Insights on Structure–Passive Permeability Relationship in Pyrrole and Furan-Containing Macrocycles

Abstract

Macrocycles have recognized therapeutic potential, but their limited cellular permeability can hinder their development as oral drugs. To better understand the structure–permeability relationship of heterocycle-containing, semipeptidic macrocycles, a library was synthesized. These compounds were created by developing two novel reactions described herein: the reduction of activated oximes by LiBH₄ and the aqueous reductive mono-N-alkylation of aldehydes using catalytic SmI₂ and stoichiometric Zn. The permeability of the macrocycles was evaluated through a parallel artificial membrane permeability assay (PAMPA), and the results indicated that macrocycles with a furan incorporated into the structure have better passive permeability than those with a pyrrole moiety. Compounds bearing a 2,5-disubstituted pyrrole (endo orientation) were shown to be implicated in intramolecular H-bonds, enhancing their permeability. This study highlighted the impact of heterocycles moieties in semipeptides, creating highly permeable macrocycles, thus showing promising avenues for passive diffusion of drugs beyond the rule-of-five chemical space.

Introduction

Macrocycles are attractive scaffolds in drug discovery.¹⁻³ In nature, they often occur as macrocycles of peptides and nonpeptidic complex molecules. Semipeptidic macrocycles display a variety of biological activities, such as antibacterial, antifungal, anticancer, and antiviral activity.⁴ Moreover, peptide macrocyclization is a well-established tool to modulate binding affinity,⁵ selectivity,⁶ signaling profile,^7,8 and PK-ADME properties^9,10 of pharmacologically relevant peptidic ligands. Some synthetic macrocycles have even been used to target previously thought “undruggable” protein–protein interactions^11,12 as these peptides are often considered to fit in a chemical space in-between the biologics and small molecules.³ However, many macrocyclic peptides display highly hydrophilic surfaces and high molecular weight (MW), confering modest-to-low oral bioavailability.^13,14

Lessons from cyclosporin A, a complex natural cyclic undecapeptide, have been an inspiring thread for medicinal chemists in developing macrocyclic drugs, particularly peptidic macrocycles beyond the rule of 5 (bRo5).¹⁵⁻¹⁷ Its outstanding permeability might be attributed to its “chameleonic” properties.¹⁸⁻²⁰ Chameleonism is defined as the ability of a molecule to change conformation from hydrophilic in water to lipophilic while crossing a membrane.²¹ Unfortunately, most macrocyclic peptides do not possess the general physicochemical attributes required for pharmaceuticals, regardless of their fascinating and attractive biological activities.^1,2,17−20 Currently, a study from our group has shown some effects of N-methylation on the increased permeability of macrocyclic tetrapeptides.^22,23 Nevertheless, the challenge in optimizing drug-like properties has led to the need for more structure–permeability relationship studies.

Aromatic heterocycles are often considered as an interesting moiety not only for their prevalence in nature²⁴ but also because of their potential to modulate physicochemical properties, biological activities, and conformations.²⁵⁻²⁹ The impact of a heterocycle moiety in a macrocycle’s skeletal flexibility may help chameleonism.³⁰ In macrocyclic peptides, replacing amino acids with heterocycles not only diversifies the structure but may also lead to increased membrane permeability. Simply by their lipophilic nature, heterocycles can affect the polarity and network of intramolecular hydrogen bonds (IMHBs), which can enhance biological activity.^1,31,32 Unlike their linear counterparts, triazoles or thiazoles, for example, often improve enzymatic stability and biological activity.^33,34 Studies on macrocyclic peptides have therefore revealed that heterocyclic scaffolds impart unique potential pharmacological properties, giving rise to more attention from medicinal chemists.

Several review articles have previously discussed pharmacologically relevant macrocycles isolated from nature.^35,36 Interestingly, some of these natural heterocycle-containing macrocycles display a wide range of pharmacologically relevant activities in vivo. For example, Bistratamide D has been shown as a central nervous system depressant,³⁷ Venturamide B has shown interesting antimalarial activity in mice,³⁸ while Nostrocyclamide M, a compound isolated from cyanobacterium, has shown allelopathic properties, which is a property of a chemical to alter the survival of other competing organisms³⁹ (Figure 1). All of these compounds display their effect in distinct pharmacological compartments.

Figure 1.

Structures of several heterocycle-containing peptidic macrocycles and their respective activities.

Synthetic methods for the generation of these peptides are however lacking and are often challenging with only a handful of examples, which have been tackled.^29,40−42 Solubility of these peptides are also often problematic as most naturally produced, cell-penetrant peptides are highly lipophilic.⁴³

Although the synthesis of these heterocycle-containing macrocycles is difficult, nature produces them readily, possibly for harnessing their high passive permeation of cellular membranes^3,44 coupled to large interaction surface area that can inhibit protein–protein interactions.⁴⁵ Their application in the field of previously thought “non-druggable“ targets is therefore of rising interest.^46,47 However, investigations in the structure–permeability relationship of heterocycle-containing macrocyclic peptides are needed to optimize both biological activities and clinical uses and will be tackled in this present work.

As a starting point, we decided to investigate the inclusion of N–H-bearing heterocycles, which were previously suggested modulate the permeability based on changes to the IMHBs.^29,48 With these previous studies in mind, we have designed a prototype of semipeptidic macrocycles in which we hypothesize that heterocycles can be used to control some of these IMHBs. Macrocycles in this study consist of a tripeptide residue linked head-to-tail via a pyrrole or furan heterocyclic linker (Scheme 1).

Scheme 1. General Approach to the Synthesis of Macrocycles.

According to previous studies, crystal structures of macrocyclic peptides containing imidazole,⁴⁹ thiophene,⁵⁰ thioxazole,⁵¹ and oxazoline⁵² have revealed that heterocycles with a 2,5-substitution pattern usually have one of their heteroatoms oriented inside the macrocycle. In contrast, heterocycles with a 2,4-substitution pattern usually have one of their heteroatoms oriented outside the macrocycle.⁴⁹⁻⁵² Therefore, we hypothesized that a pyrrole heterocycle inserted in a 2,5- or 2,4-substitution could modulate the pharmacokinetics of permeation (Scheme 1). This characteristic could be hypothesized to happen because of the inclusion of an H-bond donor, forming an IMHB with a facing carbonyl H-bond acceptor. As a control, furan rings as well as N-methylation of the pyrrole were synthesized for their lack of available hydrogen for H-bond formation.

Linkers were designed and prepared using an adaptation of known synthetic methods,⁵³⁻⁵⁵ starting from commercially available methyl pyrrole-2-carboxylate (1) and 5-hydroxymethyl furfural (HMF). The chosen linkers for this study contain a hydroxymethyl moiety and protected aminomethyl derivatives of pyrrole and furan that enabled the synthesis of the semipeptidic macrocycles. However, due to the electron-rich pyrrole moiety, the respective pyrrole derivatives are unstable because of the formation of highly reactive azafulvene. The presence of an electron-withdrawing group (EWG) on the pyrrole nitrogen, such as the tert-butoxycarbonyl (Boc) group, is thought to suppress the formation of azafulvenium species,⁵⁶ hence the present synthetic route.

Based on our recent publication, semipeptidic macrocycles containing a tripeptide H₂N-Phe-Ala-Leu-OH showed excellent passive as well as cellular permeability.^23,57 Moreover, a methyl group was added to the backbone nitrogens, either as methylation or by substituting peptides for peptoids.

All compounds were synthesized via a solid-phase peptide strategy using 2-chlorotrityl chloride (2-CTC) resin. The on-resin tripeptides were then conjugated with the linkers using the Mitsunobu–Fukuyama reaction.⁵⁸ The final macrocyclization by DEBPT or PyBOP was carried out in the solution phase after a concomitant cleavage of the linear product from the resin and of protecting groups.

Results and Discussion

Synthesis of the Pyrrole Linkers

The synthesis of pyrrole linkers (Scheme 2A) started with the Vilsmeier–Haack reaction of methyl pyrrole-2-carboxylate 1. The two aldehyde regioisomers (2a and 2b, ratio 5:2) obtained from the reaction were separated and used for the subsequent synthesis of two linkers (5a and 5b, Scheme 3). Individual treatments of 2a and 2b with hydroxylamine hydrochloride in aqueous potassium carbonate solution gave the corresponding oximes (3a and 3b) in good yields.

Scheme 2. Attempts to Prepare Tri-Boc Intermediates 4a and 4b Using Standard Boc-Protective Hydrogenation.

Scheme 3. Preactivation Using Boc₂O Enabled Protective Hydrogenation toward the Synthesis of Two Linkers 5a and 5b.

The oxime reduction and protection of the resulting amine were first attempted under the conditions shown in Scheme 2. Direct hydrogenation with in situ Boc protection was conducted efficiently with isomer 3a (72% yield) but poorly with isomer 3b (55% yield) (Scheme 2B,C). This was rationalized considering the direct conjugation of the oxime with the ester, which decreases electron density, thus facilitating the direct hydrogenation of 3a. Fortunately, this hydrogenation could be improved by preactivating the oxime 3a and 3b using Boc₂O and DMAP, followed by hydrogenation under the same conditions used for 3a (Scheme 3). Applying preactivation and hydrogenation procedures gave intermediates with two Boc protecting groups, which were then converted to the tri-Boc products (4a and 4b) in high yields (90–95%). The treatment of 4a or 4b with DIBAL-H in DCM at room temperature gave the desired product 5a or 5b in 46% yield concomitantly with the removal of a Boc protecting group.

The corresponding N-methyl pyrrole linkers were also prepared from 2a and 2b as starting materials. N-Methylation of the pyrrole ring by dimethyl carbonate (DMC) and DABCO gave products 6a and 6b in high yields. The corresponding oximes 7a and 7b were formed using similar condensation conditions as above. Oxime hydrogenation/Boc protection was successfully performed at room temperature under atmospheric pressure via the same oxime-activation strategy to deliver monoprotected amines 8a and 8b (Scheme 4). In the last step, a DIBAL-H reduction, a low yield was initially obtained when applying the same conditions used to form 5a and 5b. To explain this difference in reactivity, it was hypothesized that replacing the Boc protecting group on the pyrrole nitrogen with a methyl group might impact the complexation between reaction intermediates with aluminum from DIBAL-H. In line with the hypothesis above, THF was therefore introduced as an additive to stabilize the intermediate complex. Gratifyingly, reduction using DIBAL-H in DCM with the addition of THF at room temperature provided linkers 9a and 9b in moderate yields.

Scheme 4. Preparation of Two N-Methyl Pyrrole Linkers 9a and 9b.

Synthesis of the Furan Linker

To ascertain the impact of heterocycle identity on permeability, the corresponding furan-containing linkers were synthesized in two steps from HMF. The corresponding oxime 10 was obtained by the same procedure used above. Without the activation of the oxime by Boc₂O/DMAP, no reduction occurred of compound 10. Applying the hydrogenation of the activated oxime using the procedure used for pyrrole 3 was also unsuccessful, giving a complex mixture instead (Scheme 5).

Scheme 5. Attempts to Approach to the Furan Linker 11.

An alternative to the protective reduction was developed to obtain linker 11 using mild reaction conditions. Inspired by a nitrile reduction with lithium borohydride catalyzed by nickel(II) chloride,⁵⁹ optimization of this reduction was performed, resulting in a chemoselective protective reduction using Boc₂O as an activating reagent (Table 1). The oxime was deprotonated by LiBH₄ followed by the activation with Boc₂O to facilitate the reduction only when MeOH was present as a solvent (entries 3 and 4). The results have also shown that additional equivalents of NiCl₂ catalyst do not improve the reaction yield (entry 4). This is notably the first report of oxime reduction using lithium borohydride and Boc₂O.

Table 1. Investigations to Prepare the Furan Linker via Chemoselective Protective Reduction^a.

entry	base (1.2 equiv)	solvents	yield (%)b
1c	DMAP	DCM	0
2	LiBH4	DCM/THF	0
3d	LiBH4	DCM/MeOH	62 (1 g scale)
4e	LiBH4	DCM/MeOH	48

^aReaction scale: oxime 10 (100 mg, 0.7 mmol).

^bIsolated yields after flash column chromatography.

^cBase (0.1 equiv), LiBH₄ (3.0 equiv).

^dReaction scale: oxime 10 (1 g, 7.1 mmol).

^eNiCl₂ (0.3 equiv).

The N-methyl furan linker (13) was similarly synthesized in one pot starting from HMF. Attempts to perform the reductive amination of the aldehyde group were investigated (Table 2).

Table 2. Preparation of the N-Methylfuran Linker via Aqueous Reductive Mono-N-alkylation^a.

entry	Zn (equiv)	SmI2 (equiv)	time (h)	yields (%)b
1	0	2.0	24	11
2	2.0	0	24	0
3	1.2	0.5	3	50
4	0	0.2	3	<5%c
5d	2.0	0.2	4	60

^aReaction scale: 5-hydroxymethylfuran HMF (100 mg, 0.8 mmol).

^bIsolated yields after flash column chromatography.

^c1H NMR yield of 13 using 1,3,5-trimethoxybenzene as an internal standard.

^dReaction scale: 5-hydroxymethylfuran HMF (1 g, 7.9 mmol).

To our delight, the reductive amination was successfully conducted with a combination of samarium(II) iodide and zinc as reducing agents, giving solely the N-methylamino intermediate (12), followed by Boc protection in one pot. Stoichiometric or catalytic reactions with SmI₂ proceeded in low yield (Table 2, entries 1 and 4), and that zinc is not a suitable reductant on its own (entry 2). The reaction using an excess of Zn proved reliable with catalytic SmI₂ to proceed well on a small scale (entry 3) and on a gram scale (entry 5). To our knowledge, this is the first time a stoichiometric amount of Zn and a catalytic amount of SmI₂ in aqueous reductive mono-N-alkylation was reported.

Fukuyama–Mitsunobu and Macrocyclization

With the linkers in hand, the synthesis of the corresponding semipeptidic macrocycles was conducted on the solid phase. The tripeptides involved in this macrocycle are H₂N-Phe-Ala-Leu-OH and H₂N-Phe-Sar-Leu-OH, as well as their N-methylated derivatives, theoretically providing five compounds for each linker. However, due to the instability of the pyrrole ring in the alkylation or cyclization steps, only 10 out of 20 pyrrole macrocycles were stable enough for further experiments, delivering a total of 20 macrocycles ready for passive permeability assessment by PAMPA. The precursor peptides were built using standard Fmoc peptide synthesis, the linkers introduced using the Mitsunobu–Fukuyama reaction, and macrocyclizations performed using DEPBT or PyBOP (Scheme 6).

Scheme 6. General Solid-Phase Synthesis for Heterocycle-Containing Macrocycles.

DEPBT: 3-(diethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one, PyBOP: benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate, HATU: 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate, OBt: benzotriazol-1-yl-oxy.

Macrocyclization was the key step of the synthesis and the most challenging part of this study due to the high instability of the pyrrole-containing macrocycles. Optimization of the last step (Scheme 6) has been thoroughly performed. Surprisingly, we observed that the salt formation of pyrrole macrocycles could be a crucial factor contributing to their stability. However, the salt and neutral forms could be distinguished using NMR spectroscopy, usually giving mixtures of conformations (see the Supporting Information, Figure S1).

In addition, NMR spectra analysis showed that different conformations could be obtained using different coupling reagents (DEPBT or PyBOP). PyBOP was the most efficient coupling reagent for all pyrrole or furan macrocycles. However, as discussed above, PyBOP might form an inseparable mixture of macrocyclic conformations, whereas the formate ion and PyBOP ion compete for binding against each other for the N-terminus aminium of phenylalanine. In contrast, while DEPBT worked for all furan macrocycles (e.g., 18a), it was only favorable for one pyrrole macrocycle (15d) (see the Supporting Information, Figure 2).

Finally, this study successfully delivered 20 macrocycles, in which 10 macrocycles are pyrrole derivatives, which are also categorized into two minor groups: macrocycles with a 2,5-disubstituted pyrrole (7 compounds), macrocycles with a 2,4-disubstituted pyrrole (3 compounds), and 10 macrocycles are furan derivatives (Figure 2). These compounds were then evaluated for their permeability to characterize their structure–permeability relationship in a PAMPA assay.

Figure 2.

Newly synthesized macrocyclic peptides. (A) Pyrrole macrocycles with a 2,5-disubstitution, (B) pyrrole macrocycles with a 2,4-disubstitution, and (C.) Furan macrocycles.

Permeability Assessment of the Macrocycles

Passive permeability of the macrocycles was then assessed utilizing the parallel artificial membrane permeability assay (PAMPA), using propranolol (prop.) as a positive control (Table 3). Results are represented as −Log(P_e). A value of −LogP_e <6 is generally considered a good permeability, while −LogP_e >7 is considered poor.

Table 3. Influence of N-Methylation and Peptoids on Passive Permeability.

Cpd.	R1	R2	R3	R4	R5	linker	PAMPA −LogPe
14a	H	H	H	Me	H	2,5-disubstituted	7.72 ± 0.21
14b	H	Me	H	H	H		5.89 ± 0.08
14c	Me	H	H	Me	H		5.12 ± 0.19
14d	H	Me	H	Me	H		5.01 ± 0.01
14e	H	H	Me	Me	H		5.11 ± 0.09
15b	H	Me	H	H	Me	N-Me 2,5-disubstituted	5.93 ± 0.07
15d	H	Me	H	Me	Me		6.30 ± 0.07
16b	H	Me	H	H	H	2,4-disubstituted	6.76 ± 0.17
16d	H	Me	H	Me	H		7.53 ± 0.07
17d	H	Me	H	Me	Me	N-Me 2,4-disubstituted	6.61 ± 0.07
Prop.	N/A						5.41 ± 0.22

N-Methylation strategies have played a crucial role in the optimization of passive permeability due to increased lipophilicity and reduced hydrogen bond counts while also impacting molecular flexibility.⁶⁰ Among macrocycles with a nonmethylated 2,5-pyrrole (compounds 14a–14e), the nonmethylated precursor (14a) displayed the poorest permeability (−LogP_e = 7.72) while introducing an N-methyl group at any position (compounds 14b–e) significantly improved permeability. Replacing alanine residue with sarcosine in AA2 (14b, −LogP_e = 5.84) provided higher permeability compared to 1, although still not as high as 14d (−LogP_e = 5.01), the N-methyl alanine. In addition, N-methylation of any amide in this series (14c–e, −LogP_e = 5.12, 5.01, 5.11, respectively) increased permeability more than N-methylation of pyrrole (15b, 15d, −LogP_e = 5.93, 6.30, respectively). Despite their MWs of 439 and 453, macrocycles 14c–e displayed better passive permeability than the positive control propranolol (MW = 259). The results suggest that (1) adding an 2,5-pyrrole was important to increase permeability, an effect that might be due to increased lipophilicity or diminished flexibility, yet (2) among the N-methylated pyrrole derivatives, adding a second N-methylated amide decreased permeability in most cases, suggesting an effect on the intramolecular H-bond.

On the other hand, among macrocycles with a 2,4-pyrrole motif and similar amide N-methylation patterns, permeability was generally lower (e.g., 16b vs 14b, −logP_e = 6.75 vs 5.84; or 16d vs 14d, −LogP_e = 7.54 vs 5.01). However, N-methylation on the pyrrole ring reduced these differences (e.g., 17d vs 15d, −LogP_e = 6.62 vs 6.30). This could suggest that the 2,5-pyrrole N–H was engaged in an intramolecular H-bond as opposed to the 2,4-pyrrole N–H. In terms of positional N-methylation on the amide backbone, introducing an N-methyl group on the second amino acid gave a compound with poor permeability (16d, −LogP_e = 7.54) while replacing that amino acid with sarcosine yielded a more permeable compound (16b, −LogP_e = 6.75). Altogether, these results suggest that transannular H-bonds can be used as a positional tool to increase the permeability of the resulting macrocycles, although no clear proof of IMHB have been made in this case.

The passive permeability of furan macrocycles was also evaluated using the PAMPA assays (Table 4). Generally, furan macrocycles were postulated to have better passive permeability than pyrrole macrocycles presumably because of their reduced tPSA. In pyrrole macrocycles, only compounds containing a 2,5-NH-pyrrole ring with N-methylation on the peptide residue showed good permeability, meaning that an available hydrogen on the pyrrole ring may be beneficial for harboring the preferred conformation for membrane interaction. We hypothesize that the lack of this labile hydrogen bond on the heterocyclic ring (by replacing a pyrrole N–H moiety with the oxygen atom of the furan) could be beneficial by either providing a hydrogen bond acceptor or by removing a hydrogen bond donor (as compared to pyrrole in compounds 14a–e, 15b, and 15c). Indeed, in contrast with pyrrole-containing macrocycles, the nonmethylated version showed significantly improved permeability (18a, −LogP_e = 5.49) compared to the corresponding pyrrole compound (14a, −LogP_e = 7.72), supporting the above hypothesis. A modification on the second amino acid by replacing alanine with sarcosine (18b, −LogP_e = 5.88) displays a slightly less permeable compound compared to 18a, yet displays similar permeability to its pyrrole analog 14b. This reinforces the hypothesis on the role of the 2,5-pyrrole NH bond on passive permeability. Additionally, introducing a methyl group on any amino acid residue (18c–e, −LogP_e = 4.63, 5.14, 4.93, respectively) improves permeability, especially with the N-methylation of the leucine residue. Meanwhile, N-methylation on the linker appears to have a negative impact (19a, −LogP_e = 5.71) on the series of singly N-methylated furan macrocycles (18a–e and 19a), which also occurs with the previously discussed N-methyl pyrrole macrocycles (15b and 15d).

Table 4. Influence of N-Methylation and Peptoids on Permeability of Furan Macrocycles.

Cpd.	R1	R2	R3	R4	R5	PAMPA −LogPe
18a	H	H	H	Me	H	5.49 ± 0.09
18b	H	Me	H	H	H	5.88 ± 0.04
18c	Me	H	H	Me	H	4.63 ± 0.16
18d	H	Me	H	Me	H	5.14 ± 0.04
18e	H	H	Me	Me	H	4.93 ± 0.08
19a	H	H	H	Me	Me	5.71 ± 0.03
19b	H	Me	H	H	Me	5.30 ± 0.05
19c	Me	H	H	Me	Me	4.78 ± 0.06
19d	H	Me	H	Me	Me	4.86 ± 0.06
19e	H	H	Me	Me	Me	5.22 ± 0.08
Prop.	N/A					5.41 ± 0.22

To gain a deeper understanding on the passive permeation of methylated macrocycles, four N-methylated derivatives (including the replacement of alanine with sarcosine) of the N-methyl furan linker were prepared. None of them exhibited better passive permeability (19b–e) with −LogP_e = 5.30, 4.78, 4.86, 5.22, respectively, than compound 18c (−LogP_e = 4.62). Nevertheless, some derivatives with the N-methyl furan linker demonstrated better permeability than their nonmethylated counterparts (e.g., 18d vs 19d, −logP_e 5.88 vs 5.30; or 18d vs 19d, −LogP_e 5.14 vs 4.86). In macrocyclic peptides with 1–2 labile hydrogens (compound 19b–e), nonexchangeable IMHBs are frequently displayed as shown on the ¹H NMR spectrum (see the Supporting Information).⁶¹ The PAMPA results suggest that the passive permeability is not proportional to the number of N-methyl groups but that the N-methylation strategy can be used regioselectively to optimize IMHB patterns.⁶¹ Based on the data obtained from pyrrole- and furan-containing macrocycles, the modification on the first amino acid, leucine, leads to the most positive permeability change among a series of compounds bearing the same linker, suggesting that the Leu amide hydrogen is solvent-exposed.

Comparison with the Existing Literature

When compared to the existing prior art on permeant macrocycles, 18c ranks quite high in PAMPA permeation assays (Table 5). To our knowledge, only Le Roux’s macrocycle⁵⁷ has managed to surpass 18c, with a permeation increase of about 0.33 log. However, when compared to our most recent group’s work, permeation speed was increased by 0.69 or 0.76 log units when compared to L’Exact²² or Comeau’s²³ compounds, respectively. Quite interestingly, 18c managed to even surpass Cyclosporin A,⁶² the flag-bearer of peptide permeation by 0.38 log. Finally, even though our lead compound weighs 60% more than propanolol and has more than double its tPSA, we managed to outclass this reference compound with a permeation increase of about 0.78 log units.

Table 5. Comparison between This Work’s Most Permeant Macrocycle (18c) and Other Permeant Macrocycles Presented in the Literature.

author (year)ref	Cpd.	PAMPA (−LogPe)	permeation increase (Log)
this work (2024)	18c	4.63	N/A
	Prop.	5.41	0.78
L’Exact (2023)22	ML102	5.32	0.69
Comeau (2021)23	Nleu-5R	5.39	0.76
Le Roux (2020)57	23	4.30	–0.33
Bockus (2015)63	1NMe3	5.52	0.89
Ahlbach (2015)62	cyclosporin A	5.01	0.38

Caco-2 permeation assays were also attempted; however, pyrrole-containing macrocycles seemed to degrade during the assay. Only furan-containing macrocycles were therefore assayed in Caco-2; however, no conclusive structure–permeability data could be extracted (Table S1).

Molecular Dynamics and IMHB Formation

Although IMHB pattern modulation was only speculative in the previous paragraphs, we have decided to investigate more deeply the difference in permeability between two similar compounds having either a 2,5 or 2,4 disubstitution pattern on its pyrrole moiety (14d, −LogP_e = 5.01 vs 16d, −LogP_e = 7.53). Since 2,5-disubstituted heterocycles in peptidic scaffolds were already known to orient their heteroatom inside the macrocycle (endo position),⁴⁹⁻⁵² we hypothesized that 14d would show a similar pattern to the reported literature, while 16d would exhibit its heteroatom outside of the macrocycle (exo). The labile proton of the pyrrole moiety of 14d could also act as a HBD, promoting the formation of a IMHB with a facing carbonyl, enhancing lipophilicity (Figure 3).

Figure 3.

2D and 3D structures of 14d and 16d. (A) 2D structure of the selected macrocycles and their IMHB network (blue dashed line), (B) 3D structure of the of the lowest energy conformation of 14d (left) and 16d (right) and their IMHB network (blue dashed line) using MOE, and (C) superimposition of the lowest energy conformation (ball and stick representation) and the following 9 lowest energy conformations (line representations) of 14d (left) and 16d (right) using MOE.

Using molecular dynamic simulations in MOE,^64,65 we observe that both compounds exhibited internal H-bonds between the amide proton of leucine and the carbonyl of phenylalanine. However, 14d also presented two more IMHBs, between the linker’s amide proton and the alanine’s carbonyl and, most importantly, between the labile pyrrole proton and the leucine’s carbonyl. We believe that these IMHBs could explain in parts the enhanced permeability of 14d over 16d (−LogP_e = 5.01 vs 7.53) and 14b over 16b (−LogP_e = 5.89 vs 6.76). Despite having similar methylation patterns, these compounds show a decrease of solvent-exposed HBDs and HBAs as they are implicated in internal H-bond formation. This would therefore lead to a more lipophilic compound then previously thought, hence the enhanced permeability of 14d over 16d and 14b over 16b.

Conclusions

This project aimed to synthesize and evaluate the structure–permeability relationships of semipeptidic macrocycles incorporating heterocycles as a mean to fine-tune intramolecular H-bonds. The synthesis of heterocyclic linkers had two challenging steps requiring the development of novel reactions: the reduction of activated oximes using LiBH₄ and the aqueous reductive mono-N-alkylation of aldehydes using catalytic SmI₂ and stoichiometric Zn. We believe that these novel synthetic methods will be of interest to medicinal chemists pursuing the synthesis of bioactive, heterocycle-containing semipeptides similar to those reported herein.

Several structural modifications were studied for the permeability study: N-methylation on the tripeptide residue and linker, substitution of an amino acid by its peptoid analog, and different heterocyclic linkers with diverse orientations (2,5- vs 2,4-). Further studies using molecular dynamics revealed that the 2,5-disubstituted pyrrole exhibited an endo orientation, which contributed in better permeability than its 2,4-disubsituted counterpart (14d vs 16d). We believe that these different linker orientations had impacts on the passive permeability, likely by modulating IMHB networks, leading to enhanced permeation. Analysis of these results led to multiple observations, which produced some structure–permeability relationships.

• Limited H-bonds donors generally favor passive permeability (furan > pyrrole macrocycles).

• Mono-N-methylation on the tripeptide residue increased passive permeability regioselectively. However, di-N-methylation did not always yield better results, suggesting IMHB formation.

• Compound 18c, containing a furan ring, possesses the best permeability in PAMPA, but 14c, 14d, and 14e are very close contenders despite exibiting an added H-bond donor in the pyrrole heterocycle. These macrocycles exhibited high permeation, even surpassing the reference propanolol. This high permeation might be due to IMHB formation between the labile pyrrole N–H and a facing backbone carbonyl, as shown by molecular modeling.

Experimental Section

If not mentioned, reactions were carried out at ambient temperature and atmospheric pressure. Otherwise, reactions were conducted under a hydrogen or inert atmosphere (argon) in flame-dried glassware. Anhydrous solvents and liquid reagents were purchased from EMD Millipore DrySolv.

The thin-layer chromatography analyses were carried out on glass plates covered with silica gel (0.25 mm, Silicyle). The products by thin-layer chromatography were revealed under a UV lamp, a solution of vanillin or a solution of ninhydrin, followed by heating. Flash chromatography purifications were performed by Biotage IsoleraTM Systems equipped with various column sizes using silica gel (40–63 μm, SiliCycle), with a flow of 25–50 mL/min in varied gradients of EtOAc/Hexane.

The nuclear magnetic resonance spectra (¹H, ¹³C, HSQC, HMBC, COSY) were recorded with a Bruker Ascend 400 MHz instrument. The residual solvent peak was used as the internal standard for CDCl₃ (7.27 ppm), DMSO-d₆ (2.50 ppm), or CD₃OD (3.31 ppm) for proton resonance and CDCl₃ (77.0 ppm), DMSO-d₆ (39.5 ppm), or CD₃OD (49.2 ppm) for carbon resonance.

The UV spectra, mass spectra, and hydrogen exchange mass spectrometry experiments were recorded with an “Acquity H-Class UPLC-MS” UPLC/MS equipped with a BEH C18 column (50 × 2.1 mm, 1.7 μm spherical particle size) with a flow of 0.8 mL/min on 2.5 min in a gradient of 1–10, 1–50, 5–95, 50–95, and 80–95% acetonitrile in water +0.1% formic acid in both solvents. The UPLC-MS system and columns were purchased from Waters (Canada). The final macrocycles were purified on a preparative HPLC apparatus (Waters Sample Manager 2767, Binary gradient module 2545, SQ Detector 2) equipped with an XSelect Peptide CSH C18 OBD Prep column, column (100 × 19 mm, 5 μm spherical particle size) with a flow of 20 mL/min over 15 min in different gradients of acetonitrile in water +0.1% formic acid in both solvents. All compounds were assayed for purity (>95%) using MassLynx V4.2 built-in integration.

PAMPA Assay

The assay was carried out with 10 μL membrane (2% lecithin in dodecane), using a 100 μM solution in a phosphate butter (pH = 6.4) and shaken for 17 h (37 °C, 150 rpm). All experiments were conducted in triplicates. The PAMPA assay was performed on 200 μL hydrophobic 0.45 μm PVDF 96-well filter plates (Millipore MAIPNTR10) and 300 μL receiver plates (Millipore MATRNPS50).

Loading Procedure

To 2-chlorotrityl chloride resin (200 mg, nominal loading: 0.89 mmol/g) was added DCM (2 mL), and the mixture was left for 15 min. Meanwhile, a solution of a first amino acid (3 equiv) was prepared in DCM (2 mL). To this solution was added DIPEA (6 equiv). The DCM from the resin solution was removed by filtration, and the amino acid solution was added. The mixture was left on an orbital shaker for 4 h or overnight. To the mixture was then added 0.1 mL MeOH and left on an orbital shaker for 10 min. The resin was then filtered and washed with sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Deprotection Procedure

To the resin (200 mg) was added a solution of 20% piperidine in DMF (3 mL), and the mixture was left on an orbital shaker for 30 min. The resin was then washed 6 times with DMF.

Coupling Procedure for Amino Acids

A solution of amino acid (3 equiv) and HATU (3 equiv) in DMF (3 mL) was prepared, and then DIPEA (3 equiv) was added. The resulting yellow solution was added to the deprotected resin and left on the orbital shaker for 30 min. The resin was then filtered and washed in the following sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Nosylation Procedure

In a 20 mL vial, o-nosyl chloride (4 equiv) was dissolved in NMP (1 mL/100 mg of resin) and sym-collidine (10 equiv) was added. This mixture was poured on resin and agitated for 15 min at room temperature. The reaction was repeated twice, and then the resin was washed with a sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Procedure for Mitsunobu–Fukuyama Reaction

In a cartridge, well-dried resin was washed twice with DCM. In a vial, a solution of PPh₃ (2.5 equiv) and a hydroxylmethyl heterocycle (a linker) (2 equiv) in DCM (1 mL/100 mg resin) was prepared. The resulting solution was dispensed in the reactor. Finally, DIAD (2.5 equiv) was added dropwise and the mixture was agitated overnight at room temperature. The reaction was repeated once but only 4 h in the second cycle. The resin was washed with a sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Nosyl Deprotection Procedure

A solution of 2-mercaptoethanol (14 equiv) and DBN (5 equiv) in 3 mL DMF was poured on resin. The reaction mixture was agitated for 30 min at room temperature, and resin was filtered. The reaction was repeated twice, and then the resin was washed with the washing sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Procedure for Reductive Amination of Phenylalanine

A solution of formaldehyde (20 equiv) and THF-TMOF 1:1 mixture (3 mL) was pour on resin. The reaction mixture was agitated overnight at room temperature, and resin was filtered. In a 20 mL vial, NaBH(OAc)₃ (10 equiv) was dispersed in DCE (3 mL). The mixture was poured on the resin and agitated for 4 h at room temperature. The reaction was quenched by 1 mL MeOH and then washed with the washing sequence: 3x DMF, 3x DCM, 3x i-PrOH, 3x DCM, 3x i-PrOH, 3x DCM.

Cleavage and Deprotection Procedure

A 95% solution of TFA in DCM (2 mL) and TIPS (0.1 mL) were added to the resin, and the mixture was left on the orbital shaker for 1 h. The solution was then filtered and concentrated under reduced pressure, adding DCM when almost dry (3x) to remove most of the TFA. The crude peptide was then purified by preparative HPLC-MS to obtain a desired product.

Macrocyclization Procedure by DEBPT

The linear peptide (favorable for 15d, 18a–e, and 19a–e) was dissolved in DMF (2 mmol/L); DEBPT (3 equiv) was added, followed by DIPEA (3 equiv). The resulting solution was left to stir until completion, typically around 24 h. It was then concentrated under reduced pressure. The crude product was purified on a preparative HPLC-MS.

Macrocyclization Procedure by PyBOP

The linear peptide (favorable for all compounds) was dissolved in DMF (2 mmol/L); PyBOP (5 equiv) was added, followed by DIPEA (5 equiv). The resulting solution was left to stir until completion, typically around 1 to 2 h. It was then concentrated under reduced pressure and stored in the freezer overnight to form stable HOBt salts. The crude product was purified on a preparative HPLC-MS.

Molecular Modeling

The conformational analysis was conducted using the Molecular Operating Environment (MOE), 2022.02 Chemical Computing Group ULC, 910-1010 Sherbrooke St. W., Montreal, QC H3A 2R7, Canada, 2023. Compounds 14d and 16d were constructed using MOE’s molecular sequence editor for the peptide part and the builder module for the linker. The structures were cleaned and optimized using MOE’s built-in tools to ensure a reasonable initial geometry. Protonation states and charge distributions were assigned according to the physiological pH conditions. The Amber10:EHT force field was employed for all simulations and energy calculations, and R-field was used as an implicit solvatation model.

Conformational Search Methodology

The conformational search was performed using the LowModeMD method, a technique optimized for exploring conformational space efficiently. Default parameters were applied, including a rejection limit of 100, an iteration limit of 10,000, an RMS gradient of 0.005, and an MM iteration limit of 500. Multiple runs were conducted to ensure comprehensive sampling. Each run was initiated from a different randomly generated starting conformation to avoid potential bias. The resultant conformations were analyzed based on their relative energy levels. The top 10 conformations, ranked by their minimized energy states, were selected for further analysis and highlighted for their IMHBs. All 10 energy-minimal conformations showed similar secondary structure and H-bond networks.

Methyl 5-Formyl-1H-pyrrole-2-carboxylate (2a) and Methyl 4-Formyl-1H-pyrrole-2-carboxylate (2b)

Formylation of pyrrole 1 was performed according to the procedure reported by Hawker and Silverman.⁵³

POCl₃ (13.3 mL, 143 mmol) was added within 15 min to stirred, ice cold DMF (13.3 mL, 172 mmol). The reaction mixture was diluted with DCM (10 mL). A solution of ester (10.0 g, 79.9 mmol) in DCM (10 mL) was added dropwise within 45 min, while the temperature was kept at 2–10 °C. After the addition was complete, the reaction mixture was refluxed for 90 min under vigorous stirring. Upon completion, the reaction was cooled to room temperature, slowly added a sodium acetate aqueous solution (120 g/500 mL) (15 mL), and stirred for 15 min more. Then, phases were separated, and to the aqueous phase was slowly added NaOH 15% solution until the pH reached 10–11. Extraction was then performed with EtOAc (15 mL x 10). The combined organic extracts were washed with water and dried with Na₂SO₄. The filtrate was collected, the solvent was removed under reduced pressure, and then the solid, orange residue was purified by flash chromatography (Hexanes: EtOAc = 8:2 to obtain 2a, and Hexanes: EtOAC = 6:4 to obtain 2b) to yield desired products.

Methyl 5-Formylpyrrole-2-carboxylate (2a)

A white solid (8.40 g, 68%); ¹H NMR (400 MHz, DMSO-d₆) δ 9.70 (s, 1H), 6.97 (d, J = 4.0 Hz, 1H), 6.89 (d, J = 4.0 Hz, 1H), 3.83 (s, 3H). ¹³C NMR (101 MHz, DMSO-d₆) δ 181.47, 160.4, 135.6, 127.6, 116.6, 115.7, 51.9; HRMS [M + Na]⁺ calcd for C₇H₇NO₃: 176.0318, found 176.0323.

Methyl 4-Formylpyrrole-2-carboxylate (2b)

A yellow solid (3.21 g, 26%); ¹H NMR (400 MHz, DMSO-d₆) δ 12.72 (s, 1H), 9.75 (s, 1H), 7.82 (s, 1H), 7.13 (d, J = 1.6 Hz, 1H), 3.80 (s, 3H). ¹³C NMR (101 MHz, DMSO-d₆) δ 185.9, 160.5, 131.2, 126.7, 124.3, 113.1, 51.7; HRMS [M + Na]⁺ calcd for C₇H₇NO₃: 176.0318, found 176.0315.

Methyl (E)-5-((Hydroxyimino)methyl)-1H-pyrrole-2-carboxylate (3a)

Condensation of pyrrole 2a was performed according to the procedure reported by Hawker and Silverman.⁵³

Methyl 5-formyl-1H-pyrrole-2-carboxylate 2a (2.00 g, 13.1 mmol) was dissolved in water (75 mL) at 80 °C, and a solution of hydroxylamine hydrochloride (1.81 g, 26.1 mmol) and potassium carbonate (1.82 g, 13.1 mmol) in water (5 mL) was added dropwise. Upon cooling, a white precipitate formed, which was collected by filtration and purified by washing with a sufficient mixture of water and DCM to give the desired product as a white powder (1.98 g, 90%). Analysis by ¹H NMR spectroscopy indicated a mixture of the two isomers (2:1 in ratio). ¹H NMR (400 MHz, methanol-d₄) δ (major isomer) 7.98 (s, 1H), 6.83 (d, J = 3.9 Hz, 1H), 6.43 (d, J = 3.9 Hz, 1H), 3.82 (s, 3H); δ (minor isomer) 7.34 (s, 1H), 6.86 (d, J = 4.0 Hz, 1H), 6.69 (d, J = 3.9 Hz, 1H), 3.84 (s, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ (major isomer) 161.5, 140.5, 130.3, 123.8, 115.9, 110.9, 50.6; δ (minor isomer) 161.4, 136.7, 128.1, 123.5, 115.1, 115.0, 50.8; HRMS [M + Na]⁺ calcd for C₇H₈N₂O₃: 191.0427, found 191.0425.

Methyl (E)-4-((Hydroxyimino)methyl)-1H-pyrrole-2-carboxylate (3b)

Condensation of pyrrole 2b was performed according to the procedure reported by Hawker and Silverman.⁵³

Note: the product had been stored in the freezer for two years before it was characterized so the E-Z isomerization might take place. Similar preparation to 3a, which was mentioned above to yield 2 isomers (14:1 in ratio). The product was obtained as a yellow solid (1.97, 90%); ¹H NMR (400 MHz, methanol-d₄) δ (major isomer) 7.67 (d, J = 1.5 Hz, 1H), 7.26 (s, 1H), 7.22 (d, J = 1.5 Hz, 1H), 3.84 (s, 3H); δ (minor isomer) 8.00 (s, 1H), 7.20 (d, J = 1.6 Hz, 1H), 7.07 (d, J = 1.6 Hz, 1H), 3.83 (s, 4H). ¹³C NMR (101 MHz, methanol-d₄) δ (only major isomer) 161.6, 140.7, 127.6, 122.2, 116.8, 116.4, 50.5; HRMS [M + Na]⁺ calcd for C₇H₈N₂O₃: 191.0427, found 191.0425; HRMS [M + H]⁺ calcd for C₇H₈N₂O₃: 169.0608, found 169.0606.

1-(tert-Butyl) 2-Methyl 5-((Bis(tert-butoxycarbonyl)amino)methyl)-1H-pyrrole-1,2-dicarboxylate (4a)

Methyl 5-((hydroxyimino)methyl)-1H-pyrrole-2-carboxylate (3.00 g, 17.8 mmol) and DMAP (0.109 g, 0.05 mmol) were dissolved in ethyl acetate (900 mL, to form a 0.02 M solution), followed by the slow addition of Boc₂O (12.5 g, 57.1 mmol). The mixture was stirred for 60 min at room temperature to activate the oxime. Then, it was placed under an H₂ atmosphere in the presence of 10% Pd/C (0.300 g) and vigorously stirred overnight at room temperature. Upon completion of the hydrogenation reaction as judged by TLC or UPLC-MS, to the stirred reaction was added DMAP (0.109 g, 0.05 mmol) and the reaction mixture was left to stir under ambient conditions for 60 min. Upon completion, the reaction mixture was filtered through a Celite pad; then EtOAc was evaporated to afford the crude product as an orange liquid. The crude product was then purified by flash chromatography (EtOAc/hexanes, 0–20% EtOAc gradient) to yield the desired product as a light-yellow liquid (7.70 g, 95%). ¹H NMR (400 MHz, methanol-d₄) δ 6.81 (d, J = 3.7 Hz, 1H), 5.97 (dt, J = 3.7, 1.1 Hz, 1H), 4.87 (d, J = 1.1 Hz, 2H), 3.81 (s, 3H), 1.58 (s, 9H), 1.46 (s, 18H). ¹³C NMR (101 MHz, methanol-d₄) δ 161.1, 152.1, 149.1, 137.9, 124.8, 118.8, 107.7, 85.3, 82.9, 50.9, 43.4, 26.8, 26.4; HRMS [M + Na]⁺ calcd for C₂₂H₃₄N₂O₈: 477.2207, found 477.2213.

1-(tert-Butyl) 2-Methyl 4-((Bis(tert-butoxycarbonyl)amino)methyl)-1H-pyrrole-1,2-dicarboxylate (4b)

Similar preparation to 4a as mentioned above. The product was obtained as a light-yellow liquid (7.30 g, 90%). ¹H NMR (400 MHz, methanol-d₄) δ 7.29 (d, J = 1.8 Hz, 1H), 6.81 (d, J = 1.9 Hz, 1H), 4.55 (s, 2H), 3.81 (s, 3H), 1.56 (s, 9H), 1.49 (s, 18H). ¹³C NMR (101 MHz, methanol-d₄) δ 161.2, 152.4, 148.2, 125.1, 122.5, 120.4, 84.8, 82.6, 51.2, 41.5, 27.1, 26.7, 26.6; HRMS [M + Na]⁺ calcd for C₂₂H₃₄N₂O₈: 477.2207, found 477.2211.

tert-Butyl 2-(((tert-Butoxycarbonyl)amino)methyl)-5-(hydroxymethyl)-1H-pyrrole-1-carboxylate (5a)

To a round-bottomed flask equipped with a mechanical stirrer were added 220 mL of dry DCM and 4a (2.00 g, 4.40 mmol) to obtain 0.02 M solution. To the stirred solution at 0 °C was added 26.4 mL of a DIBAL-H solution (1 M, 26.4 mmol) over 30 min. When the addition was complete and gas evolution subsided, the mixture continued to be stirred at room temperature for 4 h more. Upon completion, to the mixture were slowly added 2.0 mL of water, 1 mL of MeOH, 1 mL of 15% NaOH solution, and 106 mL saturated Rochelle’s salt. The mixture was then vigorously stirred until becoming a quite clear solution. The product was then extracted by 4 × 50 mL of EtOAc. Then, the organic layer was washed with 20 mL of saturated NaCl solution, dried over Na₂SO₄, filtered, and then concentrated under reduced pressure. The crude was purified by flash chromatography (hexane:EtOAc = 3:2) to afford the final product as a yellow liquid (0.674 mg, 47%). On TLC, the product turned yellow after heated. ¹H NMR (400 MHz, methanol-d₄) δ 6.13 (d, J = 3.4 Hz, 1H), 6.04 (d, J = 3.4 Hz, 1H), 4.67 (s, 2H), 4.35 (s, 2H), 1.63 (s, 9H), 1.45 (s, 9H). ¹³C NMR (101 MHz, methanol-d₄) δ 156.8, 150.1, 135.7, 133.5, 111.0, 110.3, 84.6, 78.9, 58.0, 48.5, 27.4, 26.8; HRMS [M + Na]⁺ calcd for C₁₆H₂₆N₂O₅: 349.1734, found 349.1737.

tert-Butyl 4-(((tert-Butoxycarbonyl)amino)methyl)-2-(hydroxymethyl)-1H-pyrrole-1-carboxylate (5b)

Similar preparation to 5a, which was mentioned above. The product was obtained as a yellow liquid (0.670 g, 47%); on TLC, the product turned yellow after heated. ¹H NMR (400 MHz, methanol-d₄) δ 7.10 (dt, J = 2.1, 1.0 Hz, 1H), 6.24–6.12 (m, 1H), 4.79–4.57 (m, 2H), 4.00 (s, 2H), 1.60 (s, 9H), 1.45 (s, 9H). ¹³C NMR (101 MHz, methanol-d₄) δ 157.1, 149.4, 135.3, 123.7, 118.6, 112.2, 83.7, 78.7, 57.37, 36.5, 27.4, 26.8; HRMS [M + Na]⁺ calcd for C₁₆H₂₆N₂O₅: 349.1734, found 349.1745.

Methyl 5-Formyl-1-methyl-1H-pyrrole-2-carboxylate (6a)⁴³

A reaction flask was charged with 2a (2.00 g, 13.1 mmol), dimethyl carbonate (16.5 mL, 196 mmol), DABCO (0.147 g, 1.31 mmol), and DMF (13.1 mL). The resulting mixture was heated to 90–92 °C and stirred at that temperature for 23 h. UPLC analysis indicated complete consumption of the starting material. The reaction mixture was cooled to 20 °C, diluted with EtOAc (50 mL), and transferred to a separatory funnel. The solution was washed sequentially with H₂O (50 mL) and 0.1 M HCl solution (2 × 30 mL). The aqueous washes were combined and extracted with EtOAc (25 mL). The combined organic layers were washed with brine (50 mL), dried over Na₂SO₄, and concentrated in vacuo to afford the crude product. The crude product was then purified by coloumn chromatography (EtOAc/hexanes = 1:3) to yield the desired product as a light-yellow solid (1.83 g, 84%). ¹H NMR (400 MHz, CDCl₃) δ 9.72 (s, 1H), 6.93 (d, J = 4.3 Hz, 1H), 6.87 (d, J = 4.3 Hz, 1H), 4.28 (s, 3H), 3.87 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 181.5, 161.5, 135.5, 129.9, 122.1, 117.0, 52.0, 34.7; HRMS [M + Na]⁺ calcd for C₈H₉NO₅: 190.0475, found 190.0471.

Methyl 4-Formyl-1-methyl-1H-pyrrole-2-carboxylate (6b)

Similar preparation to 6a, which was mentioned above. The product was obtained as a white solid (1.92 g, 88%); ¹H NMR (400 MHz, methanol-d₄) δ 9.67 (s, 1H), 7.71–7.68 (m, 1H), 7.30 (d, J = 1.9 Hz, 1H), 3.98 (d, J = 0.6 Hz, 3H), 3.83 (s, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 186.1, 161.1, 135.7, 124.9, 124.5, 116.1, 50.6, 36.5; HRMS [M + H]⁺ calcd for C₈H₉NO₅: 168.0655, found 168.0653.

Methyl (E)-5-((Hydroxyimino)methyl)-1-methyl-1H-pyrrole-2-carboxylate (7a)

Methyl 5-formyl-1-methyl-1H-pyrrole-2-carboxylate (1.65 g, 9.90 mmol) was dispersed in water (100 mL) at 80 °C; methanol was added into the mixture dropwise until obtaining a transparent solution. A solution of hydroxylamine hydrochloride (1.37 g, 9.90 mmol) and potassium carbonate (1.37 g, 19.8 mmol) in water (10 mL) was then added dropwise. Upon cooling, a white precipitate formed, which was collected by filtration and purified by washing with sufficient mixture of water and methanol to give the desired product in a mixture of the two geometric isomers (5:1 in ratio) as a white solid (1.67 g, 93%). ¹H NMR (400 MHz, CDCl₃) δ (major isomer) 8.13 (s, 1H), 6.93 (d, J = 4.2 Hz, 1H), 6.43 (d, J = 4.2 Hz, 1H), 4.10 (s, 3H), 3.83 (s, 3H); δ (minor isomer) 7.52 (s, 1H), 7.22 (d, J = 4.3 Hz, 1H), 6.99 (d, J = 4.3 Hz, 1H), 4.06 (s, 3H), 3.85 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ (major isomer) 161.7, 142.9, 131.5, 125.9, 117.8, 112.4, 51.5, 34.5; δ (minor isomer) 169.0, 148.1, 135.8, 123.0, 121.8, 117.1, 51.6, 32.6; HRMS [M + Na]⁺ calcd for C₈H₁₀N₂O₃: 205.0584, found 205.0585.

Methyl (E)-4-((Hydroxyimino)methyl)-1-methyl-1H-pyrrole-2-carboxylate (7b)

Methyl 4-formyl-1-methyl-1H-pyrrole-2-carboxylate (1.3 g, 7.8 mmol) was dissolved in methanol (50 mL) at room temperature to obtain a transparent solution. A solution of hydroxylamine hydrochloride (1.08 g, 7.90 mmol) and potassium carbonate (1.08 g, 15.6 mmol) in water (10 mL) was then added dropwise. After 1 h, methanol was evaporated to have 25 mL left followed by an addition of 50 mL of water into a reaction mixture. Upon cooling to 0 °C, a yellow precipitate formed, which was collected by filtration and purified by washing with a sufficient mixture of water and methanol to give the desired product in a mixture of the two geometric isomers (3:2 in ratio) as a yellow solid (0.935 g, 66%). ¹H NMR (400 MHz, methanol-d₄) δ (major isomer) 7.94 (s, 1H), 7.18 (d, J = 1.8 Hz, 1H), 7.10 (d, J = 1.9, 1H), 3.90 (s, 3H), 3.80 (s, 3H); δ (minor isomer) 7.70–7.65 (m, 1H), 7.26–7.23 (m, 1H), 7.20 (s, 1H), 3.93 (s, 3H), 3.81 (s, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ (major isomer) 161.5, 143.6, 133.5, 129.6, 119.3, 114.8, 50.3, 35.9; δ (minor isomer) 161.5 (two overlapping signals), 140.35, 125.6, 123.3, 122.2, 116.5, 50.3 (two overlapping signals), 35.9; HRMS [M + H]⁺ calcd for C₈H₁₀N₂O₃: 183.0764, found 183.0761.

Methyl 5-(((tert-Butoxycarbonyl)amino)methyl)-1-methyl-1H-pyrrole-2-carboxylate (8a)

Methyl (E)-5-((hydroxyimino)methyl)-1-methyl-1H-pyrrole-2-carboxylate (1.68 g, 9.23 mmol) and DMAP (0.56 mg, 0.046 mmol) were dissolved in ethyl acetate (460 mL, to form a 0.02 M solution), followed by the slow addition of Boc₂O (5.04 g, 23.1 mmol). The mixture was stirred in 60 min at room temperature to activate the oxime. Then, it was placed under an H₂ atmosphere in the presence of 10% Pd/C (168 mg) and vigorously stirred overnight at room temperature. Upon the complete hydrogenation as judged by TLC or UPLC-MS, the reaction mixture was filtered through a Celite pad; then EtOAc was evaporated to afford the crude product as an orange liquid. The crude product was then purified by flash chromatography (EtOAc/hexanes, 0–30% EtOAc gradient) to yield the desired product as a white solid (2.10 g, 85%). ¹H NMR (400 MHz, CDCl₃) δ 6.87 (d, J = 4.0 Hz, 1H), 6.05 (d, J = 3.9 Hz, 1H), 4.67 (s, 1H), 4.32 (d, J = 5.7 Hz, 2H), 3.87 (s, 3H), 3.80 (s, 3H), 1.45 (s, 9H). ¹³C NMR (101 MHz, CDCl₃) δ 161.9, 155.5, 136.9, 123.4, 117.1, 108.6, 80.1, 51.2, 36.9, 32.8, 28.5; HRMS [M + H]⁺ calcd for C₁₃H₂₀N₂O₄: 269.1496, found 269.1503.

Methyl 4-(((tert-Butoxycarbonyl)amino)methyl)-1-methyl-1H-pyrrole-2-carboxylate (8b)

Similar preparation to 8a, which was mentioned above. The product was obtained as a white solid (2.03 g, 82%). ¹H NMR (400 MHz, methanol-d₄) δ 6.83 (d, J = 2.1 Hz, 1H), 6.82 (d, J = 2.0 Hz, 1H), 4.02 (d, J = 4.5 Hz, 2H), 3.85 (s, 3H), 3.77 (s, 3H), 1.44 (s, 9H). ¹³C NMR (101 MHz, methanol-d₄) δ 161.8, 157.0, 128.0, 121.8, 121.4, 116.7, 78.7, 50.0, 36.3, 35.5, 27.4; HRMS [M + Na]⁺ calcd for C₁₃H₂₀N₂O₄: 291.1315, found 291.1310.

tert-Butyl ((5-(Hydroxymethyl)-1-methyl-1H-pyrrol-2-yl)methyl)carbamate (9a)

9a was prepared from 8a (1.00 g, 3.73 mmol) with a similar procedure as 5a, which was mentioned above, but to the reaction were added an additional 3 equiv of THF and 5 equiv of DIBAL (18.6 mL, 18.6 mmol). The crude product was purified by flash chromatography (EtOAc/hexanes, 0–50% EtOAc gradient) to yield the desired product as a yellow solid (0.564 g, 63%); on TLC, the product turned pink to purple after heating. ¹H NMR (400 MHz, CDCl₃) δ 6.00 (d, J = 3.5 Hz, 1H), 5.95 (d, J = 3.5 Hz, 1H), 4.65 (s, 1H), 4.55 (s, 2H), 4.27 (d, J = 5.6 Hz, 2H), 3.59 (s, 3H), 1.71 (s, 1H), 1.45 (s, 9H). ¹³C NMR (101 MHz, CDCl₃) δ 155.6, 133.0, 131.1, 107.8, 107.2, 79.7, 57.0, 36.9, 30.6, 28.5; HRMS [M + Na]⁺ calcd for C₁₂H₂₀N₂O₃: 263.1366, found 263.1372.

tert-Butyl ((5-(Hydroxymethyl)-1-methyl-1H-pyrrol-3-yl)methyl)carbamate (9b)

Similar preparation to 9a, which was mentioned above. The crude product was purified by flash chromatography (EtOAc/hexanes, 0–50% EtOAc gradient) to yield the desired product as a light-yellow white solid (0.502 g, 56%). ¹H NMR (400 MHz, methanol-d₄) δ 6.55 (d, J = 2.0 Hz, 1H), 5.98 (d, J = 2.0 Hz, 1H), 4.48 (s, 2H), 4.04–3.97 (m, 2H), 3.60 (s, 3H), 1.44 (s, 9H). ¹³C NMR (101 MHz, methanol-d₄) δ 157.0, 131.9, 121.0, 119.7, 108.1, 78.5, 55.2, 36.9, 32.3, 27.4; HRMS [M + Na]⁺ calcd for C₁₂H₂₀N₂O₃: 263.1366, found 263.1372.

(E)-5-(Hydroxymethyl)furan-2-carbaldehyde Oxime (10)

5-Hydroxymethyl-2-furaldehyde (5.0 g, 40 mmol) was dissolved in water (400 mL) at 80 °C, and a solution of hydroxylamine hydrochloride (5.5 g, 79 mmol) and potassium carbonate (5.5 g, 40 mmol) in water (10 mL) was added dropwise. The reaction was gradually cooled to 0 °C while being stirred for 2 h. Upon completion, the aqueous solution was extracted with EtOAc (50 mL x 3). Then, the combined organic phases were washed with 20 mL of saturated NaCl solution, dried over Na₂SO₄, then concentrated to obtain the desired product as a mixture of the two geometric isomers (5:2 in ratio) as a yellow solid (5.17 g, 93%). ¹H NMR (400 MHz, methanol-d₄) δ (major isomer) 7.94 (s, 1H), 6.60 (d, J = 3.3 Hz, 1H), 6.38 (d, J = 3.3 Hz, 1H), 4.52 (s, 2H); δ (minor isomer) 7.40–7.35 (m, 1H), 7.21 (d, J = 3.4 Hz, 1H), 6.44 (d, J = 3.4 Hz, 1H), 4.54 (s, 2H). ¹³C NMR (101 MHz, methanol-d₄) δ (major isomer) 156.3, 147.7, 139.3, 112.2, 108.8, 56.0; δ (minor isomer) 155.2, 145.3, 135.7, 117.7, 109.2, 56.0 (two overlapping signals); HRMS [M + Na]⁺ calcd for C₆H₇NO₃: 164.0318, found 164.0323.

tert-Butyl ((5-(Hydroxymethyl)furan-2-yl)methyl)carbamate (11)

Compound 10 (1.0 g, 7.1 mmol) was dissolved in methanol (7 mL); then dichloromethane (142 mL) was added to obtain a 0.05 M solution. To the stirred solution were added NiCl₂ (181 mg, 1.42 mmol) and LiBH₄ 2 M solution in THF (4.3 mL, 8.5 mmol). After 5 min, Boc₂O (1.9 g, 8.5 mmol) was added and the mixture continued to be stirred for 30 min. Then, to the stirred reaction were added additional LiBH₄ solution (8.5 mL, 2.4 mmol) and Boc₂O (1.9 g, 8.5 mmol) and the reaction mixture was stirred under ambient condition for the next 4 h. Upon completion, water (50 mL) and EtOAc (50 mL) were added into the vigorously stirred mixture. The organic layer was then extracted, washed with saturated NaCl solution (20 mL), dried over Na₂SO₄, and concentrated to afford the crude product as a yellow solid. It was then purified by chromatography (EtOAc/hexanes = 2:3) to yield the desired product as a light-yellow solid (0.998 g, 62%). On TLC, the product cannot be detected by UV, although it was revealed by CAM staining after the TLC plate was heated. ¹H NMR (400 MHz, methanol-d₄) δ 6.21 (d, J = 3.1 Hz, 1H), 6.14 (d, J = 3.1 Hz, 1H), 4.46 (s, 2H), 4.18 (s, 2H), 1.44 (s, 9H). ¹³C NMR (101 MHz, methanol-d₄) δ 156.8, 153.9, 152.4, 107.9, 106.9, 79.0, 56.0, 37.0, 27.4; HRMS [M + Na]⁺ calcd for C₁₁H₁₇NO₄: 250.1050, found 250.1053.

tert-Butyl ((5-(Hydroxymethyl)furan-2-yl)methyl)(methyl)carbamate (13)

HMF (1.0 g, 7.9 mmol) was dissolved in 79 mL of THF:H₂O (95:5) solution, followed by the addition of methylamine 40% solution in water (12.3 mL, 159 mmol). After 10 min, zinc (1.03 g, 15.9 mmol) and SmI₂ 0.1 M solution in THF (15.9 mL, 1.59 mmol) were added into the reaction and the mixture was stirred at room temperature for 4 h. The formation of intermediate [12] was confirmed by UPLC-MS. The reaction mixture was then concentrated in vacuo, and the remaining crude solid was dispersed into 108 mL of acetonitrile followed by the addition of Boc₂O (2.08 g, 9.52 mmol). The mixture was stirred for 30–60 min. Upon completion, water (50 mL), brine (50 mL), and EtOAc (2 × 50 mL) were added into the vigorously stirred mixture. The organic layer was then extracted, washed with saturated NaCl solution (20 mL), and concentrated to afford the crude product as a yellow solid. It was then purified by chromatography (EtOAc/hexanes = 2:3) to yield the desired product as a light-yellow solid (1.148 g, 60%). ¹H NMR (400 MHz, methanol-d₄) δ 6.24 (d, J = 3.2 Hz, 1H), 6.19 (d, J = 3.2 Hz, 1H), 4.86 (s, 3H), 4.47 (s, 2H), 4.37 (s, 2H), 2.86 (s, 3H), 1.47 (s, 10H). ¹³C NMR (101 MHz, methanol-d₄) δ 156.0, 154.4, 151.3, 107.8 (two overlapping signals), 80.0, 56.0 (two overlapping signals), 33.0, 27.3; HRMS [M + Na]⁺ calcd for C₁₂H₁₉NO₄: 264.1206, found 264.1213.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8-methyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione, 1H-Benzo[d][1,2,3]triazol-1-ol (14a)

The product was obtained as a salt of HOBt. ¹H NMR (400 MHz, methanol-d₄) δ 7.89 (dt, J = 8.6, 1.0 Hz, 1H), 7.76–7.60 (m, 2H), 7.51 (ddd, J = 8.1, 5.4, 2.4 Hz, 1H), 7.37–7.22 (m, 6H), 6.27 (d, J = 3.4 Hz, 1H), 5.96 (d, J = 3.4 Hz, 1H), 5.68 (s, 2H), 4.42–4.27 (m, 2H), 4.27–4.17 (m, 2H), 4.09 (dd, J = 8.6, 5.3 Hz, 1H), 3.30–3.22 (m, 1H), 2.98 (dd, J = 14.2, 8.6 Hz, 1H), 1.71–1.44 (m, 3H), 1.30 (d, J = 7.1 Hz, 3H), 0.92 (d, J = 6.5 Hz, 3H), 0.88 (d, J = 6.5 Hz, 3H). Note: some quaternary carbons were not detectable; ¹³C NMR (101 MHz, methanol-d₄) δ 173.2 (C), 173.1 (C), 134.7 (C), 133.6 (C), 130.4 (C), 130.3 (CH), 129.1 (CH), 128.7 (CH), 127.3 (CH), 125.5 (CH), 122.8 (C), 114.0 (CH), 112.0 (CH), 110.0 (CH), 106.4 (CH), 54.5 (CH), 52.2 (CH), 48.9 (CH), 46.0 (CH₂), 40.3 (CH₂), 37.6 (CH₂), 35.9 (CH₂), 24.4 (CH), 21.9 (CH₃), 20.7 (CH₃), 16.8 (CH₃); HRMS [M + H]⁺ calcd for C₂₄H₃₃N₅O₃: 440.2656, found 440.2654.

(5S,11S)-11-Benzyl-5-isobutyl-9-methyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione (14b)

The product was obtained as a mixture of two conformations, ratio 3:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.37–7.02 (m, 5H), 5.95 (d, J = 3.3 Hz, 1H), 5.80 (d, J = 3.3 Hz, 1H), 4.61 (dd, J = 15.5, 1.0 Hz, 1H), 4.34 (d, J = 15.4 Hz, 1H), 4.28–4.06 (m, 3H), 4.06–3.73 (m, 2H), 3.62–3.47 (m, 2H), 3.26 (d, J = 15.4 Hz, 1H), 3.01 (dd, J = 12.8, 5.5 Hz, 1H), 2.84 (ddd, J = 12.9, 6.0, 3.6 Hz, 2H), 2.77 (s, 1H), 2.57 (s, 3H), 2.30–1.91 (m, 1H), 1.89–1.52 (m, 3H), 1.40–1.22 (m, 2H), 0.94 (d, J = 6.0 Hz, 3H), 0.89 (d, J = 5.4 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 173.5, 172.2, 170.8, 164.0, 138.8, 137.4, 129.3, 129.2, 129.16, 129.07, 128.98, 128.50, 128.32, 128.28, 126.97, 106.04, 104.39, 59.0, 53.4, 52.0, 51.0, 44.1, 39.0, 36.1, 35.9, 33.8, 24.5, 22.2, 21.6, 21.1, 19.3; HRMS [M + H]⁺ calcd for C₂₄H₃₃N₅O₃: 440.2656, found 440.2650.

(5S,8S,11S)-11-Benzyl-5-isobutyl-6,8-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione, 1H-Benzo[d][1,2,3]triazol-1-ol (14c)

The product was obtained as a mixture of 2 conformations of HOBt salts, ratio 5:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.90 (dt, J = 8.6, 1.0 Hz, 1H), 7.72–7.65 (m, 2H), 7.52 (ddd, J = 8.6, 5.3, 2.6 Hz, 1H), 7.40–7.22 (m, 8H), 6.24 (d, J = 3.4 Hz, 1H), 5.94 (d, J = 3.4 Hz, 1H), 5.67 (s, 2H), 4.97 (dd, J = 10.1, 5.7 Hz, 1H), 4.79 (q, J = 6.9 Hz, 1H), 4.33 (d, J = 15.2 Hz, 1H), 4.19 (d, J = 15.0 Hz, 1H), 4.12 (dd, J = 8.9, 5.2 Hz, 1H), 3.30–3.20 (m, 1H), 2.98 (s, 3H), 2.93 (dd, J = 14.4, 8.9 Hz, 1H), 2.80 (s, 1H), 1.78–1.55 (m, 2H), 1.53–1.41 (m, 1H), 1.23 (d, J = 7.0 Hz, 3H), 0.93 (d, J = 6.6 Hz, 3H), 0.88 (d, J = 6.5 Hz, 3H); ¹³C NMR (101 MHz, methanol-d₄) δ 171.5, 169.2, 161.4, 134.2, 132.5, 130.4, 129.8, 129.1, 128.75, 127.49, 125.51, 114.01, 111.91, 109.77, 106.72, 60.40, 56.10, 55.3, 54.1, 45.9, 40.6, 36.7, 30.3, 24.6, 22.1, 20.7, 15.7; HRMS [M + H]⁺ calcd for C₂₅H₃₅N₅O₃: 454.2813, found 454.2827.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8,9-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione (14d)

A mixture of 2 conformations, ratio 5:2 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.39–7.03 (m, 5H), 6.06 (d, J = 3.2 Hz, 1H), 5.92 (dd, J = 3.3, 0.8 Hz, 1H), 4.33 (dd, J = 9.0, 6.2 Hz, 1H), 4.05–3.93 (m, 2H), 3.82 (dd, J = 29.4, 14.6 Hz, 2H), 3.68 (q, J = 6.9 Hz, 1H), 3.56 (ddz, J = 10.7, 4.1 Hz, 1H), 3.25 (dd, J = 12.7, 4.1 Hz, 1H), 2.76–2.64 (m, 1H), 2.61 (s, 1H), 2.59 (s, 3H), 2.03 (d, J = 5.0 Hz, 1H), 1.77–1.49 (m, 2H), 1.43 (d, J = 7.0 Hz, 1H), 0.99 (d, J = 6.0 Hz, 1H), 0.94 (d, J = 6.1 Hz, 1H), 0.94 (d, J = 6.6 Hz, 3H), 0.89 (d, J = 6.5 Hz, 3H), 0.43 (d, J = 6.9 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 172.5, 171.2, 168.8, 132.4, 130.5, 129.5, 129.2, 129.0, 128.6, 128.4, 128.3, 126.7, 107.8, 105.9, 56.4, 54.2, 51.1, 44.4, 41.4, 37.6, 35.7, 28.6, 24.6, 22.0, 20.8, 13.1; HRMS [M + H]⁺ calcd for C₂₅H₃₅N₅O₃: 454.2813, found 454.2805.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8,12-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione, 1H-Benzo[d][1,2,3]triazol-1-ol (14e)

¹H NMR (400 MHz, methanol-d₄) δ 7.97–7.83 (m, 1H), 7.73–7.59 (m, 2H), 7.51 (ddd, J = 8.2, 5.5, 2.4 Hz, 1H), 7.37–7.24 (m, 6H), 6.28 (d, J = 3.4 Hz, 1H), 5.96 (d, J = 3.4 Hz, 1H), 5.67 (s, 2H), 4.48–4.34 (m, 2H), 4.24–4.11 (m, 3H), 3.27 (dd, J = 14.2, 6.5 Hz, 1H), 3.15 (dd, J = 14.3, 6.9 Hz, 1H), 2.64 (s, 3H), 1.64 (ddt, J = 12.6, 8.2, 6.1 Hz, 1H), 1.59–1.44 (m, 2H), 1.30 (d, J = 7.0 Hz, 3H), 0.93 (d, J = 6.6 Hz, 3H), 0.89 (d, J = 6.5 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 173.2, 172.7, 167.8, 133.6, 130.5, 130.3, 129.2, 128.7, 127.5, 125.5, 122.7, 114.0, 112.0, 110.0, 106.3, 62.4, 52.3, 48.9, 48.2, 48.0, 47.8, 47.6, 47.4, 47.2, 47.0, 46.1, 40.3, 36.4, 35.9, 31.4, 24.4, 21.9, 20.8, 16.9; HRMS [M + H]⁺ calcd for C₂₅H₃₅N₅O₃: 454.2813, found 454.2826.

(5S,11S)-11-Benzyl-5-isobutyl-1¹,9-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione (15b)

Note: the synthesis was not efficient, the yield was too low and the product was too unstable that the¹³C NMR could not be achieved. The product was obtained as a mixture of 3 conformations, ratio 6:1:1. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.35–7.06 (m, 5H), 6.01 (d, J = 3.6 Hz, 1H), 5.92 (s, 0.17H), 5.88 (d, J = 3.5 Hz, 1H), 5.35 (dd, J = 5.4, 4.1 Hz, 1H), 4.55 (d, J = 15.0 Hz, 1H), 4.34 (d, J = 13.6 Hz, 1H), 4.26–4.12 (m, 3H), 4.04 (dd, J = 18.0, 15.0 Hz, 2H), 3.80–3.71 (m, 1H), 3.47 (s, 3H), 3.07–2.97 (m, 1H), 2.89–2.77 (m, 5H), 2.66 (d, J = 1.0 Hz, 1H), 2.58 (s, 3H), 2.23–2.14 (m, 1H), 2.09–1.99 (m, 4H), 0.95 (d, J = 6.2 Hz, 3H), 0.93–0.84 (m, 5H); HRMS [M + H]⁺ calcd for C₂₅H₃₅N₅O₃: 454.2813, found 454.2809.

(5S,8S,11S)-11-Benzyl-5-isobutyl-1¹,8,9-trimethyl-1¹H-3,6,9,12-tetraaza-1(2,5)-pyrrolacyclotridecaphane-4,7,10-trione (15d)

Note: the synthesis was not efficient, the yield was too low and the product was too unstable that the¹³C NMR could not be achieved. ¹H NMR (400 MHz, methanol-d₄) δ 8.55 (s, 4H), 7.29–7.16 (m, 3H), 7.16–7.08 (m, 2H), 6.12 (d, J = 3.4 Hz, 1H), 6.01 (d, J = 3.5 Hz, 1H), 4.42–4.26 (m, 1H), 4.14 (dd, J = 9.9, 5.5 Hz, 1H), 3.97 (s, 2H), 3.84 (d, J = 14.9 Hz, 1H), 3.50 (s, 3H), 2.84 (s, 1H), 2.76–2.62 (m, 2H), 2.53 (s, 3H), 2.03 (d, J = 6.1 Hz, 1H), 1.83–1.43 (m, 2H), 1.34 (t, J = 6.9 Hz, 1H), 0.91 (d, J = 6.6 Hz, 4H), 0.84 (d, J = 6.5 Hz, 3H), 0.45 (d, J = 7.0 Hz, 3H); HRMS [M + H]⁺ calcd for C₂₆H₃₇N₅O₃: 468.2969, found 468.2975.

(5S,11S,Z)-11-Benzyl-5-isobutyl-9-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,4)-pyrrolacyclotridecaphane-4,7,10-trione (16b)

The product was obtained as a mixture of 2 conformations, ratio 2:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.42–7.27 (m, 5H), 7.15–7.06 (m, 1H), 6.65 (d, J = 1.6 Hz, 1H), 5.95 (d, J = 1.6 Hz, 1H), 4.52–4.39 (m, 2H), 4.28 (d, J = 14.9 Hz, 1H), 4.16–4.05 (m, 4H), 3.90 (d, J = 14.9 Hz, 1H), 3.19–3.12 (m, 1H), 3.05–2.99 (m, 1H), 2.68 (s, 3H), 2.66 (s, 1H), 1.78–1.47 (m, 3H), 1.31 (d, J = 17.9 Hz, 1H), 0.96 (d, J = 6.4 Hz, 3H), 0.91 (d, J = 6.4 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 176.2, 173.0, 169.2, 169.1, 133.3, 130.4, 129.3, 129.1, 128.8, 128.6, 127.8, 127.8, 121.4, 118.2, 117.8, 116.6, 115.9, 111.6, 111.1, 104.0, 58.5, 53.0, 49.8, 44.5, 40.9, 38.4, 37.8, 36.9, 35.9, 35.4, 34.9, 34.0, 25.8, 24.7, 24.5, 22.6, 22.2, 21.7, 20.8, 20.5, 18.2; HRMS [M + H]⁺ calcd for C₂₄H₃₃N₅O₃: 440.2656, found 440.2668.

(5S,8S,11S,Z)-11-Benzyl-5-isobutyl-8,9-dimethyl-1¹H-3,6,9,12-tetraaza-1(2,4)-pyrrolacyclotridecaphane-4,7,10-trione (16d)

¹H NMR (400 MHz, methanol-d₄) δ 7.37–7.32 (m, 2H), 7.32–7.28 (m, 1H), 7.18–7.11 (m, 2H), 6.77–6.72 (m, 1H), 6.16 (d, J = 1.6 Hz, 1H), 4.67 (d, J = 14.2 Hz, 1H), 4.14 (d, J = 14.6 Hz, 1H), 4.08 (d, J = 14.5 Hz, 1H), 3.96 (dd, J = 9.5, 5.8 Hz, 1H), 3.80 (dd, J = 9.6, 5.0 Hz, 1H), 3.77–3.68 (m, 2H), 3.25 (dd, J = 13.3, 5.0 Hz, 1H), 2.98–2.84 (m, 1H), 2.70 (s, 3H), 1.88–1.68 (m, 2H), 1.62–1.47 (m, 1H), 0.93 (d, J = 6.7 Hz, 3H), 0.89 (d, J = 6.6 Hz, 3H), 0.75 (d, J = 7.0 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 172.1, 170.2, 163.6, 134.5, 129.3, 129.2, 128.9, 127.5, 122.8, 116.8, 110.0, 55.9, 54.2, 52.6, 41.4, 36.9, 36.0, 35.6, 28.6, 24.8, 22.1, 20.6, 13.7; HRMS [M + Na]⁺ calcd for C₂₅H₃₅N₅O₃: 476.2632, found 454.2630.

(5S,8S,11S,Z)-11-Benzyl-5-isobutyl-1¹,8,9-trimethyl-1¹H-3,6,9,12-tetraaza-1(2,4)-pyrrolacyclotridecaphane-4,7,10-trione (17d)

Note: the synthesis was not efficient, the yield was too low and the product was too unstable that the¹³C NMR could not be achieved. The product was obtained as a mixture of 2 conformations, ratio 5:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.29–7.16 (m, 5H), 7.11 (d, J = 7.1 Hz, 2H), 6.44 (s, 1H), 5.84 (d, J = 2.0 Hz, 1H), 5.34 (t, J = 4.7 Hz, 2H), 4.20 (q, J = 7.5, 6.6 Hz, 1H), 4.11 (dd, J = 9.6, 5.9 Hz, 1H), 3.78 (d, J = 14.4 Hz, 1H), 3.69 (d, J = 12.0 Hz, 1H), 3.64 (d, J = 2.2 Hz, 1H), 3.58 (dd, J = 18.6, 3.1 Hz, 1H), 3.57 (d, J = 18.4 Hz, 2H), 3.35 (s, 3H), 2.80 (s, 3H), 2.19 (t, J = 7.5 Hz, 1H), 1.92 (d, J = 10.3 Hz, 1H), 1.85 (ddd, J = 14.7, 9.6, 5.7 Hz, 1H), 1.79–1.67 (m, 1H), 1.02 (d, J = 7.1 Hz, 3H), 0.96 (d, J = 6.7 Hz, 3H), 0.91 (d, J = 6.3 Hz, 3H); HRMS [M + H]⁺ calcd for C₂₆H₃₇N₅O₃: 468.2969, found 468.2975.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8-methyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-ttrione (18a)

¹H NMR (400 MHz, methanol-d₄) δ 7.39–7.21 (m, 5H), 6.09 (dd, J = 3.1, 0.8 Hz, 1H), 6.00 (d, J = 3.1 Hz, 1H), 4.58 (d, J = 15.2 Hz, 2H), 4.37 (q, J = 7.0 Hz, 1H), 4.11–4.01 (m, 1H), 4.05 (d, J = 15.1 Hz, 1H), 3.76 (d, J = 14.2 Hz, 1H), 3.26 (dd, J = 8.5, 4.5 Hz, 1H), 3.18 (dd, J = 13.9, 4.9 Hz, 2H), 2.82 (dd, J = 13.6, 8.5 Hz, 1H), 1.85–1.67 (m, 2H), 1.66–1.52 (m, 1H), 1.25 (d, J = 7.0 Hz, 3H), 0.95 (d, J = 6.6 Hz, 3H), 0.92 (d, J = 6.5 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 173.9, 172.2, 171.5, 151.8, 150.3, 136.7, 128.3, 127.6, 125.8, 108.2, 106.6, 61.2, 52.0, 48.8, 44.0, 37.6, 37.2, 34.9, 24.0, 21.2, 20.0, 16.2; HRMS [M + H]⁺ calcd for C₂₄H₃₂N₄O₄: 441.2496, found 441.2508.

(5S,11S)-11-Benzyl-5-isobutyl-9-methyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (18b)

The product was obtained as a mixture of 2 different conformations, ratio 5:3 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.33–7.18 (m, 4H), 7.13–7.06 (m, 1H), 6.13 (dd, J = 3.1, 0.8 Hz, 1H), 6.07 (d, J = 3.1 Hz, 1H), 4.55 (d, J = 15.0 Hz, 2H), 4.36 (d, J = 14.4 Hz, 1H), 4.18–4.07 (m, 1H), 4.07–3.95 (m, 1H), 3.94–3.81 (m, 2H), 3.54–3.37 (m, 2H), 2.98 (d, J = 14.4 Hz, 1H), 2.91 (dd, J = 13.2, 6.6 Hz, 1H), 2.81 (dd, J = 13.2, 7.6 Hz, 1H), 2.67 (s, 3H), 1.76–1.49 (m, 5H), 0.97 (d, J = 6.5 Hz, 3H), 0.91 (d, J = 6.5 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 175.1, 173.2, 170.0, 152.4, 151.4, 151.2, 137.3, 129.1, 129.1, 128.9, 128.9, 128.3, 128.0, 128.0, 126.7, 126.5, 108.7, 108.7, 107.5, 106.9, 58.2, 53.6, 51.5, 51.4, 44.8, 41.2, 40.9, 39.4, 39.4, 38.2, 37.0, 36.1, 35.6, 35.3, 33.8, 24.7, 24.6, 21.9, 21.8, 20.6, 20.5; HRMS [M + H]⁺ calcd for C₂₄H₃₂N₄O₄: 441.2496, found 441.2508.

(5S,8S,11S)-11-Benzyl-5-isobutyl-6,8-dimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (18c)

A mixture of 2 different conformations, ratio 5:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.34 (dq, J = 7.2, 1.2 Hz, 1H), 7.34–7.25 (m, 3H), 7.29–7.21 (m, 1H), 6.10 (d, J = 3.1 Hz, 1H), 5.99 (d, J = 3.0 Hz, 1H), 4.71 (dd, J = 14.5, 8.2 Hz, 2H), 3.86 (d, J = 15.1 Hz, 1H), 3.78 (d, J = 14.3 Hz, 1H), 3.61 (dd, J = 8.1, 5.9 Hz, 1H), 3.24 (s, 3H), 3.23–3.17 (m, 2H), 3.13 (d, J = 14.3 Hz, 1H), 3.03–2.89 (m, 1H), 2.85–2.72 (m, 1H), 2.09–1.97 (m, 1H), 1.78 (ddd, J = 13.9, 8.0, 5.9 Hz, 1H), 1.63 (dq, J = 13.0, 6.5 Hz, 1H), 1.20 (d, J = 6.8 Hz, 3H), 1.01 (t, J = 6.4 Hz, 1H), 0.97 (d, J = 6.5 Hz, 3H), 0.94 (d, J = 6.6 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 172.0, 170.9, 170.0, 168.9, 153.0, 151.2, 137.6, 129.1, 129.0, 128.3, 126.5, 108.8, 108.3, 63.6, 62.6, 45.3, 45.3, 38.6, 37.4, 37.1, 35.6, 28.9, 25.1, 22.1, 21.3, 16.8; HRMS [M + H]⁺ calcd for C₂₅H₃₄N₄O₄: 455.2653, found 455.2664.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8,9-dimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (18d)

¹H NMR (400 MHz, methanol-d₄) δ 8.46 (s, 1H), 7.39–7.26 (m, 2H), 7.26–7.18 (m, 1H), 7.18–7.02 (m, 2H), 6.35 (d, J = 3.0 Hz, 1H), 6.20 (d, J = 3.0 Hz, 1H), 4.53 (d, J = 15.1 Hz, 1H), 4.46 (dd, J = 9.5, 5.9 Hz, 1H), 4.16 (d, J = 15.1 Hz, 1H), 4.01 (q, J = 14.9 Hz, 2H), 3.88 (q, J = 6.8 Hz, 1H), 3.65 (dd, J = 10.5, 4.1 Hz, 1H), 3.35 (d, J = 4.1 Hz, 1H), 2.75 (dd, J = 12.6, 10.5 Hz, 1H), 2.61 (s, 3H), 1.72–1.49 (m, 3H), 0.94 (d, J = 6.4 Hz, 3H), 0.88 (d, J = 6.4 Hz, 3H), 0.45 (d, J = 6.9 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 174.1 (C), 172.6 (C), 170.8 (C), 151.8 (C), 136.7 (C), 129.1 (CH), 128.7 (CH), 126.8 (CH), 109.1 (CH), 107.0 (CH), 56.5 (CH), 55.5 (CH), 51.8 (CH), 41.1 (CH₂), 38.5 (CH₂), 36.7 (CH₂), 35.7 (CH₂), 28.7 (CH₃), 24.7 (CH), 22.0 (CH₃), 20.5 (CH₃), 13.1 (CH₃); HRMS [M + H]⁺ calcd for C₂₅H₃₄N₄O₄: 455.2653, found 455.2665.

(5S,8S,11S)-11-Benzyl-5-isobutyl-8,12-dimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (18e)

¹H NMR (400 MHz, methanol-d₄) δ 7.34–7.27 (m, 4H), 7.27–7.18 (m, 1H), 6.13 (dd, J = 3.1, 1.1 Hz, 1H), 5.88 (dd, J = 3.1, 0.7 Hz, 1H), 4.6 (s, 1H), 4.49 (q, J = 7.0 Hz, 1H), 4.12 (t, J = 7.6 Hz, 1H), 4.07 (d, J = 15.3 Hz, 1H), 3.50 (dd, J = 9.2, 4.8 Hz, 1H), 3.44 (d, J = 13.8 Hz, 1H), 3.27 (d, J = 4.7 Hz, 1H), 3.17 (d, J = 13.8 Hz, 1H), 2.95 (dd, J = 14.3, 9.3 Hz, 1H), 2.48 (s, 3H), 1.80–1.71 (m, 2H), 1.59 (dh, J = 13.3, 6.7 Hz, 1H), 1.33 (d, J = 7.1 Hz, 3H), 0.96 (d, J = 6.6 Hz, 3H), 0.93 (d, J = 6.5 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 173.1 (C), 172.8 (C), 172.1 (C), 151.5 (C), 151.2 (C), 140.0 (C), 129.1 (CH), 128.1 (CH), 126.1 (CH), 110.1 (CH), 107.3 (CH), 63.9 (CH), 52.4 (CH), 51.5 (CH), 49.7 (CH), 38.3 (CH₂), 36.3 (CH₃), 35.9 (CH₂), 31.7 (CH₂), 24.8 (CH), 21.9 (CH₃), 20.8 (CH₃), 16.8 (CH₃); HRMS [M + H]⁺ calcd for C₂₅H₃₄N₄O₄: 455.2653, found 455.2644.

(5S,8S,11S)-11-Benzyl-5-isobutyl-3,8-dimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (19a)

The product was obtained as a mixture of 2 different conformations, ratio 5:3 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.38–7.21 (m, 5H), 6.22 (dd, J = 3.1, 1.1 Hz, 1H), 6.00 (dd, J = 3.1, 0.9 Hz, 1H), 5.49 (d, J = 14.9 Hz, 1H), 4.76 (dd, J = 7.8, 6.2 Hz, 1H), 4.36 (p, J = 7.2 Hz, 1H), 3.76 (d, J = 13.9 Hz, 1H), 3.64 (d, J = 14.9 Hz, 1H), 3.27–3.01 (m, 4H), 2.83 (dd, J = 13.3, 8.3 Hz, 1H), 2.72 (s, 3H), 1.97–1.66 (m, 1H), 1.64–1.48 (m, 3H), 1.29 (d, J = 7.3 Hz, 3H), 0.95 (dt, J = 6.4, 2.2 Hz, 6H). ¹³C NMR (101 MHz, methanol-d₄) δ 174.6, 174.5, 173.2, 171.8, 171.5, 171.0, 152.4, 152.0, 150.2, 149.4, 137.5, 137.2, 129.1, 128.9, 128.3, 128.3, 126.6, 126.6, 109.9, 109.8, 109.2, 108.0, 63.5, 60.4, 50.4, 50.2, 45.8, 45.8, 44.3, 43.2, 41.0, 40.1, 39.1, 38.4, 33.2, 32.6, 24.4, 24.3, 22.0, 22.0, 21.4, 21.4, 18.4, 16.9; HRMS [M + H]⁺ calcd for C₂₅H₃₄N₄O₄: 455.2653, found 455.2664.

(5S,11S)-11-Benzyl-5-isobutyl-3,9-dimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (19b)

The product was obtained as a mixture of 3 different conformations, ratio 9:6:5 determined by ¹H NMR. Note: No attempts gave adequately resolved¹H and ¹³C NMR signals for unambiguous structural determination of the major and minor conformations. ¹H NMR (400 MHz, methanol-d₄) δ 7.41–6.97 (m, 5H), 6.37–6.08 (m, 2H), 5.62–5.07 (m, 1H), 4.60 (s, 3H), 4.23–3.57 (m, 4H), 3.04–2.88 (m, 2H), 2.88–2.81 (m, 2H), 2.80–2.73 (m, 2H), 2.58–2.37 (m, 1H), 2.25–1.92 (m, 1H), 1.82–1.47 (m, 2H), 1.82–1.47 (m, 3H), 1.33 (s, 3H), 1.00–0.89 (m, 6H). ¹³C NMR (101 MHz, methanol-d₄) δ 13C NMR (101 MHz, MeOD) δ 174.0, 171.5, 171.2, 168.0, 167.4, 163.7, 153.1, 152.8, 152.8, 151.2, 150.1, 149.1, 136.8, 136.6, 129.0, 128.9, 128.8, 128.5, 128.3, 128.2, 128.1, 126.9, 126.7, 126.7, 109.6, 109.1, 51.6, 51.1, 45.7, 43.7, 42.6, 41.8, 41.3, 39.8, 39.3, 38.4, 36.7, 34.5, 34.1, 34.0, 32.3, 24.4, 24.4, 22.2, 22.0, 21.9, 21.4, 21.2, 21.0; HRMS [M + H]⁺ calcd for C₂₅H₃₄N₄O₄: 455.2653, found 455.2667.

(5S,8S,11S)-11-Benzyl-5-isobutyl-3,6,8-trimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (19c)

The product was obtained as a mixture of 3 different conformations, ratio 10:6:3. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.45–7.17 (m, 5H), 6.22 (d, J = 3.1 Hz, 1H), 5.99 (d, J = 3.1 Hz, 1H), 5.40 (d, J = 14.9 Hz, 1H), 4.75–4.62 (m, 1H), 3.75 (d, J = 14.2 Hz, 1H), 3.62 (d, J = 14.9 Hz, 1H), 3.22 (dd, J = 4.0, 2.5 Hz, 1H), 3.19 (dd, J = 4.3, 2.4 Hz, 1H), 3.16 (s, 3H), 2.67 (s, 3H), 2.13–1.92 (m, 1H), 1.55 (dddt, J = 42.8, 36.1, 13.6, 6.6 Hz, 2H), 1.31–1.23 (m, 4H), 1.19 (d, J = 6.7 Hz, 2H), 1.05–0.94 (m, 6H). ¹³C NMR (101 MHz, methanol-d₄) δ 174.5, 171.7, 170.5, 168.9, 165.8, 164.3, 162.7, 153.1, 150.1, 137.6, 137.1, 129.0, 128.9, 128.9, 128.3, 128.3, 126.6, 126.6, 110.2, 110.1, 109.2, 62.9, 61.4, 45.9, 44.8, 43.7, 43.4, 43.4, 39.0, 38.5, 38.0, 37.7, 33.9, 33.0, 24.6, 24.6, 21.8, 21.7, 21.6, 16.6, 16.4; HRMS [M + H]⁺ calcd for C₂₆H₃₆N₄O₄: 469.2809, found 469.2822.

(5S,8S,11S)-11-Benzyl-5-isobutyl-3,8,9-trimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (19d)

The product was obtained as a mixture of 2 different conformations, ratio 2:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.41–7.15 (m, 3H), 7.16–7.03 (m, 2H), 6.42 (dd, J = 3.2, 0.9 Hz, 1H), 6.39–6.35 (m, 1H), 6.30 (dd, J = 3.1, 1.2 Hz, 1H), 4.71 (d, J = 17.0 Hz, 1H), 4.20 (d, J = 17.0 Hz, 1H), 4.02–3.82 (m, 3H), 3.75–3.57 (m, 3H), 3.08–2.91 (m, 2H), 2.8 (s, 4H), 2.64 (s, 3H), 2.61 (s, 1.5H), 1.86–1.74 (m, 1H), 1.61–1.42 (m, 3H), 1.41–1.21 (m, 1H), 0.93 (dd, J = 6.5, 1.5 Hz, 6H), 0.54 (dd, J = 6.9, 2.2 Hz, 3H). ¹³C NMR (101 MHz, methanol-d₄) δ 173.6, 171.5, 170.7, 169.5, 151.4, 151.1, 150.4, 149.5, 137.1, 136.6, 129.0, 128.9, 128.7, 128.5, 126.8, 126.5, 110.4, 109.6, 109.2, 108.6, 55.9, 55.8, 55.7, 55.1, 45.7, 43.8, 42.4, 41.4, 41.2, 39.9, 37.0, 36.3, 33.9, 32.7, 28.9, 28.6, 24.3, 24.3, 22.2, 22.0, 21.2, 13.8, 13.0; HRMS [M + H]⁺ calcd for C₂₆H₃₆N₄O₄: 469.2809, found 469.2820.

(5S,8S,11S)-11-Benzyl-5-isobutyl-3,8,12-trimethyl-3,6,9,12-tetraaza-1(2,5)-furanacyclotridecaphane-4,7,10-trione (19e)

The product was obtained as a mixture of 2 different conformations, ratio 5:1 determined by ¹H NMR. ¹H NMR (400 MHz, methanol-d₄) δ (only major conformation) 7.59–6.76 (m, 5H), 6.23 (dd, J = 3.1, 1.2 Hz, 1H), 5.80 (d, J = 3.1 Hz, 1H), 5.52 (d, J = 15.0 Hz, 1H), 4.80 (dd, J = 8.1, 6.1 Hz, 1H), 4.59 (q, J = 7.3 Hz, 1H), 3.67 (d, J = 15.1 Hz, 1H), 3.49 (d, J = 13.5 Hz, 1H), 3.44 (dd, J = 9.6, 4.2 Hz, 1H), 3.10 (d, J = 13.6 Hz, 1H), 2.93 (dd, J = 14.4, 9.5 Hz, 1H), 2.69 (s, 3H), 2.52 (s, 3H), 1.78–1.66 (m, 1H), 1.67–1.51 (m, 4H), 1.35 (d, J = 7.2 Hz, 3H), 0.97 (dd, J = 6.5, 1.4 Hz, 6H). ¹³C NMR (101 MHz, methanol-d₄) δ 172.9, 172.7, 171.6, 151.6, 150.5, 140.2, 129.2, 129.0, 128.1, 128.1, 126.2, 126.1, 110.7, 109.2, 62.6, 51.4, 50.1, 43.0, 40.1, 36.4, 32.9, 31.6, 24.6, 24.4, 22.0, 21.4, 17.7, 17.2; HRMS [M + H]⁺ calcd for C₂₆H₃₆N₄O₄: 469.2809, found 469.2820.

Glossary

Abbreviations

2-CTC

2-chlorotrityl chloride

Boc

tert-butyloxycarbonyl

DABCO

1,4-diazabicyclo[2.2.2]octane

DCM

dichloromethane

DEPBT

3-(diethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one

DIBAL-H

diisobutylaluminum hydride

DMAP

4-dimethylaminopyridine

DMC

dimethylcarbonate

EWG

electroweakening group

HBA

H-bond acceptor

HBD

H-bond donor

HMF

5-hydroxymethyl furfural

IMHB

intramolecular H-bond

MOE

Molecular Operating Environment

MW

molecular weight

NMR

nuclear magnetic resonance

PAMPA

parallel artificial membrane permeability assay

PK-ADME

pharmacokinetics–absorption, distribution, metabolism, excretion

PyBOP

benzotriazol-1-yloxytripyrrolidinophosphonium hexafluorophosphate

THF

tetrahydrofuran

bRo5

beyond the rule-of-five

tPSA

total polar surface area

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jmedchem.3c02162.

• (Table S1) Molecular formula strings; (Figure S1) multiple and distinguishable conformations of 14e in various kinds of salts; (Figure S2) a plausible “conformation-inducing effect” of the compound 18a; (Table S2) influence of N-methylation and peptoids on permeability of furan macrocycles; (Table S3) raw PAMPA results; 1H, 13C NMR spectra, and UPLC chromatograms; (Table S4) 2D NMR analysis for the characterization of compound 14a; (Table S5) 2D NMR analysis for the characterization of compound 16d; (Table S6) 2D NMR analysis for the characterization of compound 18a; (Table S7) 2D NMR analysis for the characterization of compound 18d; and (Table S8) 2D NMR analysis for the characterization of compound 18e (PDF)

• SMILE data (CSV)

The authors declare no competing financial interest.