.4 Hz, 1H), 5.11 (d, J = 2.4 Hz, 1H), 4.85 (d, J = 12.2 Hz, 1H), 4.80 (d, J = 12.2 Hz, 1H), 3.67 – 3.45 (m, 4H), 1.76 – 1.66 (m, 2H), 1.65 – 1.49 (m, 7H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.7 (t, J = 9.9 Hz, 3F), -110.7 (t, J = 14.4 Hz, 2F), -121.1 – -121.3 (m, 2F), -121.7 – -122.1 (m, 6F), -122.7 (bs, 2F), -125.8 – -126.5 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 188.3 (s), 163.6 (s), 150.7 (s), 143.9 (s), 140.1 (s), 128.8 (t, J = 24.5 Hz), (s), 127.8 (s), 127.3 (t, J = 6.5 Hz), 123.6 (s), 87.4 (s), 87.1 (s), 68.7 (s), 47.9 (s), 42.9 (s), 27.0 (s), 25.8 (s), 24.7 (s), 20.7 (s). HRMS (ESI-TOF) m/z: [M + H]⁺ calcd for 772.1350; Found 772.1352.

Synthesis of Fluorous Tagged Benzyloxy Tropolones: 2-hydroxy-5-methyl-7-((4-(perfluorooctyl)benzyl)oxy)-4-phenylcyclohepta-2,4,6-trien-1-one (12a)

To a solution of 11a (135.2 mg, 0.184 mmol) in CH₂Cl₂ (0.16 M, 1.15 mL) was added methanesulfonic acid (477 μL, 7.36 mmol, 40 equiv). The reaction mixture was stirred at room temperature for 1 h, at which time it was quenched with phosphate buffer (pH 7, 0.1 M, 5 mL) and extracted with CHCl₃ (3 × 5 mL). The organic layer was washed with brine (15 mL), dried over Na₂SO₄, filtered, and concentrated under reduced pressure to yield a golden brown solid. This solid was washed with hexane (5 × 2 mL) and dried under vacuum to give 12a as a light brown solid (131.6 mg, >95% yield). M.P. = Decomposes at 155 °C. IR (thin film, KBr) 3220 (br), 2917 (w), 2858 (w), 1575 (m), 1564 (m), 1199 (s), 1148 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.67 (d, J = 8.6 Hz, 2H), 7.63 (d, J = 8.7 Hz, 2H), 7.46 – 7.35 (m, 4H), 7.31 (s, 1H), 7.24 – 7.20 (m, 2H), 5.39 (s, 2H), 2.18 (s, 3H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.8 (t, J = 9.9 Hz, 3F), -110.6 (t, J = 14.3 Hz, 2F), -121.2 (bs, 2F), -121.6 – -122.1 (m, 6F), -122.7 (bs, 2F), -126.0 – -126.2 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 169.5 (s), 160.3 (s), 156.6 (s), 145.2 (s), 143.5 (s), 140.7 (s), 135.0 (s), 128.9 (t, J = 24.3 Hz), 128.7 (s), 128.2 (s), 127.9 (s), 127.6 (s), 127.6 (s), 127.5 (t, J = 6.5 Hz), 122.3 (s), 71.0 (s), 26.5 (s). HRMS (ESI-TOF) m/z: [M + H]⁺ calcd for C₂₉H₁₈F₁₇O₃ 737.0979 Found 737.0982.

6-hydroxy-2-methyl-5-oxo-4-((4-(perfluorooctyl)benzyl)oxy)cyclohepta-1,3,6-triene-1-carboxylic acid (12l)

To a solution of 111 (116.1 mg, 0.165 mmol) in CH₂Cl₂ (0.1 M, 1.42 mL) was added trifluoromethanesulfonic acid (58.2 μL, 0.659 mmol, 4 equiv). The reaction mixture was stirred at room temperature for 30 min, at which time it was quenched with phosphate buffer (pH 7, 1 M, 5 mL). The CH₂Cl₂ was drained and the remaining residue was extracted with EtOAc (3 × 5 mL). All organics were collected, dried over Na₂SO₄, filtered, and concentrated under reduced pressure to give a yellow solid. The solid was washed with 50% CH₂Cl₂ in hexanes (6 × 2 mL) and dried under vacuum to give 12l as a light yellow solid (107.3 mg, 92% yield). M.P. = 163-166 °C. IR (thin film, KBr) 3442 (br), 1699 (m), 1570 (m), (s) cm^-1. ¹H NMR (400 MHz, Acetone) δ 7.85 (s, 1H), 7.49 (s, 1H), 5.50 (s, 2H), 2.56 (s, 3H). ¹⁹F NMR (376 MHz, Acetone) δ -81.6 (t, J = 10.1 Hz, 3F), -110.7 (t, J = 14.5 Hz, 2F), -121.6 – -121.9 (m, 2F), -122.2 – -122.6 (m, 6F), -123.2 (bs, 2F), -126.6 – -126.9 (m, 2F). ¹³C NMR (100 MHz, Acetone) δ 171.8 (s), 170.2 (s), 160.2 (s), 159.2 (s), 142.5 (s), 137.1 (s), 133.8 (s), 128.9 (s), 128.6 (t, J = 24.3 Hz), 127.9 (t, J = 6.5 Hz), 125.9 (s), 117.0 (s), 71.0 (s), 25.4 (s). HRMS (ESI-TOF) m/z: [M + H]⁺ calcd C₂₄H₁₄F₁₇O₅ 705.0564 Found 705.0570.

2-hydroxy-5-methyl-7-((4-(perfluorooctyl)benzyl)oxy)-4-(piperidine-1-carbonyl)cyclohepta-2,4,6-trien-1-one (12m)

A solution of tropolone 12l (50 mg, 0.071 mmol), PyBOP (40.6 mg, 0.0781 mmol, 1.1 equiv.), and DIPEA (27.2 μL, 0.156 mmol, 2.2 equiv.) in EtOAc (0.04 M, 1.67 mL) was stirred at room temperature for 20 min in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL). Piperidine (7.7 μL, 0.0781 mmol, 1.1 equiv.) was then added to the sealed vessel, and the reaction mixture was subjected to microwave radiation at 85 °C for 5 min. The reaction mixture was concentrated under reduced pressure and purified by FSPE as described in Protocol A. The acetone layer was evaporated, producing a yellow solid. The solid was washed with hexane (5 × 2 mL) and dried under vacuum to give 4.30m as a light yellow solid (45.8 mg, 84% yield). M.P. = 183-186 °C. IR (thin film, KBr) 3175 (br), 2942 (w), 2861 (w), 1621 (m), 1568 (m), 1202 (s), 1151 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 8.93 (bs, 1H), 7.63 (s, 4H), 7.17 (s, 1H), 7.16 (s, 1H), 5.36 (d, J = 12.8 Hz, 1H), 5.32 (d, J = 12.8 Hz, 1H), 3.84 – 3.75 (m, 1H), 3.69 – 3.61 (m, 1H), 3.27 – 3.13 (m, 2H), 2.34 (s, 3H), 1.76 – 1.62 (m, 4H), 1.60 – 1.43 (m, 2H). ¹⁹F NMR (376 MHz, CDCl₃) δ -80.7 (t, J = 9.9 Hz, 3F), -110.7 (t, J = 14.4 Hz, 2F), -121.1 – -121.3 (m, 2F), -121.7 – -122.0 (m, 6F), -122.5 – -122.8 (m, 2F), -125.9 – -126.3 (m, 2F). ¹³C NMR (100 MHz, CDCl₃) δ 170.4 (s), 168.6 (s), 161.2 (s), 157.2 (s), 140.2 (s), 138.6 (s), 132.8 (s), 129.1 (t, J = 24.3 Hz), 127.6 (s), 127.5 (t, J = 6.5 Hz), 126.6 (s), 116.2 (s), 71.0 (s), 47.9 (s), 42.6 (s), 26.6 (s), 25.7 (s), 24.5 (s), 24.2 (s). HRMS (ESI-TOF) m/z: [M + H]⁺ calcd for C₂₉H₂₃F₁₇NO₄ 772.1350 Found 772.1351.

6-hydroxy-2-methyl-N-(naphthalen-1-ylmethyl)-5-oxo-4-((4-(perfluorooctyl)benzyl)oxy) cyclohepta-1,3,6-triene-1-carboxamide (12n)

A solution of tropolone 12l (51.7 mg, 0.0734 mmol), PyBOP (42 mg, 0.0807 mmol, 1.1 equiv.), and DIPEA (28.1 μL, 0.161 mmol, 2.2 equiv.) in EtOAc (0.04 M, 1.67 mL) was stirred at room temperature for 20 min in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL). 1-Naphthylmethylamine (11.8 μL, 0.0807 mmol, 1.1 equiv.) was then added to the sealed vessel, and the reaction mixture was subjected to microwave radiation at 85 °C for 5 min. The reaction mixture was concentrated under reduced pressure and purified by FSPE as described in Protocol A. The acetone layer was evaporated, producing a yellow solid. The solid was washed with hexane (5 × 2 mL) and ACN (5 × 2 mL) and dried under vacuum to give 12n as a light yellow solid (53.8 mg, 87% yield). M.P. = 175-177 °C. IR (thin film, KBr) 3239 (br), 1666 (m), 1541 (m), 1199 (s), 1146 (s)

Synthesis of α-Hydroxytropolones: 2,7-dihydroxy-4-methyl-5-phenylcyclohepta-2,4,6-trien-1-one (4a)

FSPE Purification. Tropolone 12a (50 mg, 0.068 mmol) and 33% HBr in acetic acid (197 μL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was heated to 120 °C in silicon bath oil for 1 hr. The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5 mL). The organic layer was evaporated under reduced pressure and purified by FSPE as described in Protocol A. The 80:20 MeOH/H₂O layer was concentrated to yield 4a as a white solid (13.2 mg, 85% yield). The ¹H NMR for 4.24a was consistent with previously reported data.⁷ FLLE Purification, Protocol A. Tropolone 4.30a (50.5 mg, 0.0686 mmol) and 33% HBr in acetic acid (199 μL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was heated to 120 °C in silicon bath oil for 1 hr. The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5 mL). The organic layer was dried over Na₂SO₄, filtered, and evaporated under reduced pressure to give a crude oil. The residue was taken up in ACN (2 mL) and washed with 50% HFE-7100 in perfluoromethylcyclohexane (5 × 2 mL). The ACN layer was concentrated to yield 4.24a as a brown oil (15.5 mg, >95% yield). The ¹H NMR for 4.24a was consistent with previously reported data.⁷

4-([1,1’-biphenyl]-3-yl)-2,7-dihydroxy-5-methylcyclohepta-2,4,6-trien-1-one (4j)

FSPE Protocol B. Compund 11j (36.1 mg, 0.0443 mmol), phenyl boronic acid (54 mg, 0.443 mmol, 10 equiv.), water soluble SPhos (23 mg, 0.0443 mmol, 1 equiv.), K₂CO₃ (61.2 mg, 0.443 mmol, 10 equiv.), Pd(OAc)₂ (5 mg, 0.0222 mmol, 0.5 equiv.) and 33% ACN in water (0.01 M, 4.5 mL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was heated to 50 °C in silicon bath oil for 5 hr. The reaction mixture was cooled to room temperature and immediately loaded to a Biotage SNAP Ultra 10 g cartridge that was filled with SiliBond^® Tridecafluoro silica gel (4 g) and preconditioned with 30:70 ACN/H₂O. Any remaining residue in the reactor was loaded onto the column with CHCl₃, and the nonfluorous material was provided by elution with ACN/H₂O (0% ACN/H₂O to 70% ACN/H₂O gradient over 12 column volumes) and MeOH/H₂O (80% MeOH/H₂O over 3 column volumes). The fluorous component was obtained by elution with MeOH (100% MeOH over 5 column volumes) and the resulting fractions were then concentrated, giving S2 as a yellow oil (26.4 mg, 73% crude yield).

To a solution of oxabicycle S2 (26.4 mg, 0.0325 mmol) in CH₂Cl₂ (0.1 M, 325 μL) was added trifluoromethanesulfonic acid (11.5 μL, 0.13 mmol, 4 equiv.). The reaction mixture was allowed to stir for 30 minutes, after which time it was quenched with phosphate buffer (pH 7, 0.1 M, 5 mL) and extracted with CHCl₃ (3 × 5 mL). The organic layer was dried over Na₂SO₄, filtered, and concentrated under reduced pressure. The crude mixture was then dissolved in 33% HBr in acetic acid (100 μL) and heated to 120 °C for 1 hr in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL). The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5 mL). The organic layer was then dried over Na₂SO₄, filtered, and concentrated to give a crude oil. The residue was purified by FLLE according to Protocol A, giving 4j as a brown oil (8.9 mg, 90% yield over two steps). IR (thin film, KBr) 3248 (br), 3059 (w), 2928 (w), 1525 (s), 1277 (s), 1231 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.65 – 7.59 (m, 4H), 7.55 (s, 1H), 7.54 – 7.42 (m, 4H), 7.40 – 7.34 (m, 1H), 7.24 – 7.20 (m, 1H), 2.31 (s, 3H). ¹³C NMR (100 MHz, CDCl₃) δ 167.3, 157.9, 156.6, 144.2, 143.8, 141.8, 140.6, 139.2, 129.2, 129.0, 127.8, 127.3, 127.3, 127.2, 126.6, 124.5, 124.3, 26.6. HRMS (ESI-TOF) m/z [M + H]⁺ calcd for C₂₀H₁₇O₃ 305.1172 Found 305.1176.

2,7-dihydroxy-4-methyl-5-(piperidine-1-carbonyl)cyclohepta-2,4,6-trien-1-one (4m)

Method A. To a solution of oxabicycle 11m (39.4 mg, 0.0511 mmol) in CH₂Cl₂ (0.1 M, 511 μL) was added trifluoromethanesulfonic acid (18 μL, 0.204 mmol, 4 equiv.). The reaction mixture was allowed to stir for 30 minutes, after which time it was quenched with phosphate buffer (pH 7, 0.1 M, 5 mL) and extracted with CHCl₃ (3 × 5 mL). The organic layer was dried over Na₂SO₄, filtered, and concentrated under reduced pressure producing a brown residue. The crude mixture was then dissolved in 33% HBr in acetic acid (150 μL), and heated to 120 °C for 1 hr in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL). The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5mL). The organic layer was then dried over Na₂SO₄, filtered, and concentrated to give a crude oil. The residue was purified by FLLE according to Protocol A, giving 4m as a brown oil (10.4 mg, 77% yield over two steps). Method B. Tropolone 12m (40 mg, 0.0518 mmol) and 33% HBr in acetic acid (150 μL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was heated to 120 °C in silicon bath oil for 1 hr. The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5 mL). The organic layer was dried over Na₂SO₄, filtered, and evaporated under reduced pressure to give a crude oil. The residue was purified by FLLE according to Protocol A, giving 4m as a brown oil (12.5 mg, 92% yield). Characterization of 4m. IR (thin film, KBr) 3246 (br), 2940 (s), 2859 (m), 1627 (s), 1541 (s), 1265 (s), 1232 (s) cm^-1. ¹H NMR (400 MHz, CDCl₃) δ 7.47 (s, 1H), 7.27 (s, 1H), 3.86 – 3.77 (m, 1H), 3.70 – 3.62 (m, 1H), 3.25 – 3.13 (m, 2H), 2.41 (s, 3H), 1.77 – 1.44 (m, 6H). ¹³C NMR (100 MHz, CDCl₃) δ 168.7, 168.2, 158.3, 157.8, 137.5, 137.0, 124.3, 118.6, 47.9, 42.7, 26.5, 25.7, 24.5, 24.1. HRMS (ESI-TOF) m/z: [M + H]⁺ calcd for C₁₄H₁₈NO₄ 264.1230 Found 264.1230.

4,6-dihydroxy-2-methyl-N-(naphthalen-1-ylmethyl)-5-oxocyclohepta-1,3,6-triene-1-carboxamide (4n)

Tropolone 12n (53 mg, 0.0628 mmol) and 33% HBr in acetic acid (185 μL) were placed in a sealed tube reactor (Biotage^® microwave reaction vial, 2-5 mL), and the reaction mixture was heated to 120 °C in silicon bath oil for 1 hr. The reaction mixture was cooled to room temperature, quenched with phosphate buffer (pH 7, 0.1 M, 5 mL), and extracted with CHCl₃ (3 × 5 mL). The organic layer was dried over Na₂SO₄, filtered, and evaporated under reduced pressure to give a crude oil. The residue was purified by FLLE according to Protocol A, giving 4n as a brown oil (18.8 mg, 89% yield). IR (thin film, KBr) 3263 (br), 3060 (m), 1697 (s), 1640 (s), 1536 (s), 1248 (s) cm^-1. ¹H NMR (400 MHz, Acetone) δ 8.28 (d, J = 8.3 Hz, 1H), 8.02 (bs, 1H), 7.96 (d, J = 8.0 Hz, 1H), 7.88 (d, J = 8.2 Hz, 1H), 7.64 – 7.46 (m, 4H), 7.42 (s, 1H), 7.34 (s, 1H), 5.07 (d, J = 5.7 Hz, 2H), 2.42 (s, 3H). ¹³C NMR (100 MHz, Acetone) δ 169.9, 169.1, 159.7, 158.3, 138.7, 138.0, 135.2, 134.9, 132.5, 129.6, 129.1, 127.6, 127.2, 126.8, 126.3, 124.7, 124.3, 119.9, 42.1, 24.3. HRMS (ESI-TOF) m/z: [M + H]⁺ calcd for C₂₀H₁₈NO₄ 336.1230 Found 336.1233.