Optimized Ring Closing Metathesis Reaction Conditions to Suppress Desallyl Side Products in the Solid Phase Synthesis of Cyclic Peptides involving Tyrosine(O-allyl)

Abstract

We are exploring constraining aromatic residues in the kappa opioid receptor selective antagonist arodyn (Ac[Phe^1,2,3,Arg⁴,D-Ala⁸]dynorphin A(1–11)-NH₂) by ring closing metathesis (RCM) involving tyrosine(O-allyl) (Tyr(All)), but desallyl products limited the yields of the desired cyclic peptide. The model dipeptide Fmoc-Tyr(All)-Tyr(All) was used to explore different reaction conditions, including the use of isomerization suppressants, to minimize formation of the desallyl products and enhance formation of the desired RCM product. Reaction conditions were identified that enhanced the RCM product yield while suppressing desallyl products using both second-generation Grubbs and second-generation Hoveyda Grubbs catalysts. These optimized reaction conditions were then applied to the cyclization of a tripeptide and an arodyn analog resulting in ≥70% conversion to the desired cyclic peptides. These strategies should be applicable to RCM involving Tyr(All) and similar residues in peptide and peptidomimetic cyclizations performed on solid phase.

Graphical abstract

INTRODUCTION

Cyclization of peptides is a useful approach to probe ligand binding at biological targets and to impart metabolic stability. A cyclic constraint can potentially enhance potency in vivo by stabilizing the bioactive conformation and increasing affinity and/or selectivity for the biological target,¹ improving metabolic stability and/or improving membrane permeability.^2–4 Ring closing metathesis (RCM) is a commonly used cyclization strategy in which intramolecular alkene precursors undergo olefin metathesis, resulting in a C-C double bond that reduces the conformational freedom of the product. Unlike standard peptide cyclization strategies such as disulfide and lactam linkages, RCM generates a non-native C-C bond with defined geometry that is completely stable to proteases.⁵ The synthesis of cyclic peptides by RCM, on-resin and in solution, has increased extensively^6,7 due to the broad applicability of RCM.^8–10 In addition to the development of stable catalysts,^11–14 the rising utility of RCM in peptide modification is due, in part, to the increasing availability of non-natural alkene amino acid precursors as a result of their improved syntheses.^15–17 It is not surprising then that a corresponding increase in the use of RCM derivatives in peptide and biological applications has occurred.^18–20

Peptide RCM generally features hydrocarbon bridges utilizing amino acid residues such as ω-alkenyl and allylglycine residues.^4,18,21,22 Additionally, several studies have used heteroatom-substituted allylic or homoallylic residues including allylserine, allylcysteine, and allyl-tyrosine (Tyr(All)) residues to increase the diversity of RCM bridges.^23–25 Incorporation of these amino acids into the peptide sequence allows the formation of RCM bridges with different properties.

A number of reports have described further development of catalysts and optimization of reaction conditions, including the use of additives to enhance olefin metathesis, highlighting the utility of RCM, but also several shortcomings.^26–29 Whereas ruthenium catalyzed olefin isomerization has been conveniently employed to synthesize various heterocyclic compounds^12,30,31 and complex natural products,³² it can decrease yields of RCM cyclizations, particularly in isomerization-prone substrates such aryl-allyl groups³³ and heteroatom-substituted allylic groups.^27,34 Heteroatom-substituted homoallylic groups are also susceptible to isomerization.³⁴ Olefin isomerization can result in ring contraction³⁵ or deallylation,^34,36 yielding complex mixtures and reducing the RCM product yield.³⁷

Olefin isomerization is reported to be caused by catalyst degradation products such as ruthenium hydrides.^27,38 Recently, Fogg and coworkers reported the involvement of ruthenium nanoparticles formed by catalyst degradation in olefin isomerization.³⁹ A number of additives, including phenol,⁴⁰ 1,4-benzoquinone,^27,41 and copper (I) halides,³⁶ have been reported to substantially suppress olefin isomerization. Phenol and copper iodide are also reported to increase the rate of olefin metathesis.^40,42 Although copper iodide retards ruthenium hydride formation,³⁶ it can be difficult to remove from reaction mixtures.⁴³ While 1,4-benzoquinone suppresses olefin isomerization, it also suppresses the catalytic activity of Grubbs catalysts.³³

On-resin RCM of peptides and peptidomimetics has particular advantages over RCM in solution. Reactions on solid phase allow easy removal of reagents and byproducts by filtration; on-resin RCM thus allows cyclization of the resin-bound peptide and removal of the catalyst by washing the solid support. Furthermore, pseudodilution on the solid support favors the desired intramolecular cyclization and limits intermolecular reactions.⁴⁴

A number of shortcomings have been reported in solid-phase RCM, however. Chelation of the catalyst to polar functionalities in peptides has been reported to decrease the amount of catalyst available for productive metathesis.⁴⁵ Additionally, aggregation of the peptide on the solid support can limit yields.²⁴ Approaches that have been reported to enhance on-resin RCM include additives, such as chaotropic salts or Lewis acids, and the use of microwave heating.^46,47 Sequence dependent effects also influence the efficiency of the cyclizations;²⁶ for example, Robinson and co-workers reported that incorporation of a turn-inducing residue and microwave irradiation were required to facilitate RCM of a human growth hormone fragment.²⁴

This study stems from initial attempts to prepare novel constrained analogs of the acetylated dynorphin A (Dyn A) analog arodyn (Ac[Phe^1,2,3,Arg⁴,D-Ala⁸]Dyn A(1–11)-NH₂, Figure 1) via cyclizations involving aromatic residues.^48,49 Arodyn is a potent and selective kappa opioid receptor (KOR) antagonist which has high affinity (K_i = 10 nM) and selectivity for KOR (K_i ratio (κ/μ/δ) = 1/174/583).⁵⁰ KOR antagonists, used classically as pharmacological tools, have recently shown potential for the treatment of depression and drug addiction.^51,52 Unlike cyclization of Dyn A analogs via RCM utilizing allylglycine residues,⁵³ attempts to cyclize arodyn analogs containing Tyr(All) via RCM resulted in mixtures of the desired cyclic peptide plus desallyl side products (Figure 2).^48,49 These side products presumably resulted from isomerization of the double bond to give the vinyl ether with subsequent cleavage to the free phenol.³⁴ The high percentage of the bis-desallyl side product 5 obtained highlights the extent of this side reaction, which limits the utility of RCM for cyclizations involving this residue. In this report, we describe the optimization of the reaction conditions to suppress desallyl product formation and enhance on-resin RCM between Tyr(All) residues.

Figure 1.

Structure of arodyn and Dyn A(1–11), a fragment of the endogenous KOR ligand Dyn A.

Figure 2.

Products observed following RCM involving Tyr(All) of an arodyn analog under the initial reaction conditions. The relative yields, determined by HPLC, are indicated under each of the structures.

In order to enhance the RCM yields of the cyclic arodyn analogs, we used the model dipeptide Fmoc-Tyr(All)-Tyr(All), 7, on resin (Figure 3A) to explore reaction conditions for solid phase cyclization that minimize olefin isomerization and deallylation that could then be used to synthesize arodyn analogs. The possible mono- and bis-desallyl side products (Figure 3B) were synthesized so that the extent of deallylation during RCM could be easily monitored. Here, we present optimization of the cyclization using this model dipeptide and application of these reaction conditions to the synthesis of a cyclic arodyn analog.

Figure 3.

A. Model dipeptide RCM reaction showing the desired RCM product 6 and starting material (7) following cleavage from the solid support, and B. the desallyl side products. The relative yields of each of the products, determined by HPLC, obtained following reaction under the same conditions as used for the arodyn analog (Figure 2) are indicated under the structures.

RESULTS AND DISCUSSION

The initial RCM reaction between Tyr(All) residues in the model dipeptide under the same reaction conditions used for arodyn analogs⁴⁸ (3 mM Grubbs II (G II) catalyst, 40 mol%, DCM/DMF(4/1), 60 °C) resulted in a low yield of the cyclic peptide (20%), as expected, due in part to formation of desallyl products (>35%, Figure 3). Higher catalyst loading can increase catalyst degradation products and result in isomerization,⁵⁴ so we explored the use of a lower catalyst loading (15 mol%) as well as a lower temperature. The inclusion of reported isomerization suppressants benzoquinone²⁷ and phenol⁴⁰ was also examined.

Reaction temperature had a significant effect on the RCM product yield (Table 1 and Figure 4). High yields of desallyl products were observed at 60 °C for all of the 2^nd generation Grubbs catalyst (G II) concentrations examined, which is consistent with catalyst degradation at elevated temperatures. The desired RCM product was only observed in appreciable yields at 40 °C, with the highest yields at 1 and 3 mM catalyst. Lowering the catalyst concentration further (0.3 mM) decreased the RCM product yield.

Table 1.

Effect of temperature and catalyst concentration on model dipeptide RCM product yields.^a

Entry	Temp. (°C)	[Cat.] (mM)	Yield (%)b
			RCM pdt.	Total desallyl pdt.	SMd
1	60	3	-c	82	-
2	60	1	-	39b	-
3	60	0.3	-	78	-
4	40	3	63	16	-
5	40	1	55	13	-
6	40	0.3	24	10	60
7	40	0.1	-	18	69

^aReaction conditions: G II 15 mol%, DCE (60 °C) or DCM (40 °C), 2 d;

^badditional side product peaks were observed in the product mixture (see Supporting Information);

^c- < 10%;

^dSM= Starting material, Fmoc-Tyr(All)-Tyr(All) (7)

Figure 4.

HPLC chromatograms of the RCM product mixtures following RCM of the model dipeptide at a) 60 °C and b) 40 °C with 3 mM catalyst concentration (Table 1, entries 1 and 4, respectively). Structures of 6–10 are shown in Figure 3.

Since catalyst degradation products such as ruthenium hydrides have been implicated in olefin isomerization during olefin metathesis,²⁷ we subsequently explored the use of phenol and 1,4-benzoquinone, additives that are reported to effectively suppress olefin isomerization during olefin metathesis reactions,^27,40 to potentially limit catalyst degradation and thereby enhance cyclic product yield. Higher RCM product yields (31–79%, Table 2 and Figure 5) were obtained at the lower temperature (40 °C) than at 60 °C (<15%) in the presence of phenol. Thus, the additive phenol was ineffective at suppressing this side reaction at the higher temperature with G II as the catalyst. This is consistent with previous reports of increased catalyst degradation at similar higher temperatures giving rise to isomerized products.⁵⁵

Table 2.

Effect of addition of phenol (1 equiv) and catalyst concentration on model dipeptide RCM product yields.^a,b

Entry	[Cat.] (mM)	Yield (%)c
		RCM pdt.	Total desallyl pdt.	SM
1	3	42	20	-d
2	1	60	-	-
3	0.3	79	-	-
4	0.1	31	15	43

^aReaction conditions: 15 mol% G II, DCM. 40 °C, 2 d with 1 equiv phenol;

^bat 60 °C in DCE phenol did not effectively suppress desallyl side product formation: RCM product (<15%), desallyl products (25–55%), and starting material (23–39%) were observed;

^cadditional side product peaks were observed in the product mixture (see Supporting Information);

^d- < 10%

Figure 5.

HPLC chromatograms of the RCM product mixtures of the model dipeptide following RCM (G II, 15 mol%, DCM, 40 °C, 2 d) in the presence or absence of the additives 1,4-benzoquinone and phenol (1 equiv each). Structures of 6–10 are shown in Figure 3.

With 1,4-benzoquinone as the additive negligible quantities of desallyl products were observed (data not shown), but a low RCM product yield was observed at both 40 and 60 °C, with starting material predominating at both 40 °C (67%, Figure 5) and 60 °C (76%), indicating poor conversion of starting material to the RCM product. Thus, 1,4-benzoquinone was effective at suppressing isomerization, but adversely affected conversion of reactant to the desired cyclic product. This is consistent with previous results where 1,4-benzoquinone suppressed both catalyst activity and olefin isomerization.³³ The marked differences observed here suggest different mechanisms for the two additives.

Given the poor conversion to the RCM product at elevated temperatures we explored the use of Hoveyda-Grubbs II (HG II) that is reported to have higher thermal stability.¹¹ Notably, HG II has an isopropoxy ether chelated to the metal center while G II has a coordinated phosphine ligand (Figure 6). We explored the performance of HG II at both 40 and 60 °C and the effect of adding phenol (Table 3). Whereas a temperature dependent effect on catalyst activity was observed for G II (<10% RCM product at 60 °C, Table 1), HG II exhibited comparable catalytic activity at both temperatures examined in the absence of an additive (80–86% RCM product, Table 3). At the higher temperature and in the absence of phenol, HG II provided the RCM product in high yield (86%, Table 3 and Figure 7) in contrast to the negligible desired product observed with G II (Table 1); this is consistent with the greater thermal stability of HG II. The presence of phenol at 60 °C, however, interfered with product formation, possibly due to competing chelating effects of phenol and the isopropyl ether ligand at the elevated temperature, which in turn could affect thermal stability and catalyst degradation. At 40 °C, on the other hand, a similar RCM product yield was observed with HG II and phenol as was previously observed for G II under similar reaction conditions (Table 2, entry 3). Increased chelation of phenol at the higher temperature could explain the difference in RCM product yield with HG II at the two temperatures. In contrast to G II, phenol did not increase the RCM product obtained with HG II.

Figure 6.

Structures of Grubbs 2^nd generation (G II) and Hoveyda-Grubbs 2^nd generation (HG II) catalysts.

Table 3.

Model dipeptide RCM product yields with HG II at 40 and 60 °C.^a

Entry	Temp. (°C)	Additive (1 equiv)	RCM pdt. (%)	Total desallyl pdt. (%)	S.M (%)
1	60	-	86	-	-b
2c	60	phenol	17	32	-
3	40	-	80	-	-
4	40	phenol	72	15	-

^aReaction conditions: 0.3 mM, 15 mol% HG II, DCE (60 °C) or DCM (40 °C), 2 d;

^b- < 10%;

^cadditional side product peaks were observed in the product mixture (see Supporting Information)

Figure 7.

HPLC chromatogram of the RCM product mixture following RCM (HG II, 15 mol%, DCM, 60 °C, 2 d) of the model dipeptide.

We subsequently explored the utility of the optimized conditions for cyclization of a tripeptide with non-adjacent Tyr(All) residues in positions 1 and 3 (Figure 8). Reasonably high yields (65–84%) of the tripeptide RCM product were observed at 40 °C with both G II and HG II while much lower yields were observed with HG II at 60 °C (Table 4). The higher yields observed at the lower temperature are consistent with the model dipeptide results.

Figure 8.

RCM of the model tripeptide Fmoc-Tyr(All)-Phe-Tyr(All) (11) to give the cyclic product 12.

Table 4.

Tripeptide RCM product yields under optimized conditions with G II and HG II.^a

Entry	Catalyst	Temp.(°C)	Additive (1 equiv)	RCM pdt. (%).	Total desallyl pdt. (%)
1b	HG II	60	-	27	28
2	HG II	40	-	84	-c
3	HG II	40	phenol	67	-
4	G II	40	phenol	65	-

^aReaction conditions: 0.3 mM, 15 mol% G II or HG II, DCE (60 °C) or DCM (40 °C), 2 d;

^badditional side product peaks were observed in the product mixture (see Supporting Information);

^c- < 10%

We then examined these conditions for the RCM of [Tyr(All)^2,3]arodyn (Table 5 and Figure 9). The yield observed with G II in the presence of phenol was comparable to the yield observed with HG II. As was found in both the dipeptide and the tripeptide, phenol appeared to decrease the yield with HG II. Overall, substantially higher yields of the desired RCM product were observed (63–70%) with the optimized conditions compared to the initial conditions^48,53 (32%, Figure 9).

Table 5.

RCM of [Tyr(All)^2,3]arodyn using optimized conditions with G II and HG II catalysts at 40 °C.^a

Entry	Catalyst	Additive	RCM pdt. (%)	Total desallyl pdt. (%)
1	HG IIb	-	63	17
2	G II	phenol	70	14

^aReaction conditions: 0.3 mM, 15 mol% G II or HG II, DCM, 2 d.

^bHG II plus phenol resulted in 51% RCM product and 11% total desallyl products.

Figure 9.

HPLC chromatograms of the RCM product mixtures from [Tyr(All)^2,3]arodyn (2) under various conditions showing the formation of the desired RCM product (1) and the bis-desallyl side product (5): A. Reactions with G II under initial conditions (3 mM G II catalyst, 40 mol%, DCM/DMF(4/1), 60 °C, 2 d) and optimized conditions (0.3 mM G II catalyst, 15 mol%, DCM, 40 °C, 2 d) B. Reactions with HG II (0.3 mM, 15 mol%, DCM, 40 °C) without and with phenol.

The RCM reaction in the arodyn analog was performed as the last step in the synthesis prior to cleavage from the solid support (Figure 1). To evaluate whether the N-terminal functionality affected cyclization yields RCM was performed on the FmoC-protected arodyn analog prior to N-terminal deprotection, acetylation and cleavage from the solid support. The yields observed with Fmoc as the N-terminal functionality in the arodyn analog appeared to be slightly (11–13%) higher when RCM was performed with HG II (76%) and G II with phenol (81%; see Supporting Information), possibly due to differences in the preference for conformation(s) of the peptide compatible with cyclization.

CONCLUSIONS

Several important modifications that improved the yields of the desired RCM product involving Tyr(All) were identified in the model dipeptide study. Key findings were the effect of temperature and the catalyst thermal stability on the yield of the desired RCM product, the temperature dependent effectiveness of phenol with the G II catalyst, and the efficiency of phenol, in contrast to 1,4-benzoquinone, to suppress deallylation in reactions utilizing the G II catalyst at 40 °C. We have identified optimized conditions with the high yields of RCM product (80%) obtained at 40 °C using either G II plus phenol or HG II without phenol. Of note, phenol enhanced the yield of the desired RCM product when G II was the catalyst, but negatively affected the yield when HG II was used. High yields of the tripeptide were also obtained at 40 °C, and the optimized conditions were effective in improving the RCM product yield from [Tyr(All)^2,3]arodyn. The synthesis of a series of RCM cyclic arodyn analogs containing Tyr(All) for pharmacological evaluation has been undertaken using this methodology, the results of which will be reported elsewhere (manuscript in preparation).

The methodology presented here could be beneficial in other RCM reactions where olefin isomerization results in side products that decrease the yields of the desired RCM product and complicate purification, particularly for peptidic substrates. For instance, macrocycles have been incorporated into antibiotics⁵⁶ and protease inhibitors^57,58 via RCM involving O-allyl protected aromatic residues, demonstrating the potential broad applicability of this methodology in drug discovery.

EXPERIMENTAL SECTION

Materials.

The Rink amide ChemMatrix resin was purchased from Biotage (Charlotte, NC), Fmoc-protected PAL-PEG-PS was purchased from APPTec LLC (Louisville, KY), and the 2-chlorotrityl chloride resin was purchased from Chem-Impex International (Wood Dale, IL). All standard protected amino acids were purchased from Bachem (King of Prussia, PA), EMD Millipore Chemicals (San Diego, CA), Peptides International (Louisville, KY), or Chem-Impex International, Fmoc-L-Tyr(All)-OH and Fmoc-L-Tyr-OH were obtained from Chem-Impex International. The coupling agent benzotriazol-1-yl-oxy-tris-pyrrolidino-phosphonium hexafluorophosphate (PyBOP) and 1-hydroxybenzotriazole hydrate (HOBt) were obtained from Peptides International. Second-generation Grubbs’ catalyst (G II) and second-generation Hoveyda Grubbs’ catalyst (HG II) were purchased from Aldrich Chemical Co. (Milwaukee, WI). 4-Methylpiperidine was purchased from Acros Organics (Morris, NJ). NMP, DCM, N,N-diisopropylethylamine (DIEA), DMF, diethyl ether, acetonitrile, methanol, and TFA were purchased from Fisher Scientific (Hampton, NH). All other chemicals were purchased from Aldrich Chemical Co.

Instruments and analysis.

Microwave-assisted peptide synthesis was performed on a Biotage microwave peptide synthesizer (Initiator+ Alstra; Biotage, Sweden).

Analytical HPLC was performed on an Agilent 1200 or 1260 system fitted with a Grace Vydac analytical column (C18, 300 Å, 5 μm, 4.6 mm × 50 mm) equipped with a Vydac C18 guard cartridge at a flow rate of 1 mL/min and monitoring at 214 nm. Analysis was performed using linear gradients of 30–70% aqueous MeCN with 0.1% TFA over 40 min for the dipeptides 6–10 and tripeptides 11 and 12 and 5–50% aqueous MeCN with 0.1% TFA over 45 min for the arodyn analogs 1–5. Preparative HPLC purification was performed on a Vydac C18 column (10 μm, 300 Å, 22 × 250 mm) equipped with a Vydac C18 guard cartridge on an LC-AD Shimadzu liquid chromatography system with monitoring at 214 nm. The cyclic di-and tripeptide RCM products were purified using a linear gradient of 30–70% aqueous MeCN (containing 0.1% TFA) over 40 min, and the [Tyr(All)^2,3]arodyn RCM product was purified using a linear gradient of 20–35% aqueous MeCN (containing 0.1% TFA) over 60 min at a flow rate of 20 mL/min. Electrospray ionization mass spectrometry (ESI/MS) was performed on an Agilent 6230 instrument with a time of flight mass analyzer or an Advion expression L compact MS (Advion, Inc. Ithaca, NY).

¹H and ¹³C NMR spectra were obtained at 25 °C in DMSO-d₆ on a Bruker AVANCE Neo-600 MHz spectrometer equipped with a 5 mm broad band Prodigy cryoprobe operating with Top Spin-4 software and standard pulse sequences. ¹H and ¹³C chemical shifts were referenced to the residual DMSO signals at 2.5 and 39.52 ppm, respectively, and coupling constants were extracted from the 1D spectra.

Solid phase peptide synthesis (SPPS) on the 2-chlorotrityl chloride resin.

Peptides were prepared by the Fmoc (9-fluorenylmethoxycarbonyl) solid-phase synthetic strategy. The linear di-and tri-peptides were synthesized on the 2-chlorotrityl resin following the general method reported previously^59,60 using a custom-made manual peptide synthesizer (CHOIR).⁶¹ Briefly, after swelling the resin in DCM (2×10 min), the Fmoc-protected C-terminal amino acid (2 equiv in DCM/DMF, 4:1) and DIEA (5 equiv) were added to the resin, and the reaction mixture was agitated with nitrogen gas for 6 h with addition of DCM every 30 min to maintain solvent volume; DIEA (5 equiv) was added every 2 h. MeOH (15%) and DIEA (5%) in DCM (2×10 min) were then used to cap unreacted sites on the resin, after which the resin was washed with DCM/DMF (1:1, 5x). After Fmoc deprotection (20% 4-methylpiperidine, 1 × 5 min, 2 × 20 min), the resin was washed with DMF (5x), DCM/DMF (1:1, 5x) and DCM (5x).

Cycles of Fmoc deprotection and amino acid couplings (2 equiv in DCM/DMF, 4:1) were performed to add amino acids to the growing peptide chain. Amino acid couplings were performed using PyBOP (2 equiv), HOBt (2 equiv), and DIEA (5 equiv) in DCM/DMF (1:1) for 2–4 h at room temperature. Nitrogen gas was used to agitate the mixture during coupling. To cleave the peptide from the resin, 1% TFA in DCM was bubbled through the resin (5 mL × 10, 2 min each), then drained into a round bottom flask and evaporated. The peptide was then lyophilized from MeCN:water (1:1).

SPPS of [Tyr(All)^2,3]arodyn (2) and desallyl biproducts 3–5.

[Tyr(All)^2,3]arodyn and its desallyl analogs was prepared by Fmoc SPPS using the automated Biotage microwave peptide synthesizer (Initiator+ Alstra; Biotage, Sweden) on a 0.14 mmol scale. Rink amide ChemMatrix (0.45–0.52 mmol/g) or Fmoc-PAL-PEG-PS resin (0.18–0.20 mmol/g) resin was swollen for 20 min at 70° C in DMF and the Fmoc group removed prior to cycles of amino acid coupling and deprotection. Coupling reactions were performed using Fmoc amino acids (4 equiv, 0.5 M), activated with PyBOP (4 equiv, 0.5 M) and HOBt (4 equiv, 0.5 M) in the presence of DIEA (8 equiv in NMP, 0.5 M), for 5 min at 75 °C. Fmoc deprotection was performed at room temperature with 20% 4-methylpiperidine in DMF (4.5 mL, 3 min, then, 2 × 10 min). The side chains of Lys and Arg were protected with tert-butoxycarbonyl (Boc) and 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl (Pbf), respectively. N-Terminal acetylation was performed using acetic anhydride (5 M in DMF) and DIEA (2 M in NMP) for 10 min at room temperature. The resulting peptide-resin was washed with DMF (5 mL, 3 × 45 sec) and DCM (5 mL, 3 × 45 sec). The crude peptide was cleaved from the resin using TFA plus scavengers (TFA/triisopropylsilane (TIPS)/H₂O (95/2.5/2.5) for the Rink amide ChemMatrix resin and TFA/phenol/H₂O/TIPS (88/5/5/2) for Fmoc-PAL-PEG-PS resin) for at least 2 h and then precipitated in cold ether following filtration; the peptide was then lyophilized from MeCN:water (1:9). When an aliquot was cleaved for reaction monitoring, the solution was diluted with 10% aqueous acetic acid (3–5 mL) and then extracted with diethyl ether (3 × 5 mL) following filtration. The aqueous solution was lyophilized to give the crude peptide.

General procedure for RCM.

Second-generation Grubbs’ catalyst or second-generation Hoveyda Grubbs’ catalyst was dissolved in DCM and added to the resin-bound peptide: Fmoc-Tyr(All)-Tyr(All)-2-chlorotrityl resin (0.56 mmol/g), Fmoc-Tyr(All)-Phe-Tyr(All)-2-chlorotrityl resin (0.51 mmol/g) or [Tyr(All)^2,3]arodyn (0.13 mmol/g). Catalyst loading was calculated relative to the resin-bound peptide loading. Additives (phenol or 1,4-benzoquinone) were added to the reaction mixture along with the catalyst, and the reaction mixture was heated as indicated for the respective reaction conditions for 2 days under nitrogen with gentle stirring. The solvent volume was varied according to the quantity of resin used in order to give the desired catalyst concentration for the reaction. After the specified reaction time, the resin was transferred to a 15 mL polypropylene syringe fitted with a polytetrafluoroethylene (PTFE) frit and washed with MeOH (3 × 5 mL) and DCM (10 × 5 mL), with N₂ agitation, to remove the catalyst. Following cleavage of an aliquot of the dried resin, analytical HPLC was performed as described above. Relative yields for the respective reactions were determined from area under the curve in the analytical HPLC chromatograms (see Supporting Information); in some cases, baseline resolution was not achieved so yields shown are estimates.

cyclo[Tyr(All)^2,3]arodyn (1).

Resin-bound 2 was subjected to RCM as described in the general RCM procedure and purified by preparative HPLC to give peptide 1 as a white lyophilized solid, R_t = 29.4 min (98.8% pure). (ESI) m/z: [M + 3H]³⁺ Calcd for C₇₇H₁₁₉N₂₅O₁₄ 540.3; Found 540.3.

[Tyr(All)^2,3]arodyn (2).

Peptide 2 was synthesized according to the general procedure for SPSS of [Tyr(All)^2,3]arodyn (2) and its analogs above, and the crude peptide obtained as a white lyophilized solid, R_t = 32.9 min (90.4% pure). (ESI) m/z: [M + 3H]³⁺ Calcd for C₇₉H₁₂₃N₂₅O₁₄ 549.7; Found 549.2.

[Tyr(All)²,Tyr³]arodyn (3).

Peptide 3 was synthesized according to the general procedure for [Tyr(All)^2,3]arodyn (2) and it analogs with the side chain of Tyr³ tert-butyl protected. The crude peptide was obtained as a white lyophilized solid, R_t = 28.6 min (90.0%, pure). (ESI) m/z: [M + 3H]³⁺ Calcd for C₇₆H₁₁₉N₂₅O₁₄ 536.7; Found 536.7.

[Tyr²,Tyr(All)³]arodyn (4).

Peptide 4 was synthesized according to the general procedure for SPSS of [Tyr(All)^2,3]arodyn (2) and its analogs above with the side chain of Tyr² tert-butyl protected. The crude peptide was obtained as a white lyophilized solid, R_t = 27.4 min (90.4%, pure). (ESI) m/z: [M + 3H]³⁺ Calcd for C₇₆H₁₁₉N₂₅O₁₄ 536.7; Found 536.6.

[Tyr^2,3]arodyn (5).

Peptide 5 was synthesized according to the general procedure for SPSS of [Tyr(All)^2,3]arodyn (2) and its analogs above with the side chains of both Tyr residues tert-butyl protected. The crude peptide was obtained as a white lyophilized solid, R_t = 22.9 min (92.1%, pure). (ESI) m/z: [M + 3H]³⁺ Calcd for C₇₃H₁₁₅N₂₅O₁₄ 523.3; Found 523.4.

Fmoc-cyclo[Tyr(CH₂CH=CHCH₂)Tyr]-OH (6).

Resin-bound 7 was subjected to RCM as described in the general RCM procedure and purified by preparative HPLC to give peptide 6 as a white lyophilized solid, R_t = 17.8 min (99.0% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.72 (br s, 1H), 7.92 (d, J = 7.6 Hz, 2H), 7.72 (m, 2H), 7.63 (d, J = 7.8 Hz, 1H), 7.43 (t, J = 7.4 Hz, 2H), 7.35 (t, J = 7.4, 2H), 6.96 (d, J = 8.6 Hz, 2H), 6.90 (d, J = 9.1 Hz, 1H), 6.73 (d, J = 8.6 Hz, 2H), 6.69 (m, 2H), 6.62 (d, J = 8.5 Hz, 2H), 5.88 (dt, J = 16.2, 4.7 Hz, 1H), 5.83 (dt, J = 16.2, 4.9 Hz, 1H), 4.66 (m, 4H), 4.54 (m, 1H), 4.39 (m, 1H), 4.35 (m, 1H), 4.26 (m, 2H), 3.01 (m, 1H), 2.90 (m, 2H), 2.66 (m, 1H). HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₃₇H₃₄N₂O₇ 619.2439; Found 619.2426.

Fmoc-Tyr(All)-Tyr(All)-OH (7).

Peptide 7 was synthesized according to the general procedure for SPSS on the 2-chlorotrityl chloride resin, and the crude peptide obtained as a white lyophilized solid, R_t = 22.5 min (98.9% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.76 (br s, 1H), 8.20 (d, J = 7.8 Hz, 1H), 7.87 (d, J = 7.6 Hz, 2H), 7.63 (m, 2H), 7.55 (d, J = 9.0 Hz, 1H), 7.40 (m, 2H), 7.29 (m, 2H), 7.20 (m, 2H), 7.14 (m, 2H), 6.81 (m, 4H), 5.98 (m, 2H), 5.35 (dd, J = 6.0, 1.8 Hz, 1H), 5.32 (dd, J = 6.0, 1.8 Hz, 1H), 5.20 (dt, J = 10.5, 1.7 Hz, 2H), 4.45 (m, 5H), 4.23 (m, 1H), 4.13 (m, 3H), 3.01 (dd, J = 14.0, 5.3 Hz, 1H), 2.91 (m, 2H), 2.67 (m, 1H); ¹³C NMR (151 MHz, DMSO-d₆) δ 172.82, 171.65, 156.87, 156.68, 155.70, 143.78, 143.72, 140.66, 140.65, 133.85, 133.78, 130.25 (2), 130.17 (2), 130.13, 129.32, 127.63, 127.60, 127.05 (2), 125.36, 125.30, 120.07 (2), 117.24 (2), 114.35 (2), 114.19 (2), 68.06, 68.05, 65.69, 56.18, 53.64, 46.55, 36.59, 35.88; HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₃₉H₃₈N₂O₇ 647.2752; Found 647.2728.

Fmoc-Tyr(All)-Tyr-OH (8).

Peptide 8 was synthesized according to the general procedure for SPSS on the 2-chlorotrityl chloride resin, and the crude peptide obtained as a white lyophilized solid, R_t = 14.4 min (95.1% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.72 (br s, 1H), 9.21 (s, 1H), 7.87 (m, 2H), 7.63 (m, 3H), 7.55 (m, 1H), 7.40 (m, 3H), 7.28 (m, 2H), 7.20 (m, 2H), 7.04 (m, 1H), 6.87 (m, 1H), 6.81 (d, J = 8.6 Hz, 2H), 6.67 (d, J = 8.4 Hz, 1H), 5.99 (m, 1H), 5.34 (m, 1H), 5.20 (m, 1H), 4.46 (m, 3H), 4.27 (m, 1H), 4.14 (m, 3H), 3.10 (m, 1H), 2.96 (m, 2H), 2.67 (m, 1H); ¹³C NMR (151 MHz, DMSO) δ 172.91, 171.65, 156.70, 156.00, 155.72, 143.82, 143.73, 140.74, 140.66, 133.86, 130.31, 130.26, 130.16 (2), 130.12 (2), 127.63 (2), 127.32, 127.07, 125.33, 125.20, 121.17, 120.07, 117.25, 115.03 (2), 114.49, 114.20, 68.08, 65.68, 56.24, 53.82, 46.59, 36.65, 35.99; HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₃₆H₃₄N₂O₇ 607.2439; Found 607.2415.

Fmoc-Tyr-Tyr(All)-OH (9).

Peptide 9 was synthesized according to the general procedure for SPSS on the 2-chlorotrityl chloride resin, and the crude peptide obtained as a white lyophilized solid, R_t = 14.2 min (97.4% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.76 (br s, 1H), 9.21 (m, 1H), 8.17 (d, J = 7.7 Hz, 1H), 7.87 (m, 2H), 7.65 (m, 3H), 7.51 (d, J = 8.9 Hz, 1H), 7.40 (m, 2H), 7.30 (m, 3H), 7.14 (m, 2H), 7.08 (m, 1H), 6.81 (d, J = 8.2 Hz, 2H), 6.64 (d, J = 8.1 Hz, 1H), 5.97 (m, 1H), 5.33 (m, 1H), 5.19 (m, 1H), 4.44 (m, 3H), 4.31 (m, 1H), 4.15 (m, 3H), 3.01 (m, 1H), 2.88 (m, 2H), 2.63 (m, 1H); ¹³C NMR (151 MHz, DMSO) δ 172.83, 171.73, 156.87, 155.74, 155.68, 143.82, 143.70, 140.73, 140.65, 133.78, 130.18 (2), 130.16 (2), 129.33, 128.17, 127.64 (2), 127.08 (2), 125.36, 125.30, 120.08 (2), 117.25, 114.83 (2), 114.35 (2), 68.05, 65.68, 56.32, 53.64, 46.56, 36.63, 35.88; HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₃₆H₃₄N₂O₇ 607.2439; Found 607.2429.

Fmoc-Tyr-Tyr-OH (10).

Peptide 10 was synthesized according to the general procedure for SPSS on the 2-chlorotrityl chloride resin, and the crude peptide obtained as a white lyophilized solid, R_t = 6.3 min (91.9% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.69 (br s, 1H), 9.20 (s, 2H), 8.13 (d, 1H), 7.86 (m, 3H), 7.63 (m, 3H), 7.50 (d, 1H), 7.40 (m, 3H), 7.30 (m, 2H), 7.07 (d, 1H), 7.02 (d, 1H), 6.64 (m, 3H), 4.38 (m, 1H), 4.29 (m, 1H), 4.14 (m, 3H), 2.95 (m, 1H), 2.83 (m, 2H), 2.61 (m, 1H); ¹³C NMR (151 MHz, DMSO) δ 172.92, 171.73, 155.97, 155.74, 155.70, 143.85, 143.70, 140.74, 140.67, 130.18 (2), 130.12 (2), 128.20 (2), 127.66 (2), 127.33, 127.10, 125.37, 125.32, 120.10 (2), 115.02 (2), 114.84 (2), 65.67, 56.37, 53.81, 46.59, 36.67, 35.97; HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₃₃H₃₀N₂O₇ 567.2126; Found 567.2112.

Fmoc-Tyr(All)-Phe-Tyr(All)-OH (11).

Peptide 11 was synthesized according to the general procedure for SPSS on the 2-chlorotrityl chloride resin, and the crude peptide obtained as a white lyophilized solid, R_t = 25.6 min (99.0% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.76 (br s, 1H), 8.29 (d, J = 7.7 Hz, 1H), 8.06 (d, J = 8.3 Hz, 1H), 7.87 (d, J = 7.6 Hz, 2H), 7.61 (m, 2H), 7.49 (d, J = 8.9 Hz, 1H), 7.40 (td, J = 7.5, 3.4 Hz, 2H), 7.29 (m, 2H), 7.22 (m, 4H), 7.13 (m, 5H), 6.83 (m, 2H), 6.78 (m, 2H), 5.98 (m, 2H), 5.34 (m, 1H), 5.32 (m, 1H), 5.20 (m, 2H), 4.59 (m, 1H), 4.45 (m, 5H), 4.16 (m, 2H), 4.10 (m, 2H), 3.02 (m, 2H), 2.87 (m, 1H), 2.80 (m, 2H), 2.61 (m, 1H); ¹³C NMR (151 MHz, DMSO-d6) δ 172.72, 171.31, 170.90, 156.87, 156.63, 155.64, 143.76, 143.71, 140.65, 140.63, 137.51, 133.83, 133.79, 130.18 (2), 130.12 (2), 129.31 (2), 129.28 (2), 127.97 (2), 127.59 (2), 127.05 (2), 126.21, 125.34, 125.27, 120.06 (2), 117.25, 117.22, 114.36 (2), 114.15 (2), 68.04 (2), 65.66, 56.27, 53.67, 53.51, 46.54, 37.62, 36.68, 35.89.HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₄₈H₄₇N₃O₈ 794.3436; Found 794.3422.

Fmoc-cyclo[Tyr¹(CH₂CH=CHCH₂)Tyr³,Phe²]-OH (12).

Resin-bound 11 was subjected to RCM as described in the general RCM procedure and purified by preparative HPLC to give peptide 12 as a white lyophilized solid, R_t = 21.1 min (99.0% pure). ¹H NMR (600 MHz, DMSO-d₆) δ 12.88 (br s, 1H), 8.56 (d, J = 8.3 Hz, 1H), 8.12 (d, J = 8.6 Hz, 1H), 7.90 (m, 2H), 7.69 (d, J = 6.4 Hz, 2H), 7.42 (m, 2H), 7.32 (m, 2H), 7.23 (d, J = 7.0 Hz, 2H), 7.17 (m, 4H), 7.12 (m, 1H), 6.76 (d, J = 8.7 Hz, 2H), 6.70 (d, J = 8.3 Hz, 2H), 6.50 (d, J = 8.5 Hz, 2H), 6.28 (d, J = 7.6 Hz, 1H), 5.87 (dt, J = 15.9, 4.5 Hz, 1H), 5.70 (dt, J = 15.9, 5.3 Hz, 1H), 4.66 (m, 2H), 4.61 (m, 1H), 4.58 (m, 2H), 4.51 (m, 1H), 4.44 (m, 1H), 4.26 (m, 1H), 4.20 (m, 2H), 3.11 (m, 2H), 2.89 (m, 1H), 2.81 (m, 2H), 2.74 (m, 1H); HRMS (ESI-QToF) m/z: [M + H]⁺ Calcd for C₄₆H₄₃N₃O₈ 766.3123; Found 766.3119.