Lead Diversification through a Prins-Driven Macrocyclization Strategy: Application to C13-Diversified Bryostatin Analogues

Abstract

The design, synthesis, and biological evaluation of a novel class of C13-diversified bryostatin analogues are described. An innovative and general strategy based on a Prins macrocyclization-nucleophilic trapping cascade was used to achieve late-stage diversification. In vitro analysis of selected library members revealed that modification at the C13 position of the bryostatin scaffold can be used as a diversification handle to regulate biological activity.

Bryostatin 1 is a marine-derived macrocyclic polyketide isolated from the bryozoan Bugula neritina (Figure 1).¹ It exhibits clinically relevant, highly potent, and diverse biological activities,² and is currently undergoing Phase I and Phase II clinical trials for treatment of a variety of cancers as well as Alzheimer’s disease.³ Bryostatin 1 has also been shown to reverse multidrug resistance,⁴ to promote immunostimulant effects,⁵ to enhance memory and learning in animal models,⁶ and to induce repair of neuronal damage resulting from stroke.⁷ More recently, bryostatin has also been shown to induce clearance of HIV viral reservoirs, an activity that could figure in the as yet unachieved goal of eradicating HIV/AIDS.⁸

Figure 1.

Bryostatin 1 (PKC binding affinity K_i = 1.4 nM)

The unique biological activities of this natural product might arise in part from its ability to target and modulate the protein kinase C (PKC) family of serine/threonine kinases. ⁹ PKCs are implicated in a wide range of cellular signal transduction cascades, including regulation of cell growth, modulation of cell membrane structure, and control of transcription. Dysfunction of certain PKC isozymes is implicated in a variety of pathologies.¹⁰ Although many ligands for PKC serve as kinase inhibitors at the ATP binding site, bryostatin 1 binds to the C1 regulatory domain, permitting up- or down-regulation of particular PKC isozymes. This additional activity of the natural product has many functional therapeutic ramifications, including induction of apoptosis in cancer cells, restoration of kinase activity in various disorders, and transcriptional regulation. In addition to interacting with PKC, bryostatin 1 also interacts with other C1-domain-containing proteins. For example, the natural product and its analogues have been shown to activate RasGRP1 in cultured T cells,¹¹ and a recent study has linked RasGRP activation with apoptotic induction in Toledo non-Hodgkin’s lymphoma.¹²

Despite its therapeutic potential, bryostatin 1, like many natural products, is isolated in variable low yields (1.4 × 10⁻⁴%) yields from its source.¹ In addition, although several members of the bryostatin family (bryostatins 1,¹³ 2,¹⁴ 3,¹⁵ 7,¹⁶ 9,¹⁷ and 16¹⁸) have been prepared by total synthesis during the past two decades, from about 40 to as many as 89 steps have been required to produce the natural products. In 1988, anticipating the problems associated with obtaining a practical supply of bryostatin and with the view that bryostatin was not evolved for clinical use, we initiated an alternative approach to address problems associated with both its supply and its potential clinical performance.¹⁹ Through bioassay-informed computer modeling, our function-oriented synthesis (FOS) strategy²⁰ sought to identify those structural features of the natural product that might affect its biological activity. The resulting hypotheses were then used to design new structures with potentially superior activities that might be produced in a practical manner.²¹ By using this computer-assisted synthesis-informed design approach, we reported in 1998 the first designed analogues of bryostatin (bryologs) with potencies that match that of bryostatin.²² We subsequently reported several analogues (Figure 2) whose potencies exceed that of the natural product while emulating its functional activity in various biological assays. Significantly, these analogues can be prepared in far fewer steps than required to synthesize the natural product (19 longest linear steps, 29 total steps).²³ Importantly, this FOS strategy provides access to analogues that exhibit PKC isoform selectivities corresponding to that of the natural product, as well as access to other analogues that exhibit unique selectivities that are not observed with the natural products but which might be exploited as tools or leads in various preclinical studies.²⁴

Figure 2.

PKC binding affinities of synthetic analogues of bryostatin 1

We recently showed that a Prins-driven macrocyclization reaction, a variant of our initially reported macrotransace-talization strategy, can be used to prepare a new class of B-ring tetrahydropyran (THP)-based bryostatin analogues. ²⁵ This transformation is presumed to proceed through formation of an oxocarbenium ion (as inferred in our earlier work) that is subsequently captured by an allylsilane nucleophile to furnish a C13-substituted exocyclic olefin (Scheme 1).²⁶ Deprotection of this compound with hydrogen fluoride·pyridine gives analogue 2, which can be elaborated by selective ozonolysis and Horner–Wadsworth–Emmons olefination, followed by global deprotection, to give bryostatin analogues 3 and 4. Importantly, the THP-based B-ring bryostatin analogues exhibit significant potencies that exceed the in vitro antiproliferative activity of 1 by up to two orders of magnitude. In addition, in terms of its PKC binding affinity, the natural C13-(Z)-enoate 4 is one of the most active analogues that has been discovered to date.

Scheme 1.

Synthetic strategies for preparing substituted THP-derived B-ring bryostatin analogues

Diversification at C13, which is needed to explore the role of this center in PKC activity, required cleavage of the C13 exo-alkene, ketone reduction, alcohol activation, and nucleophilic substitution. We reasoned that a Prins macrocyclization–nucleophilic trapping process would, in a single operation, form the macrocycle, assemble the B-ring, establish two new stereocenters, and permit incorporation of various functional groups at C13 (Scheme 1).²⁷ Although the Prins macrocyclization reaction is emerging as a versatile synthesis strategy that has been used successfully in our work^17,25a and in syntheses of kendomycin, ²⁸ neopeltolide,²⁹ and polycavernoside A,³⁰ the reaction has not been used to diversify a lead scaffold, a strategy that might be exploited in advancing many natural hydropyranyl lead compounds.³¹ In the current study, we sought to explore the applicability of Prins macrocyclization–nucleophilic trapping methods in synthesizing bryostatin scaffolds, with the aims of establishing whether modifications to the THP-based B-ring confer advantageous biological activities, such as increased potency, efficacy, or therapeutic index, and of exploring the generality of this diversification strategy. These studies were also conducted to set the stage for positron emission tomography (PET) and NMR labeling studies on the structures and functions of PKC–ligand complexes. Here, we report the successful implementation of this strategy through the efficient and convergent synthesis of a novel class of C13-diversified bryostatin analogues, together with their preliminary biological evaluation.

Our initial synthetic studies were aimed at a convergent preparation of the cyclization precursor 11. Allylation of the previously reported lactone 5,³² followed by reduction of the resulting hemiacetal gave the allyl derivative 6 in 55% yield (Scheme 2). The terminal olefin group was then oxidized with potassium osmate and sodium periodate to give the corresponding aldehyde 7 in 87% yield. Conversion of aldehyde 7 into the corresponding homoallylic alcohol was accomplished by an asymmetric Brown allylation, and the resulting secondary alcohol was protected by treatment with chloro(triethyl)silane (67% yield, two steps). To complete the synthesis of the spacer domain 9, the C1 benzyl group was removed by treatment with lithium naphthalenide (87% yield), and the resultant alcohol was oxidized to give the carboxylic acid in a two-step sequence (75% yield).

Scheme 2.

Synthesis of spacer domain 9. Reaction conditions: (a) AllMgBr, THF, −78 °C, 1.5 h; (b) TES, TFA, CH₂Cl₂, −30 °C, 1.5 h, 55%; (c) K₂OsO₄, NaIO₄, 1,4-dioxane–H₂O (3:1), 87%; (d) (−)-(+)-methoxy(diisopinocampheyl)borane, AllMgBr, Et₂O, −78 °C, 3 h, then 30 % H₂O₂, 1 M aq NaOH, 1 h, 69%; (e) TESCl, imidazole, CH₂Cl₂, 1 h, 98%; (f) lithium naphthalenide, THF, −30 °C, 30 min, 87%; (g) (1) Pr₄N⁺ RuO₄ ⁻, NMO, CH₂Cl₂, r.t., 30 min; (2) NaClO₂, NaH₂PO₄, CH₂Cl₂, r.t., 30 min, 75% (two steps).

Spacer domain 9 was then coupled to the recognition domain 10 by means of a Yamaguchi esterification (Scheme 3). Upon TES deprotection, the resulting cyclization precursor 11 was produced in a 58% yield over three steps.

Scheme 3.

Synthesis of cyclization precursor, 11. Reaction conditions: (a) 9, 2,4,6-trichlorobenzoyl chloride, Et₃N, toluene, r.t., 3 h; (b) 10, DMAP, toluene, r.t., 1 h; (c) PPTS, 20% H₂O–THF, 20 h, 58% (three steps).

Next, we turned our attention to the evaluation of the Prins macrocyclization–nucleophilic trapping transformation with a halogen nucleophile (Scheme 4).³³ Preliminary investigations indicated that enal 11 was unstable when exposed to various Lewis acids at 0 °C or above; optimizations of reaction conditions were therefore undertaken at lower temperatures. Although only modest conversion was detected when the halo-Prins macrocyclization was conducted with bromo(trimethyl)silane and iron(III) bromide at −78 °C (Scheme 4), the transformation was effective when performed at −40 °C and gave the C13 bromide in 57% yield (82% based on recovered starting material). Deprotection with hydrogen fluoride·pyridine resulted in clean production of analogue 12.

Scheme 4.

Synthesis of bryostatin analogues by Prins macrocyclization (only the B-rings are shown for products 12–18; the macrocyclization yields are given). Reaction conditions: (A, halide–Prins macrocyclization): (i) TMSBr, FeBr₃, CH₂Br₂, −40 °C, 2 h; (ii) HF·pyridine, −78 °C to r.t.; (B, fluoro-Prins macrocyclization): HF·pyridine, −78 °C to r.t.; (C, Prins–Ritter macrocyclization): TMSOTf, nitrile, CH₂Cl₂, −40 °C, 3 h; H₂O, −40 °C to r.t.; (D, Prins–Friedel–Crafts macrocyclization): (i) TMSOTf, arene (solvent), −40 °C, 2 h; (ii) HF·pyridine, −78 °C to r.t. In the case of B, C11 is the TES-protected alcohol in the starting material.

A key advantage of the Prins macrocyclization–nucleophilic trapping process is its ability to facilitate efficient synthesis of the THP-based B-ring. For example, upon coupling of fragments 9 and 10 by Yamaguchi esterification, the resulting product could be treated with hydrogen fluoride·pyridine to initiate directly a fluoro-Prins macrocyclization. This remarkable cascade assembles the macrocycle, introduces a fluorine atom at the C13 position, and achieves global removal all the silyl protecting groups to give the fluoro derivative 13 (Scheme 4). This transformation is only the second example of a hydrogen fluoride promoted Prins reaction,³⁴ and only the second report of a fluoro-Prins macrocyclization.²⁸ Although not yet optimized in terms reaction time, the reaction provides a one-step route to fluorine-labeled analogues. Importantly, the sequence to produce 13 represents the shortest synthesis (longest linear sequence 19, and 31 total steps) of a potent THP-derived B-ring bryostatin analogue (see below).

Because of the efficiency of the initial macrocyclization process, we next investigated the applicability of a range of nucleophiles, with the aim of accessing novel C13 substitution patterns. For example, employing a nitrile as the trapping agent in a tandem Prins–Ritter reaction would be expected to provide bryostatin analogues substituted with amide groups at the C13 position.³⁵ Although few examples of nitrogen-centered nucleophiles have been reported to participate in the Prins reaction, they have been restricted to formation of simple THP moieties. The application of this method to complex polyfunctionalized molecular scaffolds has yet to be realized.

The tandem Prins–Ritter macrocyclization was initially examined by treating a solution of enal 11 in acetonitrile with trimethylsilyl triflate added dropwise at −40 °C (Scheme 4). Interestingly, this protocol resulted in the desired macrocyclization with a concomitant global deprotection to give analogue 14 directly. Although only a modest yield (22%) was obtained, the one-step reaction involves four distinct transformations and many more mechanistic steps. Further optimization of solvents, amounts of reagents, and their rate of addition did not result in any appreciable increase in the yield of the desired product. However, the order of addition proved to be crucial in improving the efficiency of the reaction. For example, when a solution of enal 11 in acetonitrile was added dropwise to a solution of trimethyl triflate in dichloromethane at −40 °C, 14 was isolated in a markedly improved yield of 75%. Furthermore, this reverse addition protocol could be applied to other nitriles. When benzonitrile was used as the trapping agent, a 55% yield of analogue 15 was obtained, whereas only a 20% yield was found under the initial standard conditions.

The Prins macrocyclization–nucleophilic trapping process was further developed through the investigation of a tandem Prins–Friedel–Crafts macrocyclization.³⁶ This method proved to be successful for the production of C13-aryl bryostatin analogues (Scheme 4). In this reaction, 11 was dissolved in the appropriate arene and then treated with trimethylsilyl triflate. Lower reaction temperatures were again necessary to prevent decomposition of the starting material. When anisole was used as the trapping agent, the corresponding C13-aryl products were isolated in 56% yield as a 2.3:1.0 ratio of the ortho- and para-regioisomers. The corresponding analogues 17 and 18 could then be obtained by hydrogen fluoride·pyridine mediated silyl deprotection. This process proved to be limited in scope to electron-rich arenes, as electron-deficient arenes gave little or no product. The reverse addition protocol did not improve the yield of the desired macrocyclization product.

To determine whether a representative example of this new C13-functionalized class of bryostatin analogues would effectively target PKC, we conducted a preliminary competitive binding assay with analogue 13. Importantly, this compound displayed a 1.6 nM affinity to rodent-brain-derived PKC (used to permit comparisons with previous studies) (Figure 3).^25a This remarkably high potency toward the enzyme mixture was comparable to that of the parent natural product. However, 13 can be prepared in as few as 31 total transformations, and is therefore a more easily accessible target for synthesis. Furthermore, 13 represents one of the most potent bryostatin analogues that has been reported to date, and is therefore a superb candidate for PET and NMR studies.

Figure 3.

Binding affinity of a new C13-substituted bryostatin analogue for rat-brain PKC

The development of the Prins macrocyclization–nucleophilic trapping method provided a novel class of C13-substituted THP-based B-ring bryostatin analogues for biological evaluation. These compounds were further investigated for their growth inhibitory effects against the K562 human erythroleukemia cell line (Table 1). In our initial analysis, analogue 12 showed a promising level of activity (EC₅₀ = 471 nM). Interestingly, in the examination of the corresponding C13-fluoride bryostatin analogue 13, we found a remarkable increase in potency (EC₅₀ = 14.2 nM). A double-digit nanomolar in vitro activity was also observed for each member of the C13-amide-derived bryostatin analogue series 14–16. Furthermore, the C13-aryl-based bryostatin analogues 17 and 18 showed activity with potencies that were in the single-digit to low-double-digit nanomolar range. These results collectively show that, although the B-ring is positioned outside the proposed binding pocket of the enzyme, modifications to the spacer domain can result in significant changes in the biological profile of the corresponding bryostatin analogue, as would be expected from its potential role in translocation and membrane association. Thus, the C13 position appears to be a tunable element that can be utilized to regulate biological activity and pharmacological performance.

Table 1.

Effective Concentrations of C13-Functionalized Analogues against Human Leukemia Cell Line K562 after 48 Hours of Incubation: Average of Two Experiments.

Analogue	12	13	14	15	16	17	18
EC50 (nM)	471	14.2	55.0	30.0	45.2	5.8	23.3

In summary, we have described a novel Prins macrocyclization–nucleophilic trapping diversification strategy for accessing C13 halogen-, carbon-, and nitrogen-substituted THP-derived bryostatin analogues. This transformation was applicable to a range of nucleophiles, including halides, nitriles, and arenes, and it served as an efficient method for preparing a range of bryostatin analogues for further in vitro analysis. Notably, modifications at the C13 position of the bryostatin scaffold were found to affect the binding and potency of the analogues. Further biological evaluation of these analogues, including isozyme-selective activity, is currently underway and will be reported in due course.

Unless otherwise noted, all reactions were performed under N₂ in flame- or oven-dried glassware. Mixtures were stirred by using Teflon-coated magnetic stirrer bars. Reactions were monitored by TLC on 0.25-mm silica gel 60F plates with a fluorescent indicator (Merck). Plates were visualized by treatment with an acidic p-anisaldehyde or KMnO₄ stain and gentle heating. Products were purified by column chromatography on silica gel 60 (230–400 mesh) (EM) using the solvent systems indicated.

When necessary, solvents and reagents were purified before use. THF was distilled from sodium benzophenone ketyl. Et₂O, CH₂Cl₂, and toluene were passed through a drying column of alumina (Solv-Tek Inc.) under N₂ pressure. Anhyd DMF and DMSO were obtained from Acros Organics. EtOAc, PE, pentane, and MeOH were obtained from Fisher Scientific. DMSO used to prepare biological samples was obtained from Fisher BioReagents (Class III). DMSO used in the PKC and cellular assay protocols was purchased from Aldrich. TMSOTf was obtained from Aldrich and distilled under N₂ in a Hickman apparatus immediately before use. Powdered 4-Å MS (<5 μm) were purchased from Aldrich and stored in a vacuum oven (140 °C, 90 mmHg). All other reagents were purchased from Aldrich and used as received without additional purification. (³H)phorbol dibutyrate [(³H)PDBu] was obtained from American Radiolabeled Chemicals, Inc. (St. Louis, MO) as a soln in acetone. Samples prepared for biological evaluation were purified by preparative HPLC with a H₂O–MeCN gradient using a Varian Pro-Star (model 320) system equipped with an AllTech Alltima C18 column (10 μm, 10 × 250 mm).

NMR spectra were recorded on Varian INOVA 500 (¹H at 500 MHz, ¹³C at 125 MHz), Varian 400 (¹H at 400 MHz, ¹³C at 100 MHz), or Varian INOVA 600 MHz (¹H at 500 MHz, ¹³C at 150 MHz) spectrometers, as noted. ¹H chemical shifts are reported relative to the residual solvent peak (CHCl₃; δ = 7.26 ppm, benzene; δ = 7.15 ppm). ¹³C chemical shifts are reported relative to the ¹³C signals for the deuterated solvent (CDCl₃; δ = 77 ppm). IR spectra were recorded on a PerkinElmer 1600 Series Fourier-transform spectrometer. Optical rotation data were obtained by using a JASCO DIP. High-resolution mass spectra were recorded at the Stanford University Mass Spectrometry facility.

Protein Kinase C Isozyme Mixture Binding Assay

Preparation of assay materials

In a polypropylene vial, a buffer soln was prepared by diluting 1 M aq Tris-HCl (pH 7.4, 1 mL), 1 M aq KCl (2 mL), 0.1 M aq CaCl₂ (30 μL), and bovine serum albumin (40 mg) with deionized H₂O (20 mL). The buffer was stored on ice until required. Phosphatidyl serine (Avanti Polar Lipids, 10 mg/mL in CHCl₃, 350 μL) was isolated by removing the CHCl₃ under a stream of N₂ and then suspended as vesicles by adding the prepared buffer soln (3.5 mL), followed by sonication (Branson Sonifier 250; power = 6; 40% duty cycle) for four 30-s periods with a 30 s rest between sonications. The resulting suspension was stored on ice until required. Assay Protein Kinase C (PKC) was prepared by dissolving an aliquot of a PKC rat-brain isozyme mixture (600 μL) in buffer soln (14 mL). This mixture was stored on ice for immediate use.

Preparation of dilutions of (³H)PDBu and test compounds

A commercial 1 mCi/mL soln of (³H)PDBu in acetone (American Radiolabeled Chemicals, Inc.) was diluted 10-fold in EtOH. This EtOH soln was further diluted by a factor of 16.7 (96 μL into 1504 μL) in DMSO for use in assays. The assay concentration of (³H)PDBu for any specific batch of dilutions was measured by competitive binding of (¹H)PDBu [IC₅₀ = assay (³H)PDBu concentration]. For the data presented above (Figure 3), the assay concentration of (³H)PDBu was 9.4 nM.

Dilutions of the test compounds were prepared in DMSO, with serially dilution from a high concentration of 2.0 μM (assay concentration 130 nM) by factors of three to a low concentration of 2.7 nM (assay concentration: 180 pM).

Assay protocol

Polypropylene assay vials (in triplicate) were charged with phosphatidyl serine vesicle suspension (60 μL), PKC soln (200 μL), and the diluted test compound (20 μL). Nonspecific (³H)PDBu binding was analyzed in triplicate by using analogue 4 (75 μM, 20 μL, assay concentration: 5 μM) instead of the test compound. Maximum (³H)PDBu binding was analyzed in triplicate by substitution of DMSO (20 μL) for the test compound. Finally, (³H)PDBu [diluted in DMSO (20 μL), assay concentration: 9.4 nM, determined by titration with (¹H)PDBu] was added to all the vials. The solns were mixed with a vortexer, incubated at 37 °C for 90 min, and placed on ice for 15 min. Filters (Whatman GF/B 21 mm) were treated by soaking them for 90 min in 10 vol% aq poly(ethyleneamine) (6 mL) diluted in H₂O (200 mL). A rinse buffer of 20 mM aq Tris (500 mL, pH 7.4) was prepared and cooled on ice for the duration of the incubation period and for the remainder of the assay. The contents of the assay vials were vacuum filtered through the poly(ethyleneamine)-soaked filter papers, washing the residual contents of the vial with additional rinse buffer (0.5 mL). The filters were then washed by dropwise addition of buffer (4.5 mL) and placed into scintillation vials. These vials were filled with Bio-Safe scintillation fluid (5 mL) and their radioactivity was immediately measured for 1 min each in a scintillation counter (Beckman LS 6000SC). Counts per minute (cpm) were averaged for each of the three triplicate dilutions. The data were then plotted [cpm versus log(concentration)] by using Prism software (GraphPad), and the value of the IC₅₀ was calculated by using one-site-competition least-squares regression analysis. K_i values were then calculated by means of the equation: K_i = IC₅₀/(1 + [(³H)PDBu])/[K_d of (³H)PDBu]. The K_d of (³H)PDBu was measured under these assay conditions to be 1.55 nM.

K562 Cellular Assay: MTT Cellular Viability Readout

K562 cells were cultured in RPMI medium 1640 (Gibco) supplemented with fetal bovine serum( FBS) (10% v/v), penicillin (100 units/mL), and streptomycin (50 units/mL) at 37 °C, under 5% CO₂ in air in a humidified incubator. Cells were subcultured every 72–96 h once the cellular concentration had reached > 1,000,000 cells/mL, to a starting concentration of 100,000 cells per mL. The cell concentration was determined with a hemocytometer. To assay the compounds, cells were plated in plasma-treated 96-well polystyrene plates at a density of 10,000–14,000 cells/well in 50 μL of culture medium, and incubated for 1 h while compound dilutions were prepared. Compounds were diluted from a 4 mM DMSO stock soln with culture medium to a concentration of 20 μM. Triplicate serial dilutions were then prepared in culture medium from 20 μM to 80 pM across 10 wells of a 96-well plate (diluted by factors of 4). Plates containing cells were dosed with the compound (50 μL/well) to afford final assay concentrations of 10 μM to 40 pM. The maximum DMSO concentration was 0.25%. Cells were then incubated at 37 °C under 5% CO₂–air in a humidified incubator for 48 h. A 5 mg/mL soln of thiazolyl blue tetrazolium bromide (MTT; Aldrich) in cell culture medium (10 μl) was then added to each well. The cells were allowed to incubate for an additional 2.5 h then lysed with 100 μL of a detergent solvent [Triton-X 100 (Aldrich) (20 mL) dissolved in 0.1 M HCl in i-PrOH (180 mL)]. After thorough mixing, the plates were analyzed with a VERSAmax tunable microplate reader (Molecular Devices) using SOFTmax Pro software (version 3.1.1), reading at 570 nm and subtracting at 690 nm. The sigmoidal dose–response curves generated by plotting the corrected signal versus the log(drug concentration) were analyzed by least-squares regression using Prism software (GraphPad) to generate the EC₅₀ values.

[(2R,6S)-6-((2R)-4-(Benzyloxy)-2-{[tert-butyl(diphenyl)silyl] oxy}butyl)tetrahydro-2H-pyran-2-yl]acetaldehyde (7)

An oven-dried round-bottomed flask equipped with stirrer bar was charged with terminal olefin 6³⁷ (468 mg, 0.862 mmol) and the flask was evacuated, backfilled with N₂. A 3:1 mixture of 1,4-dioxane and H₂O (16 mL) followed by 2,6-lutidine (0.200 mL, 0.276 mmol) were added at r.t by syringe through a septum. K₂OsO₄·2H₂O (1.5 mg, 1.72 mmol) and NaIO₄ (736 mg, 3.44 mmol) were added directly to the flask under N₂, and the mixture was stirred at r.t. for 15 h. The reaction was quenched with sat. aq NaHCO₃ (25.0 mL) and diluted with Et₂O (30 mL). The mixture was then extracted with Et₂O (3 × 30 mL) and the organic layers were combined, dried (MgSO₄), filtered, and concentrated in vacuo. The resulting crude product was purified by flash chromatography (10% EtOAc–PE) to give an oil; yield: 410 mg (87%); [α]_D ^24.2 +13.7 (c 1.65, CH₂Cl₂); R_f = 0.50 (15% EtOAc–PE) (stained with KMnO₄).

IR (thin film): 3069, 2931, 2856, 1726, 1427, 1110, 821, 737, 702 cm⁻¹.

¹H NMR (500 MHz, CDCl³): δ = 9.48 (t, 3 Hz, 1 H), 7.66 (d, 8 Hz, 4 H), 7.24–7.42 (m, 11 H), 4.35 (s, 2 H), 4.12 (pent, J = 7 Hz, 1 H), 3.47 (t, J = 7 Hz, 2 H), 3.42 (m, 1 H), 3.32 (m, 1 H), 2.23–2.36 (m, 2 H), 1.78 (m, 2 H), 1.70 (m, 1 H), 1.56 (m, 2 H), 1.48 (m, 1 H), 1.25–1.34 (m, 3 H), 1.09 (m, 1 H), 1.03 (s, 9 H).

¹³C NMR (125 MHz, CDCl³): δ = 201.9, 138.5, 135.9, 134.4, 134.3, 129.5 (×2), 128.2, 127.5, 127.4 (×2), 127.3, 74.5, 72.7, 72.5, 68.2, 66.8, 49.9, 44.4, 37.7, 31.4, 31.3, 30.9, 27.0, 23.3, 19.4.

HRMS (ES+): m/z calcd for C₃₄H₄₄NaO₄Si: 567.2901; found: 567.2906.

(2S)-1-[(2R,6S)-6-((2R)-4-(Benzyloxy)-2-{[tert-butyl(diphenyl) silyl]oxy}butyl)tetrahydro-2H-pyran-2-yl]pent-4-en-2-ol (8; R = H)

(−)-(+)-Methoxy(diisopinocampheyl)borane (460 mg, 1.45 mmol) was weighed into a flask in a dry box. Et₂O (15.0 mL) was added and the soln was cooled to −78 °C. A 1.0 M soln of CH₂=CHCH₂MgBr in Et₂O (1.21 mL, 1.21 mmol) was added and the mixture stirred for 5 min at −78 °C then for 1 h at r.t. The mixture was cooled to −78 °C, a soln of aldehyde 7 (440 mg, 0.808 mmol) in Et₂O (5 mL) was added, and the mixture was stirred for 2.5 h. The reaction was quenched with 30% H₂O₂ (1 mL) and 1 M aq NaOH (400 μL), and the mixture was stirred at 30 °C for 30 min. The soln was then extracted with Et₂O (3 × 30 mL) and the organic layers were combined, dried (MgSO₄), filtered, and concentrated in vacuo. The crude product was dissolved in MeOH (3 × 10 mL) and concentrated in vacuo to remove the remaining borane. The resulting product was purified by flash chromatography (10% EtOAc–PE) to give an oil; yield: 55 mg (69%); [α]_D ^24.2 −2.4 (c 4.0, CH₂Cl₂); R_f = 0.40 (10% EtOAc–PE) [stained with anisaldehyde (red), UV active].

IR (thin film): 3497, 3070, 2932, 2857, 1640, 1589, 1427, 1106, 702 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.68 (d, 8 Hz, 4 H), 7.24–7.42 (m, 11 H), 5.85 (m, 1 H), 5.07–5.12 (m, 2 H), 4.36 (s, 2 H), 4.08 (dt, J = 2, 6 Hz, 1 H), 3.68 (m, 1 H), 3.64 (s, 1 H), 3.47 (dt, J = 3, 7 Hz, 2 H), 3.34 (m, 1 H), 3.27 (m, 1 H), 2.10–2.26 (m, 2 H), 1.78 (q, J = 7 Hz, 2 H), 1.62–1.70 (m, 2 H), 1.55–1.60 (m, 1 H), 1.39–1.49 (m, 3 H), 1.24–1.32 (m, 2 H), 1.10–1.20 (m, 1 H), 1.04 (s, 9 H).

¹³C NMR (125 MHz, CDCl₃): δ = 138.5, 135.9, 135.8, 135.2, 134.3, 134.2, 129.5 (×2), 128.2, 127.6, 127.5 (×2), 127.3, 116.8, 78.6, 74.8, 72.7, 70.9, 68.4, 66.7, 44.5, 42.2, 41.9, 37.5, 31.8, 31.3, 26.9, 23.2, 19.4.

[((1S)-1-{[(2R,6S)-6-((2R)-4-(Benzyloxy)-2-{[tert-butyl(diphenyl) silyl]oxy}butyl)tetrahydro-2H-pyran-2-yl]methyl}but-3-en-1-yl)oxy](triethyl)silane (8, R = TES)

Imidazole (22.5 mg, 0.331 mmol) was added to a soln of alcohol 8 (R = H; 55 mg, 0.094 mmol) in CH₂Cl₂ (1.0 mL) under ambient air. Upon dissolution of the solids, TMSCl (28 μL, 0.166 mmol) was added in one portion to give a white suspension. The mixture was stirred for 15 min and then diluted with sat. aq NH₄Cl (5 mL) and Et₂O (5 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with Et₂O (3 × 5 mL). The organic phases were combined, dried (MgSO₄), filtered, and concentrated. The crude product was purified chromatography (silica gel, 5% EtOAc–PE) to give a clear colorless oil; yield: 65 mg (98%); [α]_D ^24.2 +15.2 (c 3.35, CH₂Cl₂); R_f = 0.50 (5% EtOAc–pentane) [stained with anisaldehyde (green), UV active].

IR (thin film): 3071, 2932, 2875, 2857, 1641, 1589, 1455, 1427, 1106, 1005, 736, 707 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.71 (m, 4 H), 7.24–7.42 (m, 11 H), 5.84 (m, 1 H), 5.06 (s, 1 H), 5.03 (d, J = 4 Hz, 1 H), 4.37 (s, 2 H), 4.13 (dt, J = 2, 5 Hz, 1 H), 3.86 (p, J = 6 Hz, 1 H), 3.48 (dt, J = 3, 7 Hz, 2 H), 3.29 (m, 2 H), 2.17–2.29 (m, 2 H), 1.79 (q, J = 7 Hz, 2 H), 1.45–1.75 (m, 6 H), 1.28–1.38 (m, 2 H), 1.07 (s, 9 H), 1.02 (m, 2 H), 0.97 (t, J = 8 Hz, 9 H), 0.61 (q, J = 8 Hz, 6 H).

¹³C NMR (125 MHz, CDCl₃): δ = 138.6, 135.8, 135.0, 134.4 (×2), 129.4 (×2), 128.2, 127.5 (×2), 127.4, 127.3, 116.9, 74.4, 74.3, 72.7, 68.7, 68.5, 66.9, 44.8, 43.5, 42.0, 37.7, 31.9, 31.3, 27.0, 23.6, 19.4, 6.9, 5.0.

HRMS (ES+): m/z calcd for C₄₃H₆₄NaO₄Si₂: 723.4235; found: 723.4246.

(2S)-1-[(2R,6S)-6-((2R)-2-{[tert-Butyl(diphenyl)silyl]oxy}-4-hydroxybutyl) tetrahydro-2H-pyran-2-yl]pent-4-en-2-ol

A ~1 M soln of lithium naphthalenide in THF was prepared by dissolving Li metal (high sodium content; 347 mg, 50 mmol) and naphthalene (7.69 g, 60 mmol) in THF (50 mL) with sonication under N₂ at r.t. The resulting deep green soln was stored under N₂ in a Schlenk flask at −20 °C until required. Solns of this reagent could be stored for several weeks with only slight loss of activity.

A soln of 8 (R = TES; 307 mg, 0.438 mmol) in THF (7.0 mL) was cooled in a CO₂/MeCN bath (external temperature −25 °C) and then the ~1.0 M soln of lithium naphthalenide in THF (5.39 mL, 5.39 mmol) was added dropwise in 200-μL portions at 1 min intervals. When addition of the lithium naphthalenide reagent was complete, the resulting opaque green mixture was stirred for 10 min until the starting material was completely consumed (TLC). The reaction was then quenched with sat. aq NH₄Cl (5 mL), which caused the color to disappear immediately. The biphasic mixture was diluted with H₂O (10 mL) and Et₂O (20 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with Et₂O (2 × 20 mL). The organic layers were combined, dried (MgSO₄), filtered, and concentrated. The residue was purified by chromatography (silica gel, 10 to 20% EtOAc–PE) to give a clear, colorless oil; yield: 232 mg (87%); [α]_D ^24.2 +10.7 (c 1.1, CH₂Cl₂); R_f = 0.20 (10% EtOAc–pentane) [stained with anisaldehyde (green), UV active].

IR (thin film): 3443, 2932, 2876, 2857, 1642, 1427, 1110, 1068, 1046, 702 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.69 (m, 4 H), 7.36–7.45 (m, 6 H), 5.79 (m, 1 H), 5.02 (d, J = 2 Hz, 1 H), 4.98 (dd, J = 2, 7 Hz, 1 H), 4.08 (p, J = 5 Hz, 1 H), 3.75–3.81 (m, 2 H), 3.69 (m, 1 H), 3.19 (m, 1 H), 2.99 (m, 2 H), 2.31 (br s, 1 H), 2.09–2.24 (m, 2 H), 1.81–1.89 (m, 1 H), 1.59–1.76 (m, 4 H), 1.39–1.56 (m, 3 H), 1.16–1.29 (m, 2 H), 1.07 (s, 9 H), 1.02 (m, 1 H), 0.94 (t, J = 8 Hz, 9 H), 0.57 (q, J = 8 Hz, 6 H).

¹³C NMR (125 MHz, CDCl₃): δ = 135.9, 135.8, 134.9, 134.0, 133.8, 129.7, 129.6, 127.7, 127.6, 116.9, 75.2, 74.6, 71.0, 68.5, 59.6, 43.7, 43.4, 41.8, 39.1, 31.9, 31.3, 27.0, 23.4, 19.3, 6.9, 5.0.

HRMS (ES+): m/z calcd for C₃₆H₅₈NaO₄Si₂: 633.3766; found: 633.3775.

(3S)-3-{[tert-Butyl(diphenyl)silyl]oxy}-4-((2S,6R)-6-{(2S)-2-[(triethylsilyl)oxy]pent-4-en-1-yl}tetrahydro-2H-pyran-2-yl)butanoic acid (9; Spacer Domain)

An oven-dried vial was charged with the alcohol prepared as described above (30 mg, 0.049 mmol) and powdered 4-Å MS (250 mg). The vessel was purged twice with N₂ and then a soln of NMO (22 mg, 0.188 mmol) in CH₂Cl₂ (0.50 mL) was added. A soln of Pr₄N⁺ RuO₄ ⁻ (2.2 mg, 0.00625 mmol) in CH₂Cl₂ (0.50 mL) was added and the resulting green mixture was stirred at r.t. for 10 min until the starting material was completely consumed (TLC). After 15 min, the mixture was loaded directly onto a short pad of silica gel and eluted with 20% EtOAc–pentane. The filtrate was concentrated to give a colorless oil that was used directly in the next step.

The oil was dissolved in t-BuOH (1.20 mL) under air, and H₂O (600 μL) and 2-methylbut-2-ene (450 μL) were added successively. The resulting cloudy mixture was stirred vigorously and cooled in an ice water bath. The suspension was treated with sequentially with NaH₂PO₄ (75 mg, 0.625 mmol) and NaClO₂ (35 mg, 0.384 mmol), then stirred for 25 min. The reaction was then quenched with a 1:1 mixture of sat. aq NaCl and sat. aq Na₂S₂O₃ (3 mL). The resulting mixture was diluted with H₂O (10 mL) and Et₂O (10 mL), and the layers were separated. The aqueous layer was extracted with Et₂O (3 × 10 mL) and the organic phases were combined, dried (MgSO₄), filtered, and concentrated. The resulting oil was purified by chromatography (silica gel, 30% EtOAc–PE) to give the carboxylic acid 9 as a viscous colorless oil; yield: 23 mg (75%, two steps); [α]_D ^24.2 +9.0 (c 0.95, CH₂Cl₂); R_f = 0.20 (20% EtOAc–PE) [stained with anisaldehyde (green), UV active].

IR (thin film): 3072, 2931, 1712, 1641, 1590, 1428, 1111, 1084, 1006, 702 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.68 (m, 4 H), 7.36–7.45 (m, 6 H), 5.77 (m, 1 H), 5.02 (s, 1 H), 4.98 (dd, J = 2, 7 Hz, 1 H), 4.25 (p, J = 5 Hz, 1 H), 3.78 (p, J = 6 Hz, 1 H), 3.24 (m, 1 H), 3.12 (m, 1 H), 2.55 (m, 2 H), 2.19 (m, 2 H), 1.50–1.74 (m, 5 H), 1.45 (m, 1 H), 1.18–1.32 (m, 3 H), 1.04 (s, 9 H), 1.00 (m, 1 H), 0.93 (t, J = 8 Hz, 9 H), 0.57 (q, J = 8 Hz, 6 H); the CO₂H proton was not observed.

¹³C NMR (125 MHz, CDCl₃): δ = 174.6, 135.9, 135.8, 134.7, 133.8, 133.3, 129.8, 129.7, 127.7, 127.6, 117.1, 74.7 (×2), 68.7, 68.5, 43.9, 43.3, 42.6, 42.0, 31.8, 31.1, 26.9, 23.4, 19.3, 6.9, 5.0.

HRMS (ES+): m/z calcd for C₃₆H₅₆NaO₅Si₂: 647.3558; found: 647.3569.

Cyclization Precursor 11

Carboxylic acid 9 (126 mg, 0.0202 mmol) was dissolved in toluene (6.00 mL) under N₂, and the soln was treated sequentially with Et₃N (163 μL, 1.17 mmol) and 2,4,6-trichlorobenzoyl chloride (60 μL, 0.383 mmol). The resulting soln was stirred for 3 h at r.t., by which time the formation of salts was observed. A soln of recognition domain 10 (80 mg, 0.133 mmol) in toluene (500 μL) was added through a cannula with the aid of 2 × 150 μL washes. A soln of DMAP (85 mg, 0.695 mmol) in toluene (200 μL) was then immediately added and the resulting mixture was stirred for 1 h then loaded directly onto a silica gel column and eluted with 20% EtOAc–pentane. Fractions containing the product were collected and concentrated in vacuo to afford a residue that was used directly in the next step.

The residue was dissolved in a soln of TsOH (10.6 mg, 0.042 mmol) in 20% H₂O–THF (25 mL) and the soln was stirred at r.t. under air for 21 h, then diluted with Et₂O (50 mL) and sat. aq NaHCO₃ (50 mL). The organic and aqueous layers were separated, and the aqueous layer was extracted with Et₂O (3 × 50 mL). The organic layers were combined, dried (MgSO₄), filtered, and concentrated, and the residue was purified by chromatography (silica gel, 20% EtOAc–PE) to afford 11 as a colorless oil; yield: 84.7 mg (58%, two steps); [α]_D ^24.2 −39.1 (c 1.24, CH₂Cl₂); R_f = 0.45 (15% EtOAc–PE) [stained with anisaldehyde (black spot), UV active].

IR (thin film): 3497, 3072, 3049, 2931, 2857, 2721, 1723, 1689, 1471, 1429, 1388, 1257, 1229, 1156, 1106, 837, 704 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 9.56 (d, J = 8 Hz, 1 H), 7.66 (m, 4 H), 7.34–7.43 (m, 7 H), 5.97 (s, 1 H), 5.93 (dd, J = 8, 9 Hz, 1 H), 5.78 (m, 1 H), 5.21 (m, 1 H), 5.09 (s, 1 H), 5.07 (d, J = 4 Hz, 1 H), 5.03 (s, 1 H), 4.32 (m, 1 H), 3.77 (t, J = 9 Hz, 1 H), 3.64 (s, 3 H), 3.52–3.63 (m, 5 H), 3.21 (t, J = 9 Hz, 1 H), 3.10 (m, 1 H), 2.54 (dd, J = 5, 15 Hz, 1 H), 2.44 (dd, J = 5, 15 Hz, 1 H), 2.20 (m, 1 H), 1.96–2.10 (m, 6 H), 1.88 (t, J = 11 Hz, 1 H), 1.54–1.75 (m, 6 H), 1.34–1.52 (m, 9 H), 1.17–1.28 (m, 14 H), 1.19 (s, 3 H), 1.14 (s, 3 H), 1.01 (s, 9 H), 0.89 (s, 3 H), 0.86 (t, J = 7 Hz, 1 H), 0.05 (s, 3 H), 0.04 (s, 3 H).

¹³C NMR (125 MHz, CDCl₃): δ = 194.8, 171.8, 171.5, 167.0, 166.5, 135.8, 135.1, 133.8, 133.7, 129.7 (×2), 127.6, 127.2, 120.1, 116.9, 99.4, 78.5, 74.3, 72.8, 71.0, 70.9, 67.6, 66.3, 64.8, 51.1, 45.7, 44.7, 43.0, 41.7, 37.9, 34.5, 31.7, 31.6, 31.2, 31.1, 28.9 (×2), 26.9, 25.8, 24.4, 23.1, 23.0, 22.5, 20.0, 19.4, 18.2, 14.1, −5.3 (×2).

HRMS (ES+): m/z calcd for C₆₁H₉₄O₁₃Si₂Na: 1113.6125; found: 1113.6128.

Bryostatin Analogue 12

Cyclization precursor 11 (7.6 mg, 0.0070 mmol) was dissolved in CH₂Br₂ (0.250 mL) under N₂ and the resulting soln was cooled in a CO₂/MeCN bath. After 10 min, 50 μL of a stock soln of FeBr₃ (1.1 mg, 0.00367 mmol) and TMSBr (4.8 μL, 0.0367 mmol) in CH₂Br₂ (0.20 mL) was then added dropwise and the resulting mixture was stirred for 3 h. The reaction was quenched with H₂O (0.50 mL) and the mixture was diluted with Et₂O (1.0 mL) and H₂O (1 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with Et₂O (3 × 1 mL). The organic layers were combined, dried (MgSO₄), filtered, and concentrated to give a yellow oil. This oil was purified by chromatography (silica gel, 25% EtOAc–PE) to give the cyclized, C26 TBS-deprotected product as a clear colorless viscous residue; yield: 4.1 mg (57% yield); R_f = 0.40 (25% EtOAc–PE).

The crude oily product was dissolved in THF (5.0 mL) under N₂ in a polypropylene vial and the resulting soln was cooled in an ice water bath. HF·pyridine (70% HF; 1.50 mL) was added dropwise over 2 min and the mixture was stirred for 10 min. The cold bath was then removed and the soln was stirred for an additional 24 h at r.t. The reaction was then quenched with sat. aq NaHCO₃ (2 mL) and the mixture was diluted with Et₂O (5 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with Et₂O (3 × 5 mL). The organic phases were combined, dried (MgSO₄), filtered, and concentrated. The crude residue was purified by chromatography (silica gel, 60% EtOAc–PE) to give 12; yield: 2.1 mg (37%, two steps). This product was further purified by C18 reverse-phase HPLC (gradient 60% MeCN–H₂O to 100% MeCN) to give an amorphous white solid; R_f = 0.30 (60% EtOAc–PE) [stained with anisaldehyde (red)].

IR (thin film): 3395, 2955, 2919, 2850, 2360, 1720, 1656, 1570, 1540, 1464, 1413, 1377, 1261, 1113, 1023 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 5.99 (s, 1 H), 5.74 (d, J = 16 Hz, 1 H), 5.32–5.37 (m, 2 H), 5.26 (dd, J = 8, 15 Hz, 1 H), 5.13 (s, 1 H), 5.13 (s, 1 H), 4.44 (d, J = 11 Hz, 1 H), 4.16–4.24 (m, 2 H), 4.06 (t, J = 11 Hz, 1 H), 3.97 (t, J = 9 Hz, 1 H), 3.85 (dd, J = 2, 11 Hz, 1 H), 3.64–3.71 (m, 4 H), 3.55 (t, J = 10 Hz, 1 H), 3.52 (t, J = 11 Hz, 1 H), 3.40 (t, J = 11 Hz, 1 H), 2.22–2.38 (m, 3 H), 1.70–2.13 (m, 3 H), 2.26–2.30 (m, 2 H), 2.00–2.12 (m, 5 H), 1.77–1.82 (m, 3 H), 1.40–1.65 (m, 7 H), 1.18–1.33 (m, 9 H), 1.09 (s, 3 H), 0.98 (s, 3 H), 0.86 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₃₉H₆₁BrNaO₁₂: 823.3239; found: 823.3231.

Bryostatin Analogue 13

The C11-TES protected cyclization precursor 11 (10.6 mg, 0.0089 mmol) was dissolved in THF (7.0 mL) under N₂ in a polypropylene vial and the soln was cooled to −78 °C. HF·pyridine (70% HF; 2.00 mL) was added dropwise over 1 min and the mixture was stirred at −78 °C for 3 h and then allowed to warm gradually to r.t. over 24 h without removal of the cooling bath. The reaction was quenched with sat. aq NaHCO₃ (10 mL) and the mixture was diluted with Et₂O (10 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with EtOAc (3 × 10 mL). The organic phases were combined, dried (MgSO₄), filtered, and concentrated. The crude residue was purified by chromatography (silica gel, 60% EtOAc–PE) to give a colorless oil; yield: 2.3 mg (35%); R_f = 0.40 (60% EtOAc–PE) [stained with anisaldehyde (red)].

IR (thin film): 3426, 2919, 2850, 1663, 1449, 1422, 1377, 1263, 1167, 1086, 1031, 704 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 5.99 (s, 1 H), 5.75 (d, J = 16 Hz, 1 H), 5.29–5.34 (m, 2 H), 5.17 (s, 1 H), 5.14 (s, 1 H), 4.72 (m, 1 H), 4.47 (d, J = 10 Hz, 1 H), 4.18 (m, 1 H), 4.07 (t, J = 12 Hz, 1 H), 3.95 (t, J = 12 Hz, 1 H), 3.84 (m, 1 H), 3.68 (s, 3 H), 3.64–3.71 (m, 4 H), 3.50–3.57 (m, 2 H), 3.43 (t, J = 12 Hz, 1 H), 2.44–2.51 (m, 2 H), 2.22–2.31 (m, 2 H), 1.94–2.05 (m, 5 H), 1.76–1.80 (m, 3 H), 1.39–1.60 (m, 8 H), 1.18–1.33 (m, 10 H), 1.12 (s, 3 H), 1.00 (s, 3 H), 0.87 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₃₉H₆₁FNaO₁₂: 763.4039; found: 763.4045.

Bryostatin Analogue 14

TMSOTf (1.0 μL) was dissolved in CH₂Cl₂ (0.100 mL) under N₂ in an oven-dried vial and the soln was cooled in a CO₂/MeCN bath for about 10 min. A soln of 11 (3.0 mg, 0.0028 mmol) in MeCN (0.100 mL) was then added dropwise over 5 min and the resulting mixture was stirred for 2 h. The reaction was quenched with H₂O (0.300 mL) and the mixture was allowed to warm to r.t. over 30 min then diluted with Et₂O (0.50 mL). The aqueous layer was extracted with Et₂O (3 × 0.50 mL) and the organic layers were combined, dried (MgSO₄), filtered, and concentrated. The resulting oil was purified by chromatography (silica gel, EtOAc to 5% MeOH–EtOAc) to give a colorless oil; yield: 1.6 mg (75%); R_f = 0.40 (5% MeOH–EtOAc) [stained with anisaldehyde (red)].

IR (thin film): 3442, 3356, 2922, 2852, 1721, 1658, 1463, 1435, 1377, 1259, 1167, 1086, 1031 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 5.99 (s, 1 H), 5.74 (d, J = 16 Hz, 1 H), 5.34 (m, 1 H), 5.25 (dd, J = 8, 16 Hz, 1 H), 5.17 (s, 1 H), 5.13 (s, 1 H), 4.47 (d, J = 12 Hz, 1 H), 4.04–4.18 (m, 3 H), 3.85 (m, 1 H), 3.67 (s, 3 H), 3.59–3.75 (m, 4 H), 3.52–3.63 (m, 5 H), 3.53 (t, J = 12 Hz, 1 H), 3.41 (t, J = 12 Hz, 1 H), 2.49–2.63 (m, 2 H), 2.29 (q, J = 7 Hz, 1 H), 2.18 (s, 1 H), 1.88–2.19 (m, 5 H), 1.76–1.80 (m, 3 H), 1.39–1.60 (m, 7 H), 1.18–1.33 (m, 12 H), 1.10 (s, 3 H), 0.97 (s, 3 H), 0.88 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₄₁H₆₅NNaO₁₃: 802.4348; found: 802.4354.

Bryostatin Analogue 15

TMSOTf (1.3 μL) was dissolved in CH₂Cl₂ (0.100 mL) under N₂ in an oven-dried vial and the soln was cooled in a CO₂/MeCN bath for about 10 min. A soln of 11 (4.0 mg, 0.0037 mmol) in PhCN (0.100 mL) was added dropwise over 5 min, and the resulting mixture was stirred for 2 h. The reaction was then quenched with H₂O (0.300 mL) and the mixture was allowed to warm to r.t. over 30 min, then diluted with Et₂O (0.50 mL). The aqueous layer was extracted with Et₂O (3 × 0.50 mL) and the organic layers were combined, dried (MgSO₄), filtered, and concentrated. The resulting oil was purified by chromatography (silica gel, EtOAc to 5% MeOH–EtOAc) to give a colorless oil; yield: 1.7 mg (55%); R_f = 0.50 (5% MeOH–EtOAc) [stained with anisaldehyde (red)].

IR (thin film): 3351, 2923, 2852, 1723, 1691, 1664, 1642, 1586, 1462, 1328, 1260, 1158, 1084 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.73 (t, J = 7 Hz, 2 H), 7.49 (t, J = 7 Hz, 1 H), 7.42 (t, J = 7 Hz, 2 H), 5.98 (s, 1 H), 5.83 (m, 1 H), 5.78 (d, J = 15 Hz, 1 H), 5.27–5.34 (m, 2 H), 5.19 (s, 1 H), 5.14 (s, 1 H), 4.50 (d, J = 11 Hz, 1 H), 4.33 (br s, 1 H), 4.15 (m, 1 H), 4.10 (m, 1 H), 3.86 (m, 1 H), 3.68 (s, 3 H), 3.64–3.74 (m, 3 H), 3.52–3.63 (m, 5 H), 3.55 (t, J = 12 Hz, 1 H), 3.45 (t, J = 12 Hz, 1 H), 2.52–2.60 (m, 2 H), 2.21–2.35 (m, 2 H), 1.91–2.10 (m, 5 H), 1.77–1.84 (m, 3 H), 1.40–1.65 (m, 7 H), 1.18–1.33 (m, 9 H), 1.11 (s, 3 H), 0.99 (s, 3 H), 0.87 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₄₆H₆₇NNaO₁₃: 864.4505; found: 864.4487.

Bryostatin Analogue 16

Cyclization precursor 11 (3.0 mg, 0.0028 mmol) and 4-methoxy-benzonitrile (9.2 mg, 0.069 mmol) were dissolved in CH₂Cl₂ (0.200 mL) under N₂, and the resulting soln was cooled in a CO₂/MeCN for about 10 min. TMSOTf (1.0 μL) was then added in one portion and the mixture was stirred for 4 h. The reaction was then quenched with H₂O (0.300 mL) and the mixture was allowed to warm to r.t. over 30 min then diluted with Et₂O (0.50 mL). The separated aqueous layer was extracted with Et₂O (3 × 0.50 mL) and the organic layers were combined, dried (MgSO₄), filtered, and concentrated. The resulting oil was purified by chromatography (silica gel, EtOAc) to give a colorless oil; yield: 0.3 mg (14%); R_f = 0.250 (EtOAc) [stained with anisaldehyde (red)].

IR (thin film): 3425, 2950, 2919, 2850, 1641, 1547, 1463, 1379, 1261, 1033 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.78 (d, J = 8 Hz, 2 H), 6.94 (d, J = 8 Hz, 1 H), 5.98 (s, 1 H), 5.83 (m, 1 H), 5.72–5.78 (m, 2 H), 5.31–5.38 (m, 2 H), 5.28 (d, J = 7, 15 Hz, 1 H), 5.18 (s, 1 H), 5.13 (s, 1 H), 4.51 (d, J = 13 Hz, 1 H), 4.27 (m, 1 H), 4.18 (m, 1 H), 4.12 (t, J = 10 Hz, 1 H), 4.07 (t, J = 8 Hz, 1 H), 3.86 (s, 3 H), 3.66 (s, 3 H), 3.61–3.78 (m, 3 H), 3.54 (t, J = 12 Hz, 1 H), 3.44 (t, J = 12 Hz, 1 H), 2.52–2.60 (m, 2 H), 2.17–2.36 (m, 3 H), 1.97–2.06 (m, 4 H), 1.77–1.84 (m, 3 H), 1.40–1.65 (m, 6 H), 1.18–1.33 (m, 14 H), 1.11 (s, 3 H), 0.98 (s, 3 H), 0.86 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₄₇H₆₉NO₁₄Na: 894.4610; found: 894.4613

Bryostatin Analogues 17 and 18

Cyclization precursor 12 (9.0 mg, 0.0083 mmol) was dissolved in PhOMe (0.225 mL) under N₂ and the resulting soln was cooled in a CO₂/MeCN bath for about 10 min. TMSOTf (1.8 μL, 0.0099 mmol) was then added in one portion, and the resulting mixture was stirred for 1.5 h. The reaction was then quenched with NaHCO₃ (0.50 mL) and the mixture was diluted with Et₂O (1.5 mL) and H₂O (1 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with Et₂O (3 × 2 mL). The organic phases were combined and concentrated to give a yellow oil that was purified via chromatography (silica gel, 25% EtOAc–PE) to give the cyclized C26-TBS-deprotected product as a clear colorless viscous residue; yield: 4.9 mg (56%); R_f = 0.35 (25% EtOAc–PE)

This product was dissolved in THF (7.0 mL) under N₂ in a polypropylene vial and the soln was cooled in an ice water bath. HF·pyridine (70% HF; 1.82 mL) was added dropwise over 1 min, and the mixture was stirred for 10 min. The cold bath was then removed and the soln was stirred for an additional 25 h at r.t. The reaction was quenched with sat. aq NaHCO₃ (2 mL) and the mixture was diluted with EtOAc (5 mL). The organic and aqueous layers were separated and the aqueous layer was extracted with EtOAc (3 × 5 mL). The organic phases were combined, dried (Na₂SO₄), filtered, and concentrated. The crude residue was purified by chromatography (silica gel, 60% EtOAc–PE) to give a 7:3 mixture of the C13-(2-methoxyphenyl) analogue 17 and the C13-(4-methoxyphenyl) analogue 18 as a white solid; yield: 1.9 mg (27%); R_f = 0.40 (60% EtOAc–PE) [stained with anisaldehyde (red)]. The pure analogues were obtained by C18 reverse-phase HPLC (60% MeCN–H₂O to 100% MeCN) as amorphous white solids.

17

IR (thin film): 3426, 2922, 2852, 1722, 1663, 1463, 1429, 1376, 1260, 1232, 1158, 1109, 704 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.13–7.18 (m, 2 H), 6.91 (dt, J = 2, 7 Hz, 1 H), 6.84 (dd, J = 7, 8 Hz, 1 H), 5.99 (s, 1 H), 5.73 (d, J = 15 Hz, 1 H), 5.32–5.37 (m, 2 H), 5.27 (s, 1 H), 5.14 (s, 1 H), 4.64 (d, J = 11 Hz, 1 H), 4.21 (m, 1 H), 4.10 (m, 1 H), 3.88 (m, 1 H), 3.82 (s, 3 H), 3.68 (s, 3 H), 3.66–3.74 (m, 3 H), 3.52–3.63 (m, 5 H), 3.54 (m, 1 H), 3.45 (t, J = 12 Hz, 1 H), 3.38 (t, J = 12 Hz, 1 H), 2.52–2.64 (m, 2 H), 2.26–2.30 (m, 2 H), 2.00–2.12 (m, 5 H), 1.77–1.82 (m, 3 H), 1.40–1.65 (m, 7 H), 1.18–1.33 (m, 9 H), 1.09 (s, 3 H), 0.98 (s, 3 H), 0.86 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₄₆H₆₈NaO₁₃: 851.4572; found: 851.4552.

18

IR (thin film): 3426, 2922, 2852, 1722, 1663, 1463, 1429, 1376, 1260, 1232, 1158, 1109, 704 cm⁻¹.

¹H NMR (500 MHz, CDCl₃): δ = 7.08 (d, J = 8 Hz, 1 H), 6.83 (dt, J = 7, 8 Hz, 2 H), 5.99 (s, 1 H), 5.74 (d, J = 15 Hz, 1 H), 5.32–5.37 (m, 2 H), 5.22 (s, 1 H), 5.14 (s, 1 H), 4.60 (d, J = 11 Hz, 1 H), 4.21 (m, 1 H), 4.10 (m, 2 H), 3.88 (m, 1 H), 3.78 (s, 3 H), 3.68 (s, 3 H), 3.66–3.74 (m, 3 H), 3.52–3.63 (m, 5 H), 3.54 (t, J = 12 Hz, 1 H), 3.45 (t, J = 12 Hz, 1 H), 3.38 (t, J = 12 Hz, 1 H), 2.52–2.64 (m, 2 H), 2.26–2.30 (m, 2 H), 2.00–2.12 (m, 5 H), 1.77–1.82 (m, 3 H), 1.40–1.65 (m, 7 H), 1.18–1.33 (m, 9 H), 1.09 (s, 3 H), 0.98 (s, 3 H), 0.86 (t, J = 7 Hz, 3 H).

HRMS (ES+): m/z calcd for C₄₆H₆₈NaO₁₃: 851.4572; found: 851.4575.