Building Chemical Probes Based on the Natural Products YM-254890 and FR900359: Advances toward Scalability

Abstract

The biological activity of natural products YM-254890 (YM) and FR900359 (FR) has led to significant interest in both their synthesis and the construction of more simplified analogs. While the simplified analogs lose much of the potency of the natural products, they are of interest in their own right, and their synthesis has revealed synthetic barriers to the family of molecules that need to be addressed if a scalable synthesis of YM and FR analogs is to be constructed. In the work described here, a synthetic route to simplified analogs of YM is examined and strategies for circumventing some of the challenges inherent to constructing the molecules are forwarded.

Graphical Abstract

Because of their selective inhibition of the Gq/11-signaling pathway,¹ the natural products YM-254890 (YM) and FR900359 (FR) have received significant interest from both a synthetic^2–8 and biological perspective (Figure 1).^9–15 Analogs of these natural products offer an opportunity to both study the biochemistry of the Gq-pathway and develop new chemical probes for examining related signaling pathways. For example, while simplified analogs of YM and FR like WU-07047 (Figure 1) do not possess the potency and selectivity of the natural products toward the Gq signaling pathway, WU-07047 has been shown to restore uterine artery tone to wild-type in mice with RGS2-deficiencies.¹⁶ RGS2-deficiencies were found to decrease uterine arterial blood flow (needed to support pregnancy) by increasing myogenic tone, at least partly as a result of prolonged G-protein activation. RGS2 aids the G-protein ATP-ase function that shortens the time the G-protein remains in its active state. WU-07047, as an inhibitor of Gq, helps compensate for the RGS2-deficiency. In another study, WU-07047 was compared to YM and FR for its ability to reduce blood pressure.¹⁷ All three compounds served as inhibitors to block G protein-dependent vasoconstriction, although the simplified analog did so to a much lesser extent, as expected. However, YM and WU-07047 also blocked whole-cell L-type Ca²⁺ transients and vasoconstricting of intact vessels, while FR did not. Interestingly, structural studies that examined how the inhibitor bound within the 5KMD dihydropyridine binding site of bacterial CavAB indicated that while all three inhibitors bound to the active site, WU-07047 bound in a manner that was distinct from both YM and FR with the sidechain phenyl substituent pointing in an opposite direction to the natural products. This result suggested a unique opportunity to develop inhibitors that take advantage of this alternative binding mode, a suggestion that led to proposals for WU-07047 analogs with modified sidechains.

Figure 1.

YM and FR and an initial pair of simplified analogs

Yet while the potential applicability of simplified YM and FR analogs as chemical probes has led to the synthesis of the natural products and a series of very interesting analogs, the synthetic routes developed to date have frequently failed to afford the quantities of material needed for more extensive biological studies. The syntheses published to date have relied on solid-phase peptide synthesis approaches and have typically afforded the natural products and analogs on scales ranging from 2 to 8 mg. These issues have led to an effort to access larger amounts of FR through a biotechnology approach that is capitalizing on cultivation of the gram-negative bacterium Chromobacterium vaccinia MWE205.¹⁸ The method uses novel promotors for the expression system that can lead to titer ranging from 22–84 mg/L of FR.¹⁹ Of course, a scalable synthetic route to the natural products and other analogs would provide a complimentary method to access the natural products that would allow for the evaluation of analogs containing unnatural amino acids and conformational constraints. However, a scalable, solution-phase approach to the analogs has proven elusive.

With this in mind, we reexamined the syntheses that enabled the construction of simplified analogs of YM and FR and what those efforts could teach us about the synthetic challenges inherent to developing efficient, scalable routes to this family of molecules.^20,21 There were significant synthetic challenges encountered during the initial synthesis of WU-07047 that both impeded those efforts and taught us about the challenges that would need to be overcome in the future. We report here the nature of those challenges and potential strategies for circumventing them. The first-generation synthesis of WU07047 is shown in Scheme 1. This route allowed for the synthesis of analogs on the 25 mg scale. The analogs retained the left- and right-hand portions of the natural products along with the appropriate size of the macrocycle and connected the conserved portions of the molecules with simplified bridges. This design kept the key portions of the natural products initially thought to make contact with Gq (marked in Figure 1 with the asterisks) while attempting to make a synthetically more accessible analog family.

Scheme 1.

An initial synthetic approach

The strategy taken to make the simplified analogs was to construct a set of building blocks for the molecules that could then be assembled into both the simplified and more complex analogs in a convergent manner. For these first-generation analogs (Scheme 1), the macrocycle was made using a ring-closing metathesis reaction. It was easy to imagine other approaches that would include a more functionalized ‘bottom bridge’ incorporated via more traditional ester and amide bond-forming reactions with the macro-cycle being generated through a macrolactamization as in other approaches.^2–8

While the approach allowed for the synthesis and testing of a number of analogs,^18,19 it was not without problems that both interfered with the scalability of the route and complicated its use for building analogs of WU-07047 that varied the β-hydroxyleucine sidechain found in YM. Two issues in particular required attention. The first involved a β-elimination of the β-hydroxy leucine group (marked in blue) that serves as the ‘left-hand’ sidechain of the molecule as shown in Scheme 1. One example of this elimination reaction is illustrated in Scheme 2 for the acyclic intermediate used to construct the left-hand portion of the molecule. While this elimination reaction was not a problem following formation of the macrocycle, it was a significant problem for every reaction in the synthesis following the addition of the sidechain and prior to generation of the macrocycle, leading to significantly lower yields over a large portion of the synthesis. In addition, trace amounts of acid or base in the solutions containing any of the purified acyclic intermediates led to the materials undergoing the elimination reaction during storage. This combination of low reaction yields and unstable intermediates meant that it was difficult to ‘stockpile’ most of the advanced intermediates in the synthesis or to vary the left-hand sidechain. The more valuable the unnatural amino acid derivative used to replace the β-hydroxyleucine group, the more painful it was to lose the sidechain all along the way during the synthetic sequence. It is difficult to imagine that this elimination has not also arisen in previous syntheses of the natural products and limited the scale of those synthesis. It appeared that a far better approach would be to add the left-hand sidechain needed after the macrocycle is formed.

Scheme 2.

A problematic elimination

The second problem encountered with our original synthetic efforts involved the acetyl amide groups found in both YM and the initial simplified analog shown in Scheme 1. The amides are not compatible with the esterification reactions needed to assemble the larger structure because they lead to oxazolone formation with a neighboring activated ester. To circumvent this problem, the nitrogen atoms were protected with a Troc-group that was subsequently converted into the desired acetyl amide functionality following the esterification reactions. Unfortunately, the exchange of the Troc protecting group for the desired amide proved capricious in a manner that again interfered with the use of this route for a scalable, building block-based synthesis of YM and FR analogs. This is especially true if one needs to remove a carbamate late in the synthesis when a loss of yield can be catastrophic.

We report here solutions to both these issues.

A Plan for Avoiding the Elimination Reaction

The suggestion to add the left-hand sidechain to the molecule following generation of the macrocycle sprung from the rather curious observation that the undesirable β-elimination reaction happened very readily prior to formation of the macrocycle but was not observed afterwards. A hint as to the reason for this observation can be gained by MM2-based molecular dynamics calculations. In the macrocycle, it appears that the molecule adopts a conformation that prevents the sidechain from being antiperiplanar to the proton labeled H_a in Figure 2. This conformation relieves steric congestion between the sidechain and the macrocycle. In the acyclic molecule, the steric interactions between the sidechain and the rest of the molecule can be relieved by bond rotations. Hence, the acyclic molecule is free to adopt a conformation having the sidechain antiperiplanar to H_a and the elimination reaction occurs readily. The cyclic molecule is not, and the result is that the elimination does not happen as readily with the macrocycle as it does with the more flexible, acyclic molecule.

Figure 2.

(a) Conformational structure of YM showing that the proton and oxygen marked with the asterisks are not antiperiplanar to each other. (b) Conformational structure of the simplified analog 5 again showing that the proton and oxygen marked with the asterisks are not antiperiplanar.

While the late-stage addition of the sidechain was considered early in our efforts, each of the previously published synthetic routes had added the sidechain prior to the cyclization. In addition, a previous synthesis of the macrocycle without the inclusion of the sidechain suggested to us that there might be some difficulty associated with the addition of the sidechain following closure of the macrocycle.² So, we followed suit and added the β-hydroxy leucine group to the left-hand building block prior to assembly of the macrocycle. With the elimination reaction proving so problematic, we decided to reverse this earlier decision. From the start, two approaches were considered (Scheme 3). The first (Path A) would convert the β-hydroxyleucine 2 into a β-propiolactone 3 that would then be opened by the macrocycle. This would introduce the sidechain to the molecule in a fashion directly analogous to the earlier addition of a sidechain to the core Taxol ring skeleton.²² Path A had two potential advantages. First, the approach would add the sidechain in the very last step of the synthesis and, in so doing, avoid any chance for an elimination in a subsequent reaction. Second, the use of the β-propiolactone approach would avoid any need to protect the hydroxyl group of β-hydroxyleucine (2) and hence the need to remove that protecting group.

Scheme 3.

Approaches to a late addition of the left-hand sidechain

The second route (Path B) would require a protection–deprotection sequence including one step that followed coupling of the sidechain to the macrocycle in order to take advantage of a series of more traditional, and hopefully more trustworthy, reactions. The sequence would require protection of the acid, protection of the hydroxyl group, deprotection of the acid, coupling to the macrocycle, and then deprotection of the hydroxyl group. Overall, the larger number of transformations, traditional or not, in Path B, led us to select Path A as the initial approach.

Synthesis of the Macrocycle

In either plan, the first step was to synthesize the macrocycle 1 needed for the coupling reaction. The chemistry to do so followed a very similar approach to the previous route, but the yields of the process were consistently better than those obtained previously because of the absence of the elimination-prone left-hand sidechain (Scheme 4). The synthesis started by converting the simplified top bridge (7) into a mixed anhydride and then quenching the activated ester with unprotected β-hydroxyleucine (8). The resulting carboxylic acid was converted into an allyl ester, thereby installing the double bond needed for a subsequent ring-closing metathesis reaction. This part of the sequence was identical to the route used in the previous synthesis.^18,19

Scheme 4.

Synthesis of the left-hand building block and simplified top bridge

The right-hand half of the molecule was also assembled in a manner directly analogous to the previously reported chemistry (Scheme 5). In this case, a Mitsunobu reaction was used to couple the N-ethylcarbamate protected allylglycine to methyl (S)-2-hydroxy-3-phenylpropionate (l-Pla methyl ester). The reaction inverted the stereochemistry of the alcohol, as expected, to afford the diastereomer with the relative stereochemistry the same as the natural products YM and FR. Variations in this stereochemistry were wanted for biological studies, so a right-hand building block with the opposite stereochemistry was also constructed. This was accomplished by using a standard esterification reaction to couple the same two starting materials. It was desired to make the two stereochemical analogs from a common intermediate that led to a change from the original use of an acetyl group to protect the amine of the allyglycine for the Mitsunobu reaction.^18,19 The acyl group was not compatible with the esterification reaction, leading to the well-known oxazolone product. This change was only partially successful. While the Boc protecting group used for the esterification reaction could be exchanged for the acetyl group following the esterification reaction, the yield of the process was only 39%.¹⁹

Scheme 5.

The right-hand piece

With that in mind, we hoped that an alternative protecting group strategy might be compatible with a single approach to both stereoisomers of the right-hand building block while retaining biological activity. For this reason, an ethyl carbamate was examined (13c and 13c′). The ethyl carbamate was compatible with both the Mitsunobu and esterification routes and the subsequent lithium iodide cleavage of the methyl ester in the molecule needed to reveal the acid for carrying the molecule forward (Scheme 6). Yet while the plan was ideal from a synthetic perspective, the ethyl carbamate did not prove compatible with the subsequent biological studies. So, we settled on a strategy that used the acetamide, the Mitsunobu reaction, and the construction of 13a for all new analogs. For a change in stereochemistry, the enantiomer of 11 can be employed if needed.

Scheme 6.

Assembling the macrocycle

At this point, the assembly of the macrocycle was undertaken (Scheme 6). While the reported yields of the reactions shown were not significantly different from those obtained in the initial route, the stability of the products was. We found that our previously reported route had to be conducted rapidly without storing any intermediate along the way due to the elimination reaction discussed earlier. A longer chromatography would lead to diminished yield. The result is that one could not stage the synthesis of analogs of the initial target structure from an advanced intermediate. Each analog had to be made from ‘scratch’ with the same rapid approach to the entire synthesis. The chemistry shown in Scheme 6 held no possibility for an equivalent elimination reaction. Hence, any intermediate including the macrocycle could be synthesized and stored for future use, an approach that greatly simplified the synthesis of analogs. Accordingly, macrocycles incorporating both stereoisomers of the right-hand building block were synthesized.

The β-Propiolactone Approach (Path A)

With the macrocycle in hand, attention turned toward exploring method A for introducing the isoleucine sidechain. This route took advantage of a four-membered lactone as the electrophile needed for an esterification reaction (Scheme 3). These efforts began with the synthesis of the four-membered ring lactone (Scheme 7). To this end, the amine in amino β-hydroxyleucine was protected as a methylcarbamate and then the lactone was generated using TBTU-coupling conditions. Both reactions proceeded smoothly, and the lactone 3 could be isolated by column chromatography. The approach to making the lactone was general and proved compatible with the use of a variety of protecting groups and amino acids.

Scheme 7.

β-Lactone synthesis

While production of the lactone proceeded smoothly, the coupling of that lactone to the macrocycle did not (Scheme 8). For the coupling reaction, the alcohol of the macrocycle was deprotonated with LiHMDS and then the alkoxide was treated with the β-lactone. The use of the strongly basic conditions proved necessary for obtaining the desired product, and the reaction proceeded as planned in that no product from the undesirable elimination reaction was observed. Similar reactions on acyclic model studies were consistently plagued by the elimination reaction following coupling of the lactone to the alcohol. While there was no evidence of the elimination product, the reaction afforded 15% yield of the desired product along with 30% of the recovered starting material. When the product from the reaction was re-exposed to the reaction conditions it proved to be stable, with no evidence of elimination. Even with the poor yield, the trial reaction did lead to 20 mg of the desired product, a scale that was competitive with previous syntheses of the analog, using a route that was potentially much more scalable. However, the loss of material was clearly far from ideal.

Scheme 8.

β-Lactone coupling with the macrocycle

Analysis of the reaction and products generated showed that the low yield and poor mass balance of the transformation arose from a polymerization reaction that was triggered by the desired coupling reaction (Scheme 9). The alkoxide generated from the initial lactone opening proved to be a very good nucleophile for opening the next lactone. This opening competed with opening of the lactone by the macrocyclic starting material. A series of smaller amino acid test substrates were used in place of the macrocycle in an effort to avoid this polymerization reaction and test the idea that the issue resulted from the steric size of the macrocycle. None of these test reactions were successful. In each case, the alkoxide generated from the lactone opening reaction proved to be as good or better a nucleophile for the lactone opening than the original alkoxide. The polymerization reaction could not be stopped.

Scheme 9.

Polymer formation

The Traditional Esterification Approach (Path B)

The biggest problem with the polymerization reaction outlined above was that it lowered the mass balance of the reaction at the expense of consuming a macrocycle that was the product of a multistep synthesis. This cost was viewed as too great and made the slightly longer second approach to the late-stage addition of the sidechain more attractive (Scheme 3, Path B).

The route began with the conversion of an N-methyl carbamate protected β-hydroxy leucine into a suitably protected acid substrate for the coupling reaction (Scheme 10). The methyl carbamate protecting group was again selected because of its compatibility with subsequent biological studies (see below). Attempts to selectively protect the alcohol in the presence of the acid were not high yielding, so the sequence began by protecting the carboxylic acid as its benzyl ester. The alcohol was subsequently silylated and then the benzyl ester was cleaved with a hydrogenolysis to generate the desired acid for the coupling reaction.

Scheme 10.

Preparation of the acid coupling partner

The acid was then coupled to the macrocyclic ring using standard EDC and catalytic DMAP conditions to afford the product containing the sidechain in 63% isolated yield for product 6a and 84% yield for 6c (Scheme 11). The silyl protecting group was then removed from the sidechain with the use of TBAF to afford 64% yield of alcohol 5a and 76% yield of 5c.

Scheme 11.

Addition of the sidechain to the macrocycle

While it is not clear that the TBS protecting group is completely compatible with the coupling reaction, the overall method is clearly superior to any other approach taken to date. It can be consistently repeated, and it completely avoids the elimination reaction that had proven so problematic with earlier routes (Scheme 2). It is the route currently being employed to build analogs of the initial structure. For example, Scheme 12 highlights the synthesis of three analogs that were synthesized to examine the role of a donor group on the phenyl substituent of WU-07047. In this case, the macrocycles were synthesized using the exact same route described above, and then the sidechain was added in the final stages of the synthesis. This avoided having to carry the sidechain through all three syntheses with the associated loss of yield for the elimination reactions every step along the way. It also allowed for the stockpiling of building blocks other than the right-hand piece containing the phenyl ring at advanced stages of the synthesis so that the analogs could be rapidly assembled. In all three cases, the coupling reaction and the subsequent deprotection step proceeded in high yield. Again, once that macrocycle was formed, elimination of the sidechain was not an issue.

Scheme 12.

Alternative analogs synthesized

The Compatibility of Carbamate Protecting Groups with Biological Activity

The advantages of adding the sidechain onto the macrocycle using a standard esterification reaction did expose the earlier noted problems associated with converting carbamate protecting groups into the N-acetyl protecting groups contained in YM and FR (with FR containing one N-propionyl protected group). The amide groups needed in the natural product are not compatible with the formation of a neighboring activated ester, giving rise to oxazolone products. For this reason, the amide groups needed to be introduced following the coupling reactions used to assemble the structure. Moving the coupling reaction to a very late stage following formation of the macrocycle meant that the entire macrocycle had to be carried through an exchange of a carbamate protecting group for the necessary amide functionality. A number of such transformations have been used in connection with our efforts to build YM and FR analogs. The most successful has been the conversion of a Troc-protecting group into the acetyl group in one step (Scheme 13). The reaction is shown for the ‘best-case’ in combination with the removal of a TBS group, exactly the transformation needed for the late-stage addition of the sidechain described above. While the reaction was capable of leading to a decent yield of the conversion, we could not consistently obtain that yield. In practice, the conversion led to yields that ranged from 24 to 51%, a situation that was unacceptable at a late stage in the synthesis.

Scheme 13.

An example of a carbamate to amide exchange

Since the FR variant of the natural product family had a propionyl group in place of the acetyl group on the left-hand side of the molecule (Figure 1),^4,8 we wondered whether the analogs might retain activity if we left the methyl carbamate in molecule 5a and avoid the exchange altogether. The thought was that the methyl carbamate would occupy a similar volume as the propionyl-amide. With this in mind, we moved forward with the biological evaluation of 5a (Scheme 11). This evaluation was conducted by comparing the methyl carbamate analog 5a (compound 5a is represented by its official designation WU-06047 in Figure 3) with two previous analogs (WU-07047 and WU-09060 in Scheme 1) and the natural product FR900359 (FR) using an agonist-induced Ca²⁺ flux assay.²³ In this assay HEK2923 cells were transfected with a Twitch 2B Ca²⁺ Fret reporter. The cells were then treated with FR, one of the three analogs, or a vehicle, for three hours, and then stimulated with the Gq-coupled GPCR agonist carbachol. The recorded changes in Twitch 2B fluorescence and FRET are reported in Figure 3. What is clear from the data is that analog 5a (WU-06047) is superior to the previous simplified analogs studied, although it falls far short of the natural product in terms of both potency and efficacy.

Figure 3.

Biological comparison of the simplified analogs with the natural product FR900359 (FR). WU-06047 is analog 5a. WU-07047 and WU-09060 are shown in Scheme 1.

With the presence of the carbamate protecting group on the β-hydroxy leucine sidechain both simplifying the synthesis and increasing the activity of analog 5a (WU-06047) relative to the equivalent analog with the acetamide group in this position (WU-07047), we remain hopeful that its use will allow for a more scalable synthesis of YM and FR analogs in the future. If equally tolerated in more potent analogs, the use of a methyl carbamate on the β-hydroxy leucine sidechain would avoid any need to exchange protecting groups at a late stage of the synthesis.

With that result in hand, we turned our attention to the acetamide group on the right-hand side of the molecule. Earlier in connection with Scheme 5, it was pointed out that the use of an ethylcarbamate in this position (Scheme 11, c) was ideal for the synthesis of new analogs having either stereochemistry on the right-hand side of the molecule because it is compatible with both Mitsunobu and peptide coupling methods of esterifying the N-protected allylglycine (10) onto phenyllactate methyl ester (11). Unfortunately, the ethyl carbamate is not an easy group to remove and replace with an acetamide. Hence, the biological relevance of analog 5c, bearing the ethylcarbamate group, was also examined. In this case, the incorporation of a second carbamate protecting group into the molecule led to solubility issues that prevented successful completion of biological studies evaluating this analog. This forced us to conclude that the right-hand group would need to remain an acetamide in the simplified analogs. While it is possible that a molecule with only a carbamate on the right side and an acetamide on the β-hydroxy leucine sidechain might be soluble, it is far better to keep the carbamate protecting group on that sidechain. After all, the presence of the acetamide group on the right-hand building block is compatible with the Mitsunobu strategy that was used to make analog 5a, an analog that retains the stereochemistry of the natural product. If one does require an analog with the opposite stereochemical outcome from the Mitsunobu strategy, then the enantiomer of compound 11 can be used as a substrate for that reaction. This approach is far superior to having to exchange a protecting group on the β-hydroxy leucine side chain in the final stages of the synthesis.

Conclusions

While a number of total syntheses of YM, FR, and analogs have been completed, there is still a need for scalable routes to analogs that will allow for more extensive biological testing. With the knowledge that the β-hydroxy leucine sidechain can be added after formation of the macrocycle and that the analogs can tolerate a carbamate protecting group on that sidechain, it is now possible to add the sensitive β-hydroxy leucine sidechain to analogs in the penultimate step of the synthesis (followed only by the removal of a TBS protecting group). This allows for higher yields throughout a building block approach to the construction of the analogs and, in so doing, improves the scalability of the process. The building blocks used to assemble the analogs can all be made on multigram scale. Currently, the synthesis of analog 5a uses a synthetic approach that has a longest linear sequence of eight steps and leads to the product in an overall 11% yield. Work to scale the reactions further is underway in connection with analogs that possess greater levels of functionality as part of an effort to determine the minimal structural unit necessary for obtaining both high levels of selectivity and increased levels of potency and efficacy in the analogs.

All glassware was flame-dried prior to use and all reactions were conducted under argon atmosphere unless otherwise noted. All reactants were purchased from commercial suppliers and used without further purification except where otherwise noted. Solvents were bought in anhydrous form and some were also distilled before use. Tetrahydrofuran was distilled over sodium and benzophenone. Dichloromethane was distilled from calcium hydride.

Flash chromatography was carried out by using silica gel (230–400 mesh) purchased from Sorbent Technologies, Inc. All ¹H and ¹³C NMR spectra were recorded with an Agilent DD2 500 MHz, or a Varian Mercury Plus 300 MHz in a deuterated chloroform (CDCl₃) solvent with tetramethylsilane (TMS) as an internal standard unless otherwise noted. Infrared (IR) spectra were recorded with a Bruker Optics Alpha FT-IR instrument. High-resolution mass spectra (HRMS) were obtained using electrospray ionization (ESI) with Q-TOF (quadrupole time of flight) detection.

Methyl Ester Deprotection

To a flame-dried round-bottom flask with stir bar was added lithium iodide (7.07 mmol) in THF (16 mL). The flask contents was then brought to reflux before addition of methyl ester (1.86 mmol) in THF (15 mL). The solution was heated to reflux for 24–48 h, then cooled to room temperature before concentrating. The resulting oil was dissolved in chloroform (50 mL) and was then extracted with sat. NaHCO₃ (3 × 25 mL). The combined aqueous layers were then brought to pH 2, and extracted with chloroform (3 × 30 mL). The organic layers were dried with MgSO₄ and concentrated in vacuo.

Mitsunobu Reaction

Carboxylic acid (1.80 mmol) was subjected to a benzene (30 mL) azeotropic distillation to remove trace water. To this flask was added triphenylphosphine (1.80 mmol), followed by THF (4 mL). The contents of the flask was stirred at –30 °C before addition of alcohol (1.64 mmol) dissolved in additional THF (4 mL), followed by dropwise addition of a solution of diethyl azodicarboxylate (40 wt% in toluene, 0.84 mL, 1.84 mmol). The temperature was maintained at –30 °C for 30 minutes and was then allowed to reach room temperature. After 18 hours, the reaction was concentrated in vacuo. The resulting oil was dissolved in EtOAc (50 mL) and was washed with sat. NaHCO₃ (3 × 25 mL). The Aqueous layer was then back extracted with EtOAc (2 ×20 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuo. The resulting oil was purified via flash column chromatography.

Amide Coupling (HATU)

To a flame-dried RBF was added Boc-protected amine (0.057 g, 0.14 mmol), which was stirred at r.t. in dichloromethane (3 mL) with trifluoracetic acid (1 mL). After three hours, the reaction was diluted with diethyl ether (20 mL) and concentrated in vacuo. This dilution and concentration were repeated three times, followed by azeotropic distillation with benzene (25 mL) to remove trace water. To this flask was then added carboxylic acid (0.0520 g, 0.17 mmol) with HATU (0.0798 g, 0.21 mmol). The contents of the flask were then dissolved in DMF (1 mL), followed by addition of DIPEA (0.06 mL, 0.34 mmol). The reaction was then stirred at r.t. overnight. After 20 hours, the mixture was diluted with EtOAc (30 mL) and washed with saturated NaHCO₃ (3 × 25 mL). The aqueous layer was then back extracted with EtOAc (2 × 20 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuo. The resulting oil was purified via flash column chromatography.

Ring Closing Metathesis

To a flame-dried RBF was added open macrocycle (0.25 mmol) in DCM (160 mL). The reaction mixture was brought to reflux before addition of Grubbs II catalyst (0.05 mmol) in DCM (15 mL). After 20 hours at reflux, the reaction was allowed to cool to r.t. before concentration in vacuo. Crude product was then purified via flash column chromatography.

Macrocycle Esterification

To a flame-dried RBF was added macrocycle (0.092 g, 0.15 mmol) with DMAP (0.028 g, 0.23 mmol) and EDC (0.027 g, 0.14 mmol). Flask was cooled to 0 °C before addition of compound 21 (0.031 g, 0.10 mmol) in DCM (1.3 mL). Reaction was then stirred and allowed to reach RT overnight. After 48 h, the mixture was diluted with DCM (30 mL) and washed with sat. NaHCO₃ (3 × 25 mL). The aqueous layer was then back extracted with DCM (2 × 20 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuo. The resulting oil was purified via flash column chromatography.

Compound 8

A stirred solution of 0.499 g (1.93 mmol) 8-(Boc-amino)octanoic acid (7) and 0.42 mL (3.86 mmol) of 4-methylmorpholine in 5 mL THF was brought to −10 °C. To the stirring solution 0.25 mL isobutyl chloroformate was added and maintained at −10 °C for 30 minutes. The reaction was allowed to warm to 0 °C at which time 0.545 g (2.97 mmol) of the hydrochloride salt of β-hydroxyluecine (6) in 5.4 mL of 1 M NaOH. The reaction was warmed to room temperature overnight and stirred for 24 hours. The reaction was then diluted with 30 mL H₂O and washed with EtOAc (2 × 25 mL). The combinded organic layers were then extracted with saturated NaHCO₃ (3 × 25 mL). All aqueous layers were combined and acidified to pH 2 with 1 M HCl and extracted with EtOAc (3 × 30 mL). The combined organic layers were then dried with MgSO₄and concentrated in vacuo to give 0.696 g (93% yield) of compound 8 as a tan foaming oil that was carried forward without further purification.

FTIR (neat):: 3374, 2932, 2860, 1702, 1655, 1528, 1366, 1275, 1252, 1171 cm⁻¹.

¹H NMR (500 MHz, CD₃OD):δ = 4.69 (d, J = 2.5 Hz, 1 H), 3.72 (dd, J = 9.1, 2.4 Hz, 1 H), 3.03 (t, J = 7.0 Hz, 3 H)), 2.35–2.28 (m, 2 H)), 2.06 (s, 3 H)), 1.72–1.61 (m, 3 H)), 1.44 (s, 11 H), 1.36 (q, J = 6.7, 5.7 Hz, 6 H), 1.03 (d, J = 6.6 Hz, 3 H)), 0.91 (d, J = 6.7 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CD₃OD):δ = 177.70, 175.84, 159.79, 81.03, 79.36, 57.21, 42.63, 38.20, 36.20, 33.76, 32.17, 31.43, 31.35, 30.12, 29.02, 28.11, 27.28, 20.92, 20.70.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₉H₃₆O₆N₂ [M + Na]⁺ 411.2466; found 411.2468.

Compound 9

To a stirring solution of 0.662 g (1.71 mmol) compound 8 with 0.661 g NaHCO₃ (7.87 mmol) in 10 mL DMF was added 1.3 mL (15.8 mmol) allyl bromide in a dropwise fashion. The reaction was allowed to stir at room temperature overnight. After 20 h, the reaction was diluted with 30 mL H₂O, and extracted with EtOAc (3 × 30 mL). The combined organic layers were then washed with brine (1 × 20 mL). The organic layer was then dried with MgSO₄and concentrated in vacuoin vacuo. The crude product was then purified by flash chromatography (silica gel, 50% hexane: 50% EtOAc) to give 0.438 g (60% yield) of compound 9 as a clear oil.

FTIR (neat): 3346, 2931, 2858, 1743, 1692, 1660, 1530, 1366, 1252, 1172 cm⁻¹.

¹H NMR (300 MHz, CDCl₃) δ = 6.75 (d, J= 9.4 Hz, 1 H), 5.78 (ddt, J = 17.2, 10.4, 5.7 Hz, 1 H), 5.20 (dd, J = 17.0, 1.2 Hz, 1 H), 5.11 (dd, J = 10.6, 1.2 Hz, 1 H), 4.69 (dd, J = 9.1, 2.1 Hz, 1 H), 4.51 (d, J= 5.3 Hz, 2 H)), 3.61 (dd, J = 9.4, 1.8 Hz, 1 H), 2.96 (dd, J= 5.9 Hz, 2 H)), 2.14 (t, J = 7.3 Hz, 2 H)), 1.42–1.70 (m, 4 H)), 1.31 (s, 9 H), 1.30 (br, 2 H)), 1.20 (br, 6 H), 0.89 (d, J= 7.0 Hz, 3 H)), 0.79 (d, J = 7.0 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 173.6, 171.1, 156.0, 131.5, 118.1, 78.7, 76.9, 65.6, 54.4, 40.2, 36.0, 31.0, 29.6, 28.7, 28.6, 28.1, 26.3, 25.2, 18.9, 18.7.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C_{22 H)40}O₆N₂ [M + Na]⁺ 451.2779; found 451.2771.

Compound 10a

A stirring solution of 0.499 g (4.33 mmol) L-allylglycine in 15 mL H₂O was brought to 0 °C. The pH was adjusted to pH 10 with 1 M NaOH. 0.46 mL (4.74 mmol) freshly distilled acetic anhydride was then added dropwise to the stirring solution. The pH was re-adjusted to 10 and was allowed to stir and reach room temperature overnight. After 18 hours, the reaction was brought to pH 2 with 1 M HCl. The reaction was then extracted with EtOAc (3 × 30 mL). The organic layer was then dried with MgSO₄ and concentrated in vacuoto afford 0.500 g (73% yield) of compound 10a as a white solid. This crude product was carried forward without any further purification.

FTIR (neat): 3317, 3083, 3984, 2934, 2540, 1725, 1644, 1550, 1439, 1377, 1203, 1144 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 11.50 (s, 1 H), 6.32 (d, J = 7.6 Hz, 1 H), 5.66 (ddt, J = 16.5, 10.4, 7.2 Hz, 1 H), 5.13–5.05 (m, 2 H)), 4.64–4.56 (m, 1 H), 2.62–2.53 (m, 1 H), 2.49 (dt, J = 14.1, 6.8 Hz, 1 H).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 173.40, 170.27, 131.02, 118.45, 50.85, 35.03, 21.84.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₇H₁₁O₃N₁ [M + Na]⁺ 180.0631, 180.0627 observed.

Compound 12a

The reaction was set up according to the general procedure for the Mistunobu esterification. The crude product was then purified by flash chromatography (silica gel, 100% Et₂O) to give 0.739 g (77% yield) of compound 12a as a white solid.

FTIR (neat): 3282, 3064, 3031, 3006, 2954, 1748, 1657, 1542, 1438, 1374, 1285, 1219, 1192, 1147 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 7.36–7.19 (m, 5 H), 5.88 (d, J = 7.9 Hz, 1 H), 5.40 (ddt, J = 17.2, 10.2, 7.2 Hz, 1 H), 5.25 (dd, J = 9.5, 4.0 Hz, 1 H), 5.00–4.95 (dd, 1 H), 4.84–4.77 (m, 2 H)), 3.76 (s, 3 H)), 3.25 (dd, J = 14.3, 4.0 Hz, 1 H), 3.15–3.08 (dd, 1 H), 2.48 (m, 1 H), 2.35 (m, 1 H), 1.98 (s, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 173.72, 172.07, 172.03, 138.25, 134.40, 131.89, 131.89, 131.87, 131.26, 129.89, 121.90, 76.40, 55.19, 54.16, 39.93, 38.88, 25.76.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₇H₂₁O₅N₁ [M + Na]⁺ 342.1312, 342.1316 observed.

Compound 13a

The reaction was set up according to general procedure for the deprotection of a methyl ester and the workup conducted after the reaction was refluxed for 24 h. The reaction afforded 0.239 g (69% yield) of compound 13a as a yellow oil. Recrystallization from ethyl acetate, followed by slow evaporation in methanol afforded the pure diastereomeris product as clear crystals.

FTIR (neat): 3337, 3054, 2986, 1740, 1704, 1601, 1562, 1421, 1265 cm⁻¹.

¹H NMR (500 MHz, CD₃OD):δ = 7.34–7.19 (m, 5 H), 5.61 (ddt, J = 14.1, 10.3, 7.1 Hz, 1 H), 5.19 (dd, J = 9.4, 3.9, 1.9 Hz, 1 H), 5.02 (dd, 1 H) 4.98–−4.91 (dd, 2 H)), 4.54 (td, J = 7.5, 5.3, 2.1 Hz, 1 H), 3.25 (dd, J = 14.4, 3.8 Hz, 1 H), 3.09 (dd, J = 14.4, 9.3, 1.9 Hz, 1 H), 2.44–2.31 (m, 2 H)), 1.93 (s, J = 2.0 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CD₃OD):δ = 174.29, 173.72, 173.47, 138.93, 135.31, 135.26, 131.72, 130.77, 129.37, 129.26, 120.06, 76.22, 54.70, 39.50, 38.05, 23.58.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₆H₁₉O₅N₁ [M + Na]⁺ 328.115; found 328.1157.

Compound 10c

To a flame dried round bottom flask were added 0.501 g (4.36 mmol) of L-allylglycine, and 0.915 g (10.9 mmol) of NaHCO₃. The flask was brought to 0 °C before the addition of 17.5 mL of THF, followed by 11.3 mL of H₂O. In two separate portions, 2.1 mL (21.8 mmol) of ethyl chloroformate was added over thirty minutes. The flask was allowed to stir and reach room temperature overnight. After 20 hours, the reaction was diluted with H₂O (30 mL) and then washed with EtOAc (3 × 30 mL). The organic layer was then back extracted with saturated NaHCO₃(2 × 25 mL). The combined aqueous layers were then acidified to pH = 2 with 1 M HCl and extracted with EtOAc (3 × 30 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuoto afford 0.731 g (90% yield) of compound 10c as a clear oil. This crude product was carried forward without any further purification.

FTIR (neat): 3324, 2983, 1694, 1519, 1419, 1381, 1339, 1220, 1096, 1054 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 10.73 (s, 1 H), 5.74 (dq, J = 16.9, 7.9 Hz, 1 H), 5.38 (d, J = 8.2 Hz, 1 H), 5.16 (dd, J = 13.7, 8.6 Hz, 2 H)), 4.46 (q, J = 6.7 Hz, 1 H), 4.15 (dq, J = 21.6, 7.3 Hz, 2 H)), 2.58 (ddt, J = 40.4, 15.1, 6.9 Hz, 2 H)), 1.25 (t, J = 7.2 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 176.12, 156.42, 132.04, 119.47, 61.48, 53.08, 36.42, 14.47.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₈H₁₃O₄N₁ [M + Na]⁺210.0737; found 210.0751.

Compound 12c

The reaction was set up according to the general procedure for the Mitsunobu esterification. The crude product was then purified by flash chromatography (silica gel, 80% hexane: 20% EtOAc) to give 0.484 g (85% yield) of compound 12c as a white solid.

FTIR (neat): 3325, 2982, 1743, 1720, 1524, 1439, 1376, 2338, 1196, 1060 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 7.30 (t, J = 7.4 Hz, 2 H)), 7.27–7.19 (m, 3 H)), 5.44 (tt, J = 17.1, 7.8 Hz, 1 H), 5.25 (dd, J = 9.5, 4.0 Hz, 1 H), 5.11 (d, J = 8.4 Hz, 1 H), 4.98 (d, J = 10.2 Hz, 1 H), 4.87 (d, J = 17.0 Hz, 1 H), 4.51 (q, J = 6.4 Hz, 1 H), 4.10 (q, J = 7.2 Hz, 2 H)), 3.74 (s, 3 H)), 3.23 (dd, J = 14.3, 4.0 Hz, 1 H), 3.09 (dd, J = 14.3, 9.4 Hz, 1 H), 2.45 (dt, J = 13.5, 6.3 Hz, 1 H), 2.35 (dt, J = 13.8, 6.6 Hz, 1 H), 1.24 (t, J = 7.2 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 171.09, 169.45, 155.73, 135.63, 131.71, 129.24, 128.58, 127.19, 119.33, 73.69, 61.14, 53.02, 52.49, 37.32, 36.53, 14.5.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₈H₂₃O₆N₁ [M + Na]⁺372.1418; found 372.1413.

Compound 13c

The reaction was set up according to the general procedure for a methyl ester deprotection. The reaction was refluxed for 48 h, and it led to the formation of 0.352 g (82% yield) of compound 13c as a clear oil that was then carried forward without any further purification.

FTIR (neat): 3324, 2982, 1717, 1519, 1437, 1377, 1339, 1187, 1061 cm⁻¹.

¹H NMR (500 MHz, CD₃OD):δ = 7.33–7.21 (m, 6 H), 5.62 (ddt, J = 17.2, 10.3, 7.0 Hz, 1 H), 5.19 (dd, J = 9.3, 3.8 Hz, 1 H), 5.02–4.92 (m, 2 H)), 4.29 (dd, J = 8.2, 5.1 Hz, 1 H), 4.06 (qd, J = 7.1, 2.6 Hz, 2 H)), 3.25 (dd, J = 14.3, 3.9 Hz, 1 H), 3.10 (dd, J = 14.3, 9.3 Hz, 1 H), 2.46–2.41 (m, 1 H), 2.30 (dt, J = 14.7, 7.7 Hz, 1 H), 1.25 (dt, J = 14.1, 7.1 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CD₃OD):δ = 173.88, 173.70, 138.93, 135.42, 131.72, 130.73, 129.21, 120.01, 76.22, 63.31, 56.31, 50.53, 39.52, 38.27, 16.15.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₇H₂₁O₆N₁ [M + Na]⁺ 358.1261; found 358.1276.

Compound 12c’

To a flame dried RBF were added 0.803 g (2.50 mmol) of TBTU and 0.378 g (2.10 mmol) of methyl L-3-phenyllactate. To this flask was transferred 0.430 g (2.30 mmol) of compound 10c in 15 mL dichloromethane. The stirring solution was brought to 0 °C before the drop-wise addition of 0.91 mL (6.50 mmol) of triethylamine. The reaction was allowed to stir and reach RT overnight. After 20 h, the reaction was diluted with DCM (50 mL), transferred to a separatory funnel, and the mixture washed with saturated NaHCO₃(3 × 25 mL). The organic layer was then dried with MgSO₄ and concentrated in vacuo. The resulting oil was then then purified by flash chromatography (silica gel, 70% hexane: 30% EtOAc) to give 0.650 g (89% yield) of compound 12c’ as a clear oil.

FTIR (neat): 3324, 2982, 1721, 1523, 1439, 1379, 1341, 1191, 1064 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 7.32–7.18 (m, 5 H), 5.71 (ddt, J = 17.3, 10.6, 7.4 Hz, 1 H), 5.28 (dd, J = 8.1, 4.6 Hz, 1 H), 5.21 (d, J = 8.5 Hz, 1 H), 5.16–5.06 (m, 2 H)), 4.44 (m, 1 H), 4.09 (q, J = 7.1 Hz, 2 H)), 3.68 (s, 3 H)), 3.12 (d, 1 H), 2.97 (d, J = 6.5 Hz, 1 H), 2.55 (dd, J = 7.2, 6.6 Hz, 2 H)), 1.21 (t, J = 7.1 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 177.18, 173.94, 172.10, 158.60, 139.12, 138.08, 134.79, 131.10, 122.07, 80.83, 79.18, 76.13, 74.00, 63.79, 55.41, 43.22, 39.87, 39.13, 17.16.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₈H₂₃O₆N₁ [M + Na]⁺ 372.1418; found 372.1412.

Compound 13c’

The reaction was set up according to general procedure for the deprotection of a methyl ester. It afforded 0.245 g (80% yield) compound 13c’ as a clear oil was carried forward without any further purification.

FTIR (neat): 3362, 2947, 1836, 2506, 2074, 1700, 1439, 1379, 1275, 1188, 1118, 1023 cm⁻¹.

¹H NMR (500 MHz CD₃OD):δ = 7.34–7.18 (m, 6 H), 5.77 (dddd, J = 18.5, 10.3, 7.7, 6.4 Hz, 1 H), 5.23 (ddd, J = 8.1, 4.3, 1.4 Hz, 1 H), 5.14–5.02 (m, 2 H)), 4.90 (s, 3 H)), 4.25 (dd, J = 8.7, 5.0 Hz, 1 H), 4.11–4.03 (m, 2 H)), 3.32 (dq, J = 3.2, 1.6 Hz, 1 H), 3.23 (dd, J = 14.4, 4.4 Hz, 1 H), 3.18–3.08 (m, 1 H), 2.58 (ddd, J = 12.1, 9.2, 5.4 Hz, 1 H), 2.47–2.36 (m, 1 H), 1.24 (tdd, J = 7.0, 5.4, 1.3 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CD₃OD):δ = 174.04, 173.52, 159.87, 138.75, 135.78, 131.82, 131.78, 131.73, 130.68, 130.47, 129.21, 128.75, 119.98, 76.06, 74.04, 63.34, 56.80, 56.10, 42.86, 39.43, 38.11, 16.16, 15.73.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₇H₂₁O₆N₁ [M + Na]⁺ 358.1261; found 358.1277.

Compound 14a

The reaction was set up according to genera procedure for amide couplings with HATU. The crude product was purified by flash chromatography (silica gel, 10% hexane: 90% EtOAc) to give 0.063 g (77% yield) of compound 14a as a yellow oil.

FTIR (neat): 3296, 3079, 2931, 2858, 1748, 1650, 1544, 1439, 1374, 1177, 1061 cm⁻¹.

1H NMR (500 MHz, CDCl₃) δ = 7.29–7.18 (m, 5 H), 6.90 (t, J = 5.6 Hz, 1 H), 6.51 (d, J = 9.3 Hz, 1 H), 6.26 (d, J = 6.1 Hz, 1 H), 5.90 (ddt, J = 17.2, 10.4, 5.7 Hz, 1 H), 5.42–5.29 (m, 3 H)), 5.24 (dq, J = 10.5, 1.3 Hz, 1 H), 5.03–4.95 (m, 2 H)), 4.86 (dd, J = 9.3, 2.2 Hz, 1 H), 4.63 (dt, J = 5.7, 1.4 Hz, 2 H)), 4.29 (dt, J = 7.5, 6.3 Hz, 1 H), 3.73 (dd, J = 9.1, 2.2 Hz, 1 H), 3.44–3.30 (m, 2 H)), 3.13 (m, 1 H), 3.02 (dd, J = 14.4, 9.6 Hz, 1 H), 2.28–2.16 (m, 5 H), 1.97 (s, 3 H)), 1.72–1.58 (m, 3 H)), 1.48 (m, 2 H)), 1.37–1.19 (m, 6 H), 1.02 (d, J = 6.6 Hz, 3 H)), 0.94 (d, J = 6.6 Hz, 3 H)).

13C{¹H} NMR (126 MHz, CDCl₃) δ = 173.66, 173.60, 173.43, 171.40, 171.29, 171.24, 139.24, 139.17, 134.32, 134.30, 134.26, 134.06, 132.07, 132.05, 132.02, 131.17, 131.02, 130.99, 129.69, 129.50, 129.46, 129.45, 122.24, 122.22, 122.17, 121.99, 121.83, 121.37, 121.33, 89.59, 80.07, 79.94, 79.69, 79.43, 77.88, 77.70, 77.65, 73.44, 68.69, 68.64, 56.98, 55.42, 55.38, 42.12, 42.07, 41.97, 40.46, 40.44, 39.08, 37.55, 37.48, 34.91, 31.83, 31.77, 31.68, 31.64, 31.53, 31.42, 31.25, 31.02, 30.78, 29.22, 29.17, 28.76, 28.57, 25.29, 19.97, 19.85.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C_{33 H)49}O₈N₃ [M + Na]⁺638.3412; found 638.3385.

Compound 1a

The reaction was set up according to the general procedure for a ring closing metathesis. The crude product was then purified by flash chromatography (silica gel, 10% hexane: 90% EtOAc) to give 0.052 g (57% yield) of compound 1a as a clear foaming oil.

FTIR (neat): 3294, 2931, 2858, 1746, 1653, 1542, 1439, 1375, 1176, 1061 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 7.32–7.20 (m, 7 H), 6.72 (t, J = 5.6 Hz, 1 H), 6.15 (d, J = 6.3 Hz, 1 H), 5.54–5.49 (m, 2 H)), 5.37 (dd, J = 9.3, 3.8 Hz, 1 H), 4.64 (dd, J = 13.1, 3.6 Hz, 1 H), 4.49–4.42 (m, 2 H)), 4.41–4.35 (m, 1 H), 4.34–4.28 (m, 1 H), 3.43–3.34 (m, 2 H)), 3.23 (ddd, J = 13.3, 8.7, 3.7 Hz, 1 H), 3.07 (dd, J = 14.4, 9.3 Hz, 1 H), 2.27 (t, J = 7.5 Hz, 3 H)), 2.17–2.09 (m, 1 H), 2.00 (s, 3 H)), 1.85 (dt, J = 13.3, 6.7 Hz, 1 H), 1.70–1.50 (m, 4 H)), 1.39–1.20 (m, 4 H)), 0.94 (dd, J = 14.7, 6.8 Hz, 6 H).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 178.31, 175.43, 174.98, 174.03, 172.89, 160.59, 160.02, 140.38, 133.18, 133.13, 132.20, 131.95, 131.86, 130.57, 130.41, 130.08, 81.45, 80.94, 79.32, 68.45, 65.15, 64.73, 60.46, 60.19, 57.22, 56.26, 56.14, 42.60, 41.37, 39.02, 38.19, 34.44, 33.26, 33.19, 31.93, 31.84, 31.59, 30.86, 30.76, 29.51, 29.23, 28.00, 26.70, 22.94, 22.53, 22.50, 22.42, 22.33, 21.87, 18.16, 18.06.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₃₁H₄₅O₈N₃ [M + Na]⁺610.3099; found 610.3068.

Compound 14c

The reaction was set up according to the general procedure for an amide coupling with HATU. The crude product was purified via flash chromatography (silica gel, 50% hexane: 50% EtOAc) to give 0.197 g (84% yield) of compound 14c as a clear oil.

FTIR (neat): 3306, 2931, 2858, 1747, 1699, 1650, 1536, 1440, 1372, 1259, 1156, 1156, 1058 cm⁻¹.

1H NMR (500 MHz, CDCl₃) δ = 7.31–7.16 (m, 5 H), 6.87–6.79 (m, 1 H), 6.55 (d, J = 9.3 Hz, 1 H), 5.89 (ddt, J = 17.2, 10.4, 5.7 Hz, 1 H), 5.40 (m, 2 H)), 5.35 (d, 13 Hz, 1 H), 5.25 (m, 2 H)), 5.0 (m, 2 H)), 4.84 (dd, J = 9.2, 2.2 Hz, 1 H), 4.67–4.60 (m, 2 H)), 4.17–4.02 (m, 3 H)), 3.73 (dd, J = 9.0, 2.2 Hz, 1 H), 3.35 (dd, J = 14.4, 3.6 Hz, 1 H), 3.27–3.13 (m, 2 H)), 3.06–2.99 (m, 1 H), 2.29–2.21 (m, 5 H), 1.73–1.60 (m, 4 H)), 1.51–1.41 (m, 3 H)), 1.35–1.28 (m, 2 H)), 1.27–1.20 (m, 3 H)), 1.02 (d, J = 6.7 Hz, 3 H)), 0.96–0.92 (m, 5 H).

13C{¹H} NMR (126 MHz, CDCl₃) δ = 176.21, 174.32, 174.18, 173.15, 171.41, 171.24, 159.27, 139.04, 134.35, 134.31, 134.28, 132.08, 132.06, 131.23, 131.00, 129.49, 122.18, 121.33, 121.20, 89.59, 80.18, 77.72, 77.65, 73.41, 68.61, 64.17, 64.13, 57.04, 56.40, 56.37, 42.10, 41.99, 41.26, 41.24, 40.44, 40.43, 39.04, 37.83, 34.90, 33.81, 31.72, 31.65, 31.54, 31.52, 31.43, 31.23, 30.75, 29.25, 28.95, 28.57, 27.99, 21.67, 21.61, 19.95, 19.83, 17.12.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C_{34 H)51}O₉N₃ [M + Na]⁺ 668.3518; found 668.3484.

Compound 1c

The reaction was set up according to the general procedure for a ring closing metathesis. The crude product was purified via flash chromatography (silica gel, 40% hexane: 60% EtOAc) to give 0.108 g (64% yield) of compound 1c as a tan oil.

FTIR (neat): 3305, 2931, 1700, 1652, 1537, 1454, 1372, 1260, 1175, 1059 cm⁻¹.

1H NMR (500 MHz, CDCl₃) δ = 7.32–7.19 (m, 5 H), 6.67 (m, 1 H), 6.47 (d, J = 13.1, 1 H), 5.75 (d, J = 7.2 Hz, 1 H), 5.56 (m, 2 H)), 5.5–5.3 (m, 2 H)), 5.25 (dd, J = 9.3, 3.7 Hz, 1 H), 5.15 (dt, J = 15.5, 5.0 Hz, 1 H), 4.74 (dd, J = 8.7, 2.0 Hz, 1 H), 4.62 (dd, J = 13.3, 5.1 Hz, 1 H), 4.38–4.29 (m, 2 H)), 4.11 (m, 3 H)), 3.74 (m, 1 H), 3.48–3.37 (m, 1 H), 3.32 (m, 1 H), 3.19–3.03 (m, 3 H)), 2.87 (m, 1 H), 2.39–2.22 (m, 3 H)), 2.14 (m, 1 H), 1.78–1.69 (m, 1 H), 1.68–1.60 (m, 2 H)), 1.55 (m, 1 H), 1.40–1.19 (m, 5 H), 1.02 (d, J = 16.3, 3 H)), 0.97–0.90 (m, 3 H)).

13C{¹H} NMR (126 MHz, CDCl₃) δ = 176.77, 174.04, 173.29, 171.68, 159.20, 139.13, 132.2, 130.40, 129.80, 78.11, 66.68, 64.19, 56.82, 41.72, 40.31, 36.84, 34.91, 33.77, 30.85, 28.81, 27.74, 26.60, 23.73, 22.17, 19.87, 17.16.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₃₂H₄₇O₉N₃ [M + Na]⁺640.3205; found 640.3181.

Compound 2

A stirred solution 0.501 g (2.73 mmol) of the hydrochloride salt of β-hydroxyluecine with 0.686 g NaHCO₃ (8.16 mmol) in 9 mL H₂O was brought to 0 °C. THF (9 mL) was then added to the stirring solution, followed by 1.2 mL (16.3 mmol) of methyl chloroformate in three 0.4 mL portions over the next 30 minutes. The reaction was allowed to stir and reach room temperature overnight. After 24 hours, the reaction was diluted with H₂O (30 mL), and then washed with EtOAc (2 × 30 mL). The organic layer was then back extracted with saturated NaHCO₃(3 × 25 mL). The combined aqueous layers were then acidified to pH = 2 with 1 M HCl and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried with MgSO₄ and concentrated in vacuo to give 0.497 g (89% yield) as a white foaming oil that was used without further purification.

FTIR (neat): 3338, 2946, 2835, 2477, 2071, 1703, 1404, 1120, 1027 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 6.03 (d, J = 9.6 Hz, 1 H), 4.55 (d, J = 9.5 Hz, 1 H), 3.85–3.77 (m, 1 H), 3.70 (s, 5 H), 1.83–1.72 (m, 1 H), 1.03 (d, J = 6.7 Hz, 4 H)), 0.93 (t, J = 7.7 Hz, 5 H).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 178.48, 160.44, 80.19, 59.48, 58.97, 55.80, 55.31, 33.35, 21.77, 21.50.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₈H₁₅O₅N₁ [M + Na]⁺ 228.0842; found 228.0852.

Compound 3; Lactone Formation

Compound 2(0.526 g, 2.13 mmol, 1 equivalent) was added to a flame dried flask with 0.816 g (2.54 mmol, 1.2 equivalents) of TBTU. The compounds were then dissolved in 100 mL DCM and stirred at room temperature prior to addition of 0.92 mL (6.60 mmol, 3.1 equivalents) triethylamine. After 20 hours, the reaction was washed with saturated NaHCO₃ (3 × 35 mL). The aqueous layer was then back extracted with DCM (2 × 30 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuo. The crude product was then purified by flash chromatography (silica gel, 80% hexane: 20% EtOAc) to give 0.341 g (75% yield) of compound 3 as a white solid.

FTIR (neat): 3310, 1826, 1695, 1549,1351, 1276, 1132, 1077, 1003, 861 cm⁻¹.

¹H NMR (300 MHz, CDCl₃) δ = 5.50 (d, J = 5.8 Hz, 2 H)), 4.25 (dd, J = 10.2, 5.6 Hz, 1 H), 3.74 (s, 3 H)), 1.91 (dp, J = 10.3, 6.6 Hz, 1 H), 1.09 (d, J = 6.5, 0.9 Hz, 3 H)), 0.92 (d, J = 6.6, 0.9 Hz, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 169.70, 156.08, 82.82, 59.53, 53.05, 28.60, 18.46, 17.26.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₈H₁₃NO₄[M + Na]⁺ 210.0737; found 210.0738.

Compound 17

To a stirred solution at room temperature of 0.228 g (1.11 mmol) compound 2 with 0.452 g (1.40mmol) Cs₂CO₃ in 4.5 mL of DMF was added 0.18 mL (1.51 mmol) benzyl bromide drop-wise. Reaction was stirred at room temperature overnight. After 24 hours, the reaction was diluted with 50 mL EtOAc. The organic layer was then washed with saturated NaHCO₃(3 × 30 mL). The aqueous layer was then back extracted with EtOAc (2 × 20 mL). The combined organic layers were then dried with MgSO₄ and concentrated in vacuo. The crude product was then purified by flash chromatography (silica gel, 70% hexane:30% EtOAc) to give 0.250 g (76% yield) of compound 17 as a yellow oil.

FTIR (neat): 3392, 2961, 1703, 1523, 1456, 1380, 1340, 1274, 1210, 1166, 1117, 1060, 1001 cm⁻¹.

¹H NMR (500 MHz CDCl₃) δ = 7.41–7.30 (m, 5 H), 5.59 (d, J = 9.7 Hz, 1 H), 5.20 (s, 2 H)), 4.62–4.56 (m, 1 H), 3.73 (dt, J = 9.1, 3.8 Hz, 1 H), 3.70 (s, 4 H)), 2.25 (d, J = 5.3 Hz, 1 H), 0.98 (dd, J = 31.2, 6.7 Hz, 8 H).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 174.54, 159.95, 138.04, 131.26, 131.05, 130.76, 80.16, 69.93, 58.94, 55.16, 33.47, 21.53.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₁₅H₂₁O₅N₁ [M + Na]⁺ 318.1312; found 318.1320.

Compound 18

To a stirring solution of 0.240 g (0.81 mmol) compound 17 in 5 mL DCM at 0 °C was added 0.47 mL (4.1 mmol) of 2,6-lutidine, followed by dropwise addition of 0.47 mL (2.0 mmol) TBDMS-triflate. The reaction was allowed to stir and reach RT overnight. After 24 hours, the reaction was quenched with 1 M HCl (20 mL). The reaction was then extracted with DCM (3 × 25 mL). The combined organic layers were then washed with 1 M KHSO₄ (30 mL) and brine (30 mL). The organic layer was then dried with MgSO₄ and concentrated in vacuo. The crude product was then purified by flash chromatography (silica gel, 90% hexane: 10% EtOAc) to give 0.230 g (69% yield) of compound as a clear oil.

FTIR (neat): 3450, 2956, 2857, 1728, 1500, 1463, 1337, 1252, 1202, 1165, 1081, 1059, 1003.

¹H NMR (500 MHz, CDCl₃) δ = 7.42–7.29 (m, 6 H), 5.39 (d, J = 9.5 Hz, 1 H), 5.16 (d, J = 1.7 Hz, 2 H)), 4.48 (dd, J = 9.5, 1.3 Hz, 1 H), 3.97 (dd, J = 6.6, 1.3 Hz, 1 H), 3.70 (s, 3 H)), 1.89–1.78 (m, 1 H), 1.02–0.90 (m, 10 H), 0.88 (s, 10 H), 0.03 (s, 4 H)), −0.08 (s, 3 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 174.39, 159.51, 137.93, 131.22, 130.99, 130.87, 79.55, 69.89, 69.71, 59.11, 58.48, 55.00, 35.68, 28.55, 28.31, 21.92, 21.83, 20.91, 20.79, −1.63, −2.08.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₂₁H₃₅O₅N₁Si₁ [M + Na]⁺ 432.2177; found 432.2177.

Compound 19

Pd on C (0.015 g, 0.13 mmol) was added to a flame dried round bottom flask containing 0.172 g (0.42 mmol) of compound 18and 1.8 mL of methanol. The flask was put under positive pressure of H₂, and the reaction stirred vigorously for 6 h. The reaction was then diluted with MeOH (25 mL) and passed through a plug column of celite. The reaction was then concentrated in vacuoto give 0.116 g (86% yield) compound 19 as a clear oil that was used without further purification.

FTIR (neat): 2957, 2930, 2858, 1715, 1511, 1465, 1362, 1253, 1213, 1080 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 5.40 (d, J = 9.1 Hz, 1 H), 4.45 (dd, J = 9.1, 1.5 Hz, 1 H), 4.01 (dd, J = 6.7, 1.5 Hz, 1 H), 3.71 (s, 4 H)), 1.90–1.77 (m, 1 H), 0.95 (dd, J = 14.0, 6.9 Hz, 8 H), 0.89 (s, 12 H)), 0.08 (s, 4 H)), 0.01 (d, J = 8.5 Hz, 4 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 159.61, 80.57–78.28 (m), 58.37, 55.17, 35.40, 28.56, 22.86–19.95 (m), −1.82 (d, J = 30.6 Hz).

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C_{14 H)29}O₅N₁Si₁ [M + Na]⁺ 342.1707; found 342.1706.

Compound 6a

The reaction was set up according to the general procedure for generating the macrocycle. The crude product was purified via flash chromatography (silica gel, 50–90% EtOAc) to give 0.016 g (53% yield) of compound 6a as a clear foaming oil. Efforts were not made to fully characterize compound 6a as it was deemed unnecessary to purify material prior to subsequent TBDMS deprotection.

¹H NMR (300 MHz, CDCl₃) δ = 7.38–7.24 (m, 5.09 H), 7.07 (d, J = 11.8 Hz, 0.56 H), 6.91 (d, J = 5.9 Hz, 0.23 H)), 6.77 (m, 0.62 H)), 6.56 (dd, J =19 Hz, 8.5 Hz, 0.64 H)), 5.50 (d, J = 9 Hz, 0.91 H), 5.46–5.28 (m, 0.87 H), 5.14 (dd, J = 12 Hz, 3.2 Hz, 0.69 H) 5.00 (m, 0.83 H)), 4.96–4.75 (m, 0.91 H) 4.70 (dd, J = 8.6 Hz, 4.0 Hz, 0.62 H)), 4.55 (m, 1.40 H), 4.30–4.20 (m, 1.78 H), 3.88 (m, 1.78 H), 3.71 (s, 2.62 H)) 3.40–2.97 (m, 2.8 H), 2.80 (s, 1.31 H) 2.30 (m, 2.43 H)), 1.88–1.05 (m, 9.85 H), 1.04–0.73 (m, 19.78 H), 0.09 (s, 2.21 H), 0.03 (s, 2.18 H).

Compound 5a

The crude material was then subjected to the general procedure for a TBDMS deprotection. The crude product was then purified via flash chromatography (silica gel, 100% EtOAc) to give 0.008 g (56% yield) of compound 5a as a tan oil over two steps.

FTIR (neat): 3306, 2924, 2854, 1745, 1652, 1536, 1456, 1375, 1264, 1172, 1060 cm⁻¹.

¹H NMR (500 MHz, CDCl₃) δ = 7.32–7.14 (m, 6 H), 6.27 (d, J = 9.4 Hz, 2 H)), 6.18 (s, 1 H), 5.60 (dp, J = 27.5, 7.3, 6.4 Hz, 2 H)), 5.46–5.37 (m, 1 H), 5.29 (d, J = 7.5 Hz, 1 H), 5.06 (ddd, J = 11.2, 9.2, 1.9 Hz, 1 H), 4.92–4.77 (m, 1 H), 4.75–4.65 (m, 1 H), 4.46–4.37 (m, 2 H)), 4.34–4.28 (m, 2 H)), 4.12 (dq, J = 10.5, 7.1 Hz, 4 H)), 3.75–3.63 (m, 5 H), 3.62 (d, J = 8.8 Hz, 1 H), 3.28–3.17 (m, 4 H)), 3.16 (dt, J = 14.4, 4.3 Hz, 3 H)), 2.55 (d, J = 14.7 Hz, 2 H)), 2.43 (q, J = 7.1 Hz, 1 H), 2.32 (s, 2 H)), 2.20 (ddd, J = 14.5, 8.5, 6.0 Hz, 1 H), 2.05 (s, 2 H)), 1.95 (dp, J = 9.2, 6.7 Hz, 1 H), 1.75 (s, 7 H), 1.58–1.51 (m, 1 H), 1.42 (dd, J = 13.7, 7.0 Hz, 2 H)), 1.27 (ddd, J = 9.3, 6.1, 2.5 Hz, 13 H)), 1.11–0.96 (m, 8 H), 0.93 (q, J = 6.0, 5.5 Hz, 10 H).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 177.81, 174.66, 174.00, 173.07, 173.05, 172.11, 159.63, 139.75, 132.25, 132.17, 131.09, 131.07, 129.53, 128.36, 128.34, 78.95, 67.40, 59.60, 59.57, 55.28, 55.23, 54.61, 41.67, 40.51, 37.98, 37.24, 33.60, 32.37, 32.34, 30.88, 30.86, 29.61, 29.45, 26.69, 25.30, 22.11, 21.63, 21.43, 21.02.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₃₉H₅₈O₁₂N₄ [M + Na]⁺775.4124; found 775.4089.

Compound 6c

The reaction was set up according to the general procedure for generating the macrocycle. The crude product was purified via flash chromatography (silica gel, 60% hexane: 40% EtOAc) to give 0.056 g (84% yield) of compound 6c as a clear foaming oil. Efforts were not made to fully characterize intermediate 6c as it was deemed unnecessary to purify material prior to subsequent TBDMS deprotection.

¹H NMR (500 MHz, CDCl₃) δ = 7.35–7.14 (m, 5 H), 6.73 (m, 1.10 H), 6.53 (m, 1.15 H), 5.81 (m, 1.44 H)), 5.54 (m, 1.12 H)), 5.48 (d, J = 7.5 Hz, 0.92 H)), 5.39 (m, 0.43 H)), 5.33–5.19 (m, 1.37 H), 5.11 (d, J = 15.1 Hz, 1.09 H), 5.04–4.87 (m, 1.62 H)), 4.87–4.71 (m, 1.33 H)), 4.60 (m, 1.25 H), 4.34–4.17 (m, 2.72 H)), 4.10 (m, 1.99 H), 3.90–3.81 (m, 1.07 H), 3.69 (s, 2.66), 3.38–3.16 (m, 2.96 H), 3.15–3.00 (m, 1.48 H), 2.29 (s, 3.43 H)), 2.19–1.10 (m, 18.52 H)), 1.09–0.80 (m, 18.1 H), 0.08 (s, 2.43 H)), 0.04 (s, 2.26 H).

Compound 5c

The reaction was set up according to the general procedure for deprotection of a TBDMS group. The crude product was purified via flash chromatography (silica gel, 70% hexane: 30% EtOAc) to give 0.011 g (76% yield) of compound 5c as a tan oil.

FTIR (neat): 3308, 2930, 1703, 1652, 1529, 1455, 1372, 1260, 1169, 1114, 1058 cm⁻¹.

1H NMR (500 MHz, CDCl₃) δ = 7.33–7.20 (m, 5 H), 6.85–6.79 (m, 1 H), 6.28 (d, J = 8.4 Hz, 1 H), 6.02 (d, J = 7.1 Hz, 1 H), 5.56 (d, J = 8.8 Hz, 1 H), 5.49– 5.34 (m, 2 H)), 5.10 (dd, J = 10.1, 3.3 Hz, 1 H), 5.00 (dd, J = 7.5 Hz, 1Hz, 1 H), 4.81 (d, J = 12.0 Hz, 1 H), 4.73 (d, J = 8.5 Hz, 1 H), 4.52 (m, 1 H), 4.44 (d, J = 9.7 Hz, 1 H), 4.38 (m, 1 H), 4.22 (d, J = 13.3 Hz, 1 H), 4.16–4.04 (m, 3 H)), 3.74–3.67 (m, 4 H)), 3.45–3.05 (m, 4 H)), 2.38–2.21 (m, 3 H)), 2.12 (m, 1 H), 1.94 (m, 1 H), 1.80 (m, 1 H), 1.75–1.67 (m, 1 H), 1.65–1.56 (m, 1 H), 1.53–1.43 (m, 2 H)), 1.41–1.31 (m, 5 H), 1.29–1.20 (m, 3 H)), 1.12–0.87 (m, 12 H)).

¹³C{¹H} NMR (126 MHz, CDCl₃) δ = 177.34, 174.52, 174.05, 173.11, 171.91, 159.64, 139.47, 132.26, 132.21, 131.04, 130.94, 129.63, 129.54, 129.48, 129.16, 80.51, 80.00, 78.37, 67.55, 64.22, 63.80, 63.03, 59.53, 56.30, 55.34, 55.22, 41.69, 40.45, 38.80, 38.13, 37.28, 33.53, 33.39, 32.27, 30.95, 30.87, 30.66, 29.95, 29.84, 28.53, 27.10, 25.82, 23.69, 22.02, 21.79, 21.62, 21.50, 21.42, 20.95, 17.24, 17.14, 16.85, 16.76, 2.64.

HRMS (ESI/TOF-Q) m/z: [M + Na]⁺ calcd for C₄₀H₆₀O₁₃N₄ [M + Na]⁺827.4049; found 827.3993.

Compound 20a

Grubbs II catalyst [0.132 mmol (0.2 equiv)] was added to a flame-dried round bottom flask. The flask was then charged with 120 mL of dry DCM, and the solution brought to reflux under Ar. To this flask was added 0.66 mmol (1 equiv.) of the acyclic starting material in 20 mL of dry DCM, and the reaction mixture was refluxed overnight under Ar. The reaction was then cooled, and silica gel was added to the mixture. The solvent was evaporated and then the silica coated with the crude reaction product added to the top of a silica gel column. The column was eluted with a 1:1 to 1:9 hexanes/ethyl acetate mixture to to yield 204 mg (50%) of the cyclic product as a brown glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.23–7.14 (m, 2 H)), 6.91–6.80 (m, 3 H)), 6.59–6.45 (m, 1 H), 6.41 (d, J = 8.5 Hz, 1 H), 5.51–5.33 (m, 1 H), 5.23 (dd, J = 10.4, 3.5 Hz, 1 H), 5.01 (dt, J = 15.6, 4.6 Hz, 1 H), 4.69 (dd, J = 8.5, 2.0 Hz, 1 H), 4.65–4.51 (m, 2 H)), 4.32 (m, 1 H), 3.85 (s, 3 H)), 3.75–3.68 (m, 1 H), 3.44 (dd, J = 12.0 3.0 Hz, 1 H), 3.41–3.35 (m, 1 H), 3.28–3.2 (m, 1 H), 2.95 (dd, J = 10.0, 7.0 Hz, 1 H), 2.70 (d, J = 4.6 Hz, 1 H), 2.35–2.17 (m, 3 H)), 2.04 (s, 3 H)), 2.02–1.94 (m, 1 H), 1.79–1.59 (m, 3 H)), 1.57–1.43 (m, 2 H)), 1.38–1.28 (m, 6 H), 1.03 (d, J = 10.3, 3 H)), 0.92 (d, J = 11.3, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 174.47, 170.88, 170.84, 169.53, 157.80, 131.70, 128.42, 127.47, 127.39, 125.22, 120.43, 110.53, 76.71, 74.56, 63.88, 55.61, 54.44, 52.24, 39.27, 35.90, 34.14, 32.62, 31.14, 28.71, 27.89, 27.84, 25.14, 23.74, 22.88, 19.06, 19.02.

HRMS (ESI): calculated for C_{32 H)47}N₃O⁹ [M + Na]⁺: 640.3210, found 640.3202.

Compound 20b

Grubbs II catalyst [0.2 mmol (0.2 equiv)] was added to a flame-dried round bottom flask, 180 mL of dry DCM added, and the solution brought to reflux under Ar. To this flask was added 1 mmol of the acyclic starting material (1 equiv) that had been dissolved in 20 mL dry DCM. The reaction mixture was refluxed overnight under Ar. Silica gel was then added directly to the cooled reaction mixture, the solvent removed, and the silica gel now coated with the crude reaction product placed on the top of a chromatography column. The column was eluted with 1:1 to 1:9 hexanes/ethyl acetate to yield 204 mg (49%) of the purified, cyclic product as a brown glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.19 (t, J = 7.4 Hz, 1 H), 6.93 (d, J = 6.9 Hz, 1 H), 6.89–6.73 (m, 3 H)), 6.70–6.62 (m, 1 H), 6.50 (d, J = 8.7 Hz, 1 H), 5.59–5.43 (m, 1 H), 5.20 (dd, J = 8.0, 2.0 Hz, 1 H), 5.14 (dt, J = 15.0, 3.0 Hz, 1 H), 4.68 (dd, J = 8.4, 2.0 Hz, 1 H), 4.61 (dd, J = 13.7, 5.1 Hz, 1 H), 4.53 (dt, J = 7.1, 5.0 Hz, 1 H), 4.42–4.34 (m, 1 H), 3.78 (s, 3 H)), 3.72 (m, 1 H), 3.54–3.39 (m, 1 H), 3.33 (dd, J = 14.3, 3.5 Hz, 1 H), 3.15–3.05 (m, 1 H), 3.00 (dd, J = 14.0, 9.0 Hz, 1 H), 2.77 (m, 1 H), 2.38–2.17 (m, 3 H)), 2.12–1.97 (m, 1 H), 2.04 (s, 3 H)), 1.93 (m, 1 H), 1.82–1.15 (m, 10 H), 1.04 (d, J = 8.8 Hz, 3 H)), 0.92 (d, J = 9.2 Hz, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 174.44, 171.48, 171.15, 170.65, 168.99, 159.60, 138.40, 129.32, 127.59, 126.55, 121.90, 114.85, 112.33, 76.46, 75.65, 63.76, 55.22, 55.20, 54.48, 54.39, 52.18, 39.13, 37.56, 35.93, 33.78, 31.11, 28.41, 27.89, 27.77, 24.77, 24.60, 23.61, 22.69, 21.06, 18.93, 18.90, 14.20.

HRMS (ESI): calculated for C_{32 H)47}N₃O₉ [M + H]⁺: 618.3385, found 618.3372.

Compound 20c

Grubbs II catalyst (1:1 to 1:9 hexanes/ethyl acetate) was added to the acyclic substrate and the ring closing metathesis reaction used to yield 12 mg (78%) of the desired product as outlined in the previous two experiments. Once again, the product was obtained as a brown glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.14 (d, J = 8.9 Hz, 2 H)), 6.93 (d, J = 7.0 Hz, 1 H), 6.81 (d, J = 8.9 Hz, 2 H)), 6.62 (m, 1 H), 6.46 (d, J = 18.2 Hz, 1 H), 5.61–5.39 (m, 1 H), 5.17 (dt, J = 14.0, 4.0 Hz, 1 H), 4.66 (dd, J = 7.0, 1.0 Hz, 1 H), 4.61 (dd, J = 12.0, 4.0 Hz, 1 H), 4.55 (dt, J = 7.0, 5.1 Hz, 1 H), 4.47–4.35 (m, 1 H), 3.77 (s, 3 H)), 3.72 (dd, J = 8.7, 2.0 Hz, 1 H), 3.49–3.39 (m, 1 H), 3.24 (dd, J = 14.4, 3.6 Hz, 1 H), 3.14–3.06 (m, 1 H), 2.98 (dd, J = 14.4, 9.0 Hz, 1 H), 2.39–2.07 (m, 6 H), 2.02 (s, 3 H)), 1.74–1.10 (m, 11 H)1 H), 1.02 (d, J = 8.2 Hz, 3 H)), 0.90 (d, J = 9.1 Hz, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 174.61, 171.67, 171.23, 170.78, 169.23, 158.69, 130.70, 130.66, 128.84, 127.73, 127.06, 126.73, 113.91, 113.85, 76.61, 76.13, 64.02, 60.54, 55.44, 54.60, 52.32, 39.21, 36.89, 36.00, 34.09, 31.25, 29.23, 28.51, 27.96, 27.88, 24.93, 24.73, 23.76, 22.84, 19.05, 19.00, 14.34.

HRMS (ESI): calculated for C₃₂H₄₇N₃O₉ [M + H]⁺: 617.3385, found 618.3385.

Compound 21a

In a flame dried round bottom flask, 0.081 mmol (1 equiv) of 20a and 0.16 mmol (2 equiv) of compound 19were dissolved in 5 mL dry DCM under Ar. The flask was cooled to 0 °C, and 0.16 mmol (2 equiv) of EDC-HCl and 0.012 mmol (0.15 equiv) of DMAP dissolved in 1 mL dry DCM were added. The reaction was gradually warmed to room temperature and stirred overnight. The reaction mixture was then diluted with DCM and washed with saturated NaHCO₃ (3 × 15 mL) to remove any unreacted 19. The aqueous layers were combined and extracted with DCM (2 × 10 mL). The organic layers were then combined, dried over Na₂SO₄, and concentrated. The crude material was purified by silica gel chromatography (1:1 to 1:6 hexanes/ethyl acetate) to yield 54 mg (73%) of the coupled product as a pale glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.24–7.13 (m, 2 H)), 7.10 (d, J = 7.1 Hz, 1 H), 6.90–6.79 (m, 2 H)), 6.61–6.51 (m, 2 H)), 5.48 (d, J = 12.4 Hz, 1 H), 5.35 (m, 1 H), 5.26–5.19 (m, 1 H), 5.07–4.96 (m, 1 H), 4.90–4.80 (m, 1 H), 4.73 (dd, J = 8.6, 4.0 Hz, 1 H), 4.56 (m, 2 H)), 4.30–4.18 (m, 2 H)), 3.88 (m, 1 H), 3.84 (s, 3 H)), 3.71 (s, 3 H)), 3.44 (dd, J = 13.9, 3.3 Hz, 1 H), 3.38–3.31 (m, 1 H), 3.26 (m, 1 H), 2.93 (dd, J = 14.1, 10.7 Hz, 1 H), 2.36 (m, 1 H), 2.26 (m, 2 H)), 2.04 (s, 3 H)), 1.95 (m, 2 H)), 1.82 (m, 1 H), 1.64 (m, 2 H)), 1.52–1.15 (m, 8 H), 1.03–0.84 (m, 21 H), 0.09 (s, 3 H)), 0.04 (s, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 175.08, 170.88, 170.55, 169.85, 169.57, 157.91, 156.96, 131.74, 128.28, 127.30, 126.28, 125.43, 120.28, 110.22, 78.05, 75.54, 74.40, 63.87, 55.40, 55.16, 53.72, 52.67, 52.14, 39.17, 35.46, 34.27, 33.74, 32.75, 29.86, 28.69, 27.75, 26.01, 24.67, 22.96, 22.77, 19.37, 18.37, 18.24, 17.92, −4.20, −4.32.

HRMS (ESI): calculated for C₄₆H₇₄N₄O₁₃Si [M + Na]⁺: 941.4919, found 941.4885.

Compound 21b

In a flame-dried round bottom flask, 0.274 mmol (1 equiv) of 20b and 0.548 mmol (2 equiv) of compound 19 were dissolved in 9 mL of dry DCM under Ar. The flask was cooled to 0 °C, and 0.548 mmol (2 equiv) of EDC-HCl and 0.041 mmol (0.15 equiv) of DMAP dissolved in 5 mL dry DCM were added. The reaction was gradually warmed to room temperature and stirred overnight. The reaction mixture was diluted with DCM and washed with saturated NaHCO₃ (3 × 15 mL) to remove any unreacted 19. The aqueous layers were combined and extracted with DCM (2 × 10 mL). The organic layers were then combined, dried over Na₂SO₄, and concentrated. The crude material was purified by silica gel chromatography (1:1 to 1:9 hexanes/ethyl acetate) to yield 203 mg (81%) product as a pale glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.18 (t, J = 7.4 Hz, 1 H), 7.09 (d, J = 6.9 Hz, 0.5 H), 6.86–6.71 (m, 3.5 H), 6.53 (d, J = 7.1 Hz, 1 H), 5.49 (d, J = 8.1 Hz, 1 H), 5.45–5.36 (m, 1 H), 5.15 (dd, J = 10.4, 3.3 Hz, 1 H), 5.08–4.95 (m, 2 H)), 4.72 (dd, J = 8.1, 4.1 Hz, 1 H), 4.61–4.53 (m, 2 H)), 4.39–4.30 (m, 2 H)), 4.01–3.92 (m, 1 H), 3.87 (m, 1 H), 3.78 (s, 3 H)), 3.71 (s, 3 H)), 3.42–3.35 (m, 1 H), 3.29 (dd, J = 14.3, 3.3 Hz, 1 H), 3.25–3.15 (m, 1 H), 3.03 (dd, J = 24.6, 14.4 Hz, 1 H), 2.40–2.21 (m, 3 H)), 2.11–1.76 (m, 7 H), 1.68–1.46 (m, 9 H), 1.42–1.15 (m, 7 H), 1.04–0.85 (m, 21 H), 0.09 (s, 3 H)), 0.04 (s, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 174.93, 171.71, 170.99, 170.40, 169.75, 169.04, 159.62, 156.85, 138.55, 129.29, 127.39, 126.24, 121.87, 114.68, 112.54, 107.91, 77.89, 75.64, 75.43, 67.66, 63.90, 55.16, 55.09, 53.76, 52.53, 52.13, 39.12, 37.68, 35.46, 33.95, 33.60, 30.32, 29.72, 29.16, 28.39, 27.68, 25.87, 24.55, 23.92, 23.04, 22.64, 19.25, 18.23, 18.10, 17.79, 17.73, −4.34, −4.44.

HRMS (ESI): calculated for C₄₆H₇₄N₄O₁₃Si [M + H]⁺: 919.5094, found 919.5089.

Compound 21c

In a flame-dried round bottom flask, 0.1 mmol (1 equiv) of 20cand 0.2 mmol (2 equiv) of compound 19were dissolved in 3 mL of dry DCM under Ar. The flask was cooled to 0 °C, and 0.2 mmol (2 equiv) of EDC-HCl and 0.015 mmol (0.15 equiv) of DMAP dissolved in 1 mL dry DCM were added. The reaction was gradually warmed to room temperature and stirred overnight. The reaction mixture was diluted with DCM and washed with saturated NaHCO₃ (3 × 15 mL) to remove any unreacted 19. The aqueous layers were combined and extracted with DCM (2 × 10 mL). The organic layers were then combined, dried over Na₂SO₄, and concentrated. The crude material was purified by silica gel chromatography (1:1 to 1:4 hexanes/ethyl acetate) to yield 90 mg (98%) product as a pale glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.18 (q, J = 9 Hz, 2 H)), 7.09 (d, J = 6.9 Hz, 1 H), 6.81 (d, J = 9 Hz, 2 H)), 6.71 (m, 1 H), 6.52 (d, J = 8.1 Hz, 1 H), 5.49 (dd, J = 8.1, 2.7 Hz, 1 H), 5.45–5.36(m, 1 H), 5.15 (dd, J = 10.4, 3.3 Hz, 1 H), 5.08–4.95 (m, 2 H)), 4.72 (dd, J = 8.1, 4.1 Hz, 1 H), 4.61–4.53 (m, 2 H)), 4.39–4.18 (m, 2 H)), 3.87 (m, 1 H), 3.78 (s, 3 H)), 3.71 (s, 3 H)), 3.42–3.20 (m, 2 H)), 3.03 (dd, J = 24.6, 14.4 Hz, 1 H), 2.40–2.21 (m, 3 H)), 2.13–1.77 (m, 4 H)), 2.04 (s, 3 H)), 1.68–1.16 (m, 10 H), 1.04–0.85 (m, 21 H), 0.09 (s, 3 H)), 0.04 (s, 3 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 175.02, 171.84, 171.05, 170.57, 169.94, 169.22, 158.64, 156.99, 130.66, 130.64, 129.07, 127.55, 126.70, 113.88, 113.80, 78.76, 78.06, 76.00, 75.56, 64.12, 60.52, 55.43, 55.37, 55.23, 53.92, 52.66, 52.27, 39.26, 39.19, 36.90, 35.57, 34.10, 33.76, 29.97, 29.88, 28.84, 28.48, 27.83, 26.01, 24.72, 23.20, 22.79, 21.18, 19.36, 19.19, 18.36, 18.23, 18.15, 18.10, 17.96, 17.91, 17.89, 14.34, −4.21, −4.30.

HRMS (ESI): calculated for C₄₆H₇₄N₄O₁₃Si [M + H]⁺: 919.5094, found 919.5115.

Compound 22a

In a flame-dried round bottom flask, 0.0544 mmol (1 equiv) of 21a was dissolved in 2 mL of dry THF was brought to 0 °C under Ar. To this mixture was added 160 μL (3 equiv) of a 1M TBAF solution in THF, and then the reaction was stirred for 1.75 hours at 0 °C. At this point, TLC indicated the complete consumption of starting material. The reaction was quenched by addition of saturated NH₄Cl (10 mL) and extracted with ethyl acetate (3 × 10 mL). The combined organic layers were dried over Na₂SO₄ and concentrated. The crude material was purified by silica gel chromatography (1:3 to 1:9 hexanes/ethyl acetate) to yield 27 mg (62%) product as a pale glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.25–7.12 (m, 3 H)), 6.89 (t, J = 7.4 Hz, 1 H), 6.84 (d, J = 8.0 Hz, 1 H), 6.64 (t, J = 5.7 Hz, 1 H), 6.35 (d, J = 8.4 Hz, 1 H), 5.59 (d, J = 9.0 Hz, 1 H), 5.34–5.24 (m, 1 H), 5.00 (m, 2 H)), 4.73–4.63 (m, 2 H)), 4.52 (m, 2 H)), 4.44 (d, J = 7.0 Hz, 1 H), 4.23–4.05 (m, 2 H)), 3.84 (s, 3 H)), 3.75–3.57 (m, 4 H)), 3.41 (m, 1 H), 3.32 (dd, J = 14.0, 3.2 Hz, 1 H), 3.18 (m, 1 H), 2.98 (m, 1 H), 2.37–2.21 (m, 3 H)), 2.03 (s, 3 H)), 2.01–1.91 (m, 2 H)), 1.87–1.76 (m, 1 H), 1.75–1.56 (m, 2 H)), 1.57–1.27 (m, 8 H), 1.12– 0.84 (m, 12 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 175.16, 171.83, 171.14, 170.39, 170.33, 169.83, 157.78, 157.01, 131.83, 128.27, 126.68, 125.42, 125.26, 120.29, 110.14, 77.69, 74.77, 64.43, 56.89, 55.40, 52.60, 52.52, 51.91, 39.00, 35.28, 34.71, 32.56, 30.87, 29.73, 28.39, 27.04, 26.87, 24.12, 22.64, 19.38, 18.92, 18.81, 18.32.

HRMS (ESI): calculated for C₄₀H₆₀N₄O₁₃ [M + H]⁺: 805.4230, found 805.4225.

Compound 22b

In a flame-dried round bottom flask, 0.0794 mmol (1 equiv) of 21b was dissolved in 10 mL of dry THF and then the mixture brought to 0 °C under Ar. To this solution was added 240 μL (3 equiv) of a 1 M TBAF solution in THF. The reaction was stirred for 2 hrs at 0 °C, at which point TLC indicated complete consumption of the starting material. The reaction was quenched by addition of saturated NH₄Cl (10 mL) and extracted with ethyl acetate (3 × 10 mL). The combined organic layers were dried over MgSO₄ and concentrated. The crude material was purified by silica gel chromatography (1:3 to 1:9 hexanes/ethyl acetate) to yield 50 mg (78%) product as a pale glass.

¹H NMR (500 MHz, CDCl₃) δ = 7.24–7.13 (m, 2 H)), 6.85–6.72 (m, 4 H)), 6.35 (d, J = 8.3 Hz, 1 H), 5.60 (d, J = 9 Hz, 1 H), 5.41–5.21 (m, 2 H)), 5.04 (dd, J = 9.0, 1.5 Hz, 1 H), 4.95 (dd, J = 10.7, 3.1 Hz, 1 H), 4.74–4.56 (m, 3 H)), 4.53–4.39 (m, 2 H)), 3.78 (s, 3 H)), 3.75–3.68 (m, 4 H)), 3.43 (m, 1 H), 3.22–3.09 (m, 2 H)), 3.02 (dd, J = 14.3, 10.7 Hz, 1 H), 2.41–2.18 (m, 3 H)), 2.13–1.91 (m, 6 H), 1.89–1.74 (m, 1 H), 1.77–1.56 (m, 2 H)), 1.56–1.28 (m, 8 H), 1.10–0.87 (m, 12 H)).

¹³C NMR (126 MHz, CDCl₃) δ = 177.80, 174.63, 173.94, 173.05, 173.01, 172.09, 162.34, 159.61, 141.28, 132.04, 129.49, 124.54, 117.54, 115.12, 80.27, 80.15, 78.86, 67.36, 59.56, 57.83, 55.28, 55.21, 54.60, 41.67, 40.52, 37.99, 37.28, 33.60, 32.33, 30.86, 29.57, 29.38, 26.64, 25.37, 25.30, 22.10, 21.65, 21.43, 21.00, 16.86, 2.65.

HRMS (ESI): calculated for C₄₀H₆₀N₄O₁₃ [M + H]⁺: 805.4230, found 805.4221.