molecules Article
Traceless Solid-Phase Synthesis of [6,7,8 + 5,6,7]-Fused Molecular Frameworks 1,2, * Vanesa Giménez-Navarro 1 and Viktor Krchnák ˇ 1 2
*
ID
Department of Organic Chemistry, Faculty of Science, Palacky University, 17. Listopadu 12, 771 46 Olomouc, Czech Republic;
[email protected] Department of Chemistry and Biochemistry, University of Notre Dame, 251 Nieuwland Science Center, University of Notre Dame, Notre Dame, IN 46556, USA Correspondence:
[email protected]; Tel.: +1-574-631-5113
Received: 14 April 2018; Accepted: 2 May 2018; Published: 4 May 2018
Abstract: We report two synthetic strategies for traceless solid-phase synthesis of molecular scaffolds comprising 6- to 8-membered rings fused with 5- to 7-membered rings. Traceless synthesis facilitated preparation of target molecules without any trace of polymer-supported linkers. The cyclization proceeded via acid-mediated tandem N-acylium ion formation followed by the nucleophilic addition of O- and C-nucleophiles. The presented synthetic strategy enabled, through the use of simple building blocks without any conformational preferences, the evaluation of the predisposition of different combinations of ring sizes to form fused ring molecular scaffolds. Compounds with any combination of [6,7 + 5,6,7] ring sizes were accessible with excellent crude purity. The 8-membered cyclic iminium was successfully fused only with the 5-membered cycle and larger fused ring systems were not formed, probably due to their instability. Keywords: bicyclic compounds; heterocycles; iminiums; scaffold; solid-phase synthesis
1. Introduction The enormous structural diversity of natural products comprising fused ring systems and, in particular, their wide range of biological activities, make them attractive synthetic targets. More than 60% of the drugs on the market have structures derived from natural origins [1,2]. Therefore, the search for pharmacologically relevant molecules that mimic natural products has become an integral part of drug discovery. Specific structural features of these molecular scaffolds are the presence of sp3 hybridized carbons and stereogenic centers [3]. Although these characteristics are typical of natural products, they are often missing in traditional compound collections for High Throughput Screening (HTS) [4]. Generally, the chemical space occupied by natural products overlays with biological space. Therefore, the exploitation of core structures derived from bioactive natural products has led to the design and implementation of biology-oriented synthesis (BIOS) [5,6] instead of synthesis focused only on the structural dissimilarity of compounds, i.e., diversity-oriented synthesis (DOS) [6,7]. Natural products comprising fused and bridged carbo- and heterocycles represent particularly intriguing molecular scaffolds with 3D architecture. Among the various chemical routes used to synthesize diverse nitrogenous heterocycles, tandem iminium ion cyclization-nucleophilic addition represents one of the most versatile strategies [8–11]. Typically, a cyclic iminium is formed from an amide nitrogen and an aldehyde attached to acyclic intermediate. Then, a cyclic N-acyliminium ion is transformed into fused or bridged structures by intramolecular nucleophilic addition [12,13]. Specifically, oxygen has previously served as an internal nucleophile in the preparation of five- [14,15], six- [16–19], seven- [13,20] and eight-membered lactam rings [13]. Molecules 2018, 23, 1090; doi:10.3390/molecules23051090
www.mdpi.com/journal/molecules
Molecules 2018, 23, 1090
2 of 12
Incorporation of fused lactam rings was also frequently applied for the synthesis Molecules 2018, 23, x FOR PEER REVIEW 2 of 11 of peptidomimetics [8,21]. The constrained structures were designed to mimic β-turns, the most Incorporation of fused lactam rings was Amino also frequently applied for the within synthesis of frequent secondary structure found in proteins. acid motifs constrained a bicyclic peptidomimetics [8,21]. The constrained structures were designed to mimic β‐turns, the most lactam architecture have been applied to the design of modified peptides and peptidomimetics to frequent secondary structure found in proteins. Amino acid motifs constrained within a bicyclic discover new therapeutics and better understand the interactions of ligands with target enzymes and lactam architecture have been applied to the design of modified peptides and peptidomimetics to receptors. Compounds I [19,22] and II [23] (Figure 1) were designed as β-turn mimetics, and scaffold discover new therapeutics and better understand the interactions of ligands with target enzymes and III [24] was synthesized as a conformationally constrained analogue for the Phe7 -Phe8 region of receptors. Compounds I [19,22] and II [23] (Figure 1) were designed as β‐turn mimetics, and scaffold substance P. The of constrained dipeptide surrogates [6 + 5]-fused III [24] was solution-phase synthesized as synthesis a conformationally constrained analogue for the comprising Phe7‐Phe8 region of substance P. The of constrained dipeptide [6 + 5]‐ lactams (structure IV)solution‐phase synthesis from 3-aza-1,5-diketoacids and amino alcoholssurrogates comprising was developed using a one-pot fused lactams (structure IV) from 3‐aza‐1,5‐diketoacids and amino alcohols was developed using a strategy [17]. Tricyclic constrained scaffold V was reported as a carbapenem mimetic [25]. Nielsen one‐pot strategy [17]. Tricyclic constrained scaffold V was reported as a carbapenem mimetic [25]. et al. synthesized a range of bicyclic dipeptide mimetics forming 5-hydroxylactam/N-acyliminium Nielsen et al. [15]. synthesized a range of amino bicyclic acid dipeptide mimetics forming 5‐hydroxylactam/N‐ ion intermediates Depending on the side-chain, the cyclization occurs as side-chain acyliminium ion intermediates [15]. Depending on the amino acid side‐chain, the cyclization occurs cyclization VI if a nucleophilic moiety (X = O, S, NH, n = 1–3) is presented in the side-chain and as side‐chain cyclization VI if a nucleophilic moiety (X = O, S, NH, n = 1–3) is presented in the side‐ as backbone cyclization VII in absence of heteroatomic moieties in the side-chain. Their studies chain and as backbone cyclization VII in absence of heteroatomic moieties in the side‐chain. Their on N-acyliminium Pictet-SpenglerPictet‐Spengler reactions [26] were extended synthesize pyrroloisoquinolines studies on N‐acyliminium reactions [26] to were extended to synthesize VIII [27] or with an indole moiety and heterocycles as and furans and thiophenes yieldand tri- and pyrroloisoquinolines VIII [27] or with an indole such moiety heterocycles such as to furans tetracyclic scaffolds [28]. Additionally N-carbamyliminium Pictet-Spengler reaction was used to form thiophenes to yield tri‐ and tetracyclic scaffolds [28]. Additionally N‐carbamyliminium Pictet‐ Spengler reaction was used to form tetrahydro‐β‐carbolines and tetrahydroisoquinolines [29]. tetrahydro-β-carbolines and tetrahydroisoquinolines [29]. Synthesis of fused natural product-like Synthesis of fused natural different product‐like comprising different ring combination diketopiperazines comprising ringdiketopiperazines combination yielded molecular scaffolds IX [30]. yielded molecular scaffolds IX [30]. Recently, our group reported a synthetic strategy enabling the incorporation of bicyclic scaffolds Recently, our group reported a synthetic strategy enabling the incorporation of bicyclic scaffolds as peptide constraints during traditional peptide synthesis. The synthesis of as peptide constraints during traditional peptide synthesis. The synthesis of bridged 6‐oxa‐3,8‐ bridged 6-oxa-3,8-diazabicyclo[3.2.1]octan-2-ones [31], fused tetrahydropyrazino[2,1-b][1,3]oxazinediazabicyclo[3.2.1]octan‐2‐ones [31], fused tetrahydropyrazino[2,1‐b][1,3]oxazine‐4,7(6H,8H)‐diones 4,7(6H,8H)-diones X (X = O, n = 1) [32], hexahydro-4H-pyrazino[1,2-a]pyrimidine-4,7(6H)-diones X (X = O, n = 1) [32], hexahydro‐4H‐pyrazino[1,2‐a]pyrimidine‐4,7(6H)‐diones X (X = NR3, n = 1) [32] 3 X (X =and NRhexahydropyrimido[1,2‐d][1,4]diazepine‐4,7(1H,6H)‐diones , n = 1) [32] and hexahydropyrimido[1,2-d][1,4]diazepine-4,7(1H,6H)-diones X (X = NR3 , X (X = NR3, n = 2) [33] were n = 2)performed [33] were via performed viadirection westbound direction (towards the amino terminus) N-acyliminium westbound (towards the amino terminus) N‐acyliminium cyclization– cyclization–nucleophilic addition with Ser, Thr and 2,4-diaminobutyric acid as the sources of nucleophilic addition with Ser, Thr and 2,4‐diaminobutyric acid as the sources of oxygen, sulfur and nitrogen The solid‐phase The synthesis of tetrahydro‐2H‐oxazolo[3,2‐ oxygen, sulfurnucleophiles, and nitrogenrespectively. nucleophiles, respectively. solid-phase synthesis of tetrahydro-2Ha]pyrazine‐5(3H)‐ones IV (R2 = H) [34] was carried out via eastbound direction (towards the carboxyl oxazolo[3,2-a]pyrazine-5(3H)-ones IV (R2 = H) [34] was carried out via eastbound direction (towards terminus) cyclization, and the scope of this synthetic strategy was expanded recently to yield [7,8,9 + the carboxyl terminus) cyclization, and the scope of this synthetic strategy was expanded recently to 5]‐fused rings from Ser, Cys and C‐nucleophiles (structure XI) [13]. yield [7,8,9 + 5]-fused rings from Ser, Cys and C-nucleophiles (structure XI) [13]. R1
X
R1 2
N
H2N
R O X = O, S I
H2N
1
O
O N
X
O R1
AA1 OH H
O N
2N
R
OH
O
R1
H
n
O
R3
O IV
1 HN R
O R1
O
N
H
O
X = O, S, NH
O
VIII
VII
VI
n N N
IX
OH
O
H O
O
N
R2
O N
R3
O
X
O
Ph
R2 N
III
O NH
R1
N
N
H
V
Ph
N
R2
OH
O
O
II
O O
R
N
H N
R1 O
R2 X
R1 R
X = O, S, NR3
3
n N
X R2
m N O
O
XI
Figure 1. Reported synthesized bicyclic and tricyclic lactams. Figure 1. Reported synthesized bicyclic and tricyclic lactams.
R4
Molecules 2018, 23, x FOR PEER REVIEW Molecules 2018, 23, 1090
3 of 11 3 of 12
In this report, we describe two synthetic strategies for traceless solid‐phase synthesis of bicyclic scaffolds. Traceless synthesis provided access to compounds without any trace of a linker, a critical In this report, we describe two synthetic strategies for traceless solid-phase synthesis of bicyclic feature for preparation of compounds for structure‐activity synthesis was scaffolds. Traceless synthesis provided access to compoundsrelationship without any(SAR). The trace of a linker, a critical carried out in a modular fashion that allowed us to test any combination of potential ring sizes, with feature for preparation of compounds for structure-activity relationship (SAR). The synthesis was the first out ring by fashion a cyclic N‐acyliminium and the second one fused addition of carried in formed a modular that allowed us to species, test any combination of potential ringby sizes, with the oxygen and carbon nucleophiles, and to obtain [6,7,8 + 5,6,7]‐fused ring systems. first ring formed by a cyclic N-acyliminium species, and the second one fused by addition of oxygen and carbon nucleophiles, and to obtain [6,7,8 + 5,6,7]-fused ring systems. 2. Results 2. Results To prepare compounds on solid phase in a traceless manner we selected two potential anchoring To prepare compounds on solid phase in a traceless manner we selected two potential anchoring functional groups that enabled immobilization of the first building blocks and polymer‐supported functionalof groups that enabled immobilization of the first building blocks and polymer-supported synthesis the acyclic precursor. The selected functional groups were not involved in any synthesis of the acyclic precursor. The selected functional groups were not involved in any transformation during solid‐phase synthesis, however, they participated in ring‐closing reactions (C‐ transformation during solid-phase synthesis, however, they participated in ring-closing reactions N and C‐O/C‐C bond formations) in the final cleavage/cyclization step. The attachment to the resin (C-Nachieved and C-O/C-C bond formations) in served the finalas cleavage/cyclization step. Thegroup. attachment the resin was via specific linkers that polymer‐bound protecting Both tostrategies was achieved via specific linkers that served as polymer-bound protecting group. Both strategies using using either amide nitrogen or nucleophile for attachment to solid support were used (Figure 2, L either amide or nucleophile for attachment to solid released support were used (Figure L stands stands for an nitrogen acid‐labile linker). Acid‐mediated cleavage compounds from 2,the resin, for an acid-labile linker). Acid-mediated cleavage released compounds from the resin, the demasked the aldehyde and triggered cyclic N‐acyliminium formation followed by demasked nucleophilic aldehyde and triggered cyclic N-acyliminium formation followed by nucleophilic addition. addition.
Figure 2. Two ways for immobilization of acyclic intermediates. Figure 2. Two ways for immobilization of acyclic intermediates.
To perform the synthesis, we used simple and commercially available building blocks: amino To perform the synthesis, we used simple commercially available building blocks: amino alcohols (Fmoc‐glycinol, Fmoc‐β‐alaninol and and 4‐(Fmoc‐amino)butanol), α‐bromocarboxylic acids alcohols (Fmoc-glycinol, Fmoc-β-alaninol and 4-(Fmoc-amino)butanol), α-bromocarboxylic acids (bromoacetic acid and (S)‐ and (R)‐bromopropionic acids), protected amino aldehydes (containing (bromoacetic acid and (S)and (R)-bromopropionic acids), protected amino aldehydes (containing one, one, two and three‐carbon spacers) and sulfonyl chlorides or aryl fluorides. Because each ring is two and three-carbon spacers) and sulfonyl chlorides or aryl fluorides. Because each ring is formed formed from different building blocks, this strategy allowed us to use an identical synthetic strategy from different building blocks, this strategy allowed us to use an identical synthetic strategy to prepare to prepare any combination of different ring sizes in a truly combinatorial fashion. The size of the any combination of different ring sizes in a truly combinatorial fashion. The size of the first ring, first ring, the cyclic N‐acyliminium, can be controlled by different amino acids (α, β, γ) and by the the cyclic N-acyliminium, can be controlled by different amino acids (α, β, γ) and by the length of the length of the carbon spacer bearing the protected aldehyde. In this study, we focused on varying the carbon spacer bearing the protected aldehyde. In this study, we focused on varying the length of the length of the aldehyde spacer. aldehyde spacer. of resin‐bound acyclic precursors was accomplished using well‐documented The synthesis The synthesis of resin-bound acyclic precursors was accomplished using well-documented transformations. Briefly, the synthesis began with attachment via the oxygen atom of Fmoc‐protected transformations. Briefly, the synthesis began with attachment via the oxygen atom of Fmoc-protected amino alcohols to Wang resin 1 by using trichloroacetimidate activation [35] to yield resin 2 (Scheme amino alcohols to Wang resin 1 by using trichloroacetimidate activation [35] to yield resin 2 (Scheme 1). 1). The Fmoc group was cleaved by piperidine, and the polymer‐supported amines were acylated The Fmoc group was cleaved by piperidine, and the polymer-supported amines were acylated with with α‐bromocarboxylic acids to afford resin 3. Nucleophilic substitution of bromine with different α-bromocarboxylic acids to afford resin 3. Nucleophilic substitution of bromine with different protected amino aldehydes provided resin‐bound secondary amines. To obtain resin 4, protected the final amino aldehydes provided resin-bound secondary amines. To obtain resin 4, the final derivatization derivatization was performed using 4‐nitrobenzenesulfonyl chloride (Ns‐Cl), 4‐methylbenzene‐ was performed using 4-nitrobenzenesulfonyl chloride (Ns-Cl), 4-methylbenzene-sulfonyl chloride sulfonyl chloride (Tos‐Cl) or 4‐fluoro‐3‐nitrobenzotrifluoride. TFA exposure of the polymer‐ (Tos-Cl) or 4-fluoro-3-nitrobenzotrifluoride. TFArelease exposure of the supported acyclic precursors triggered their from the polymer-supported resin, the removal acyclic of the precursors aldehyde triggered their release from the resin, the removal of the aldehyde protecting group, and the formation protecting group, and the formation of the six‐, seven‐ and eight‐membered N‐acyliminium ions via of the six-, seven- and eight-membered N-acyliminium ions via hemiaminals 5, followed by internal hemiaminals 5, followed by internal nucleophilic attack to provide target molecular scaffolds 6. We nucleophilic attack to provide target molecular scaffolds 6. We typically used 50% TFA in DCM in the typically used 50% TFA in DCM in the final step. However, the LC/MS analysis of crude compound final step. However, the LC/MS analysis of crude compound 6{3,2,1,3} revealed partial decomposition, 6{3,2,1,3} revealed partial decomposition, i.e., the purity and yield of the crude products were only i.e., the purity yield of thefrom crude products were only 38% 16%, respectively, from in theDCM cleavage 38% and 16%, and respectively, the cleavage cocktail. We and therefore used 10% TFA for cocktail. We therefore used 10% TFA in DCM for cleavage followed by extraction of DCM/TFA with cleavage followed by extraction of DCM/TFA with aqueous sodium bicarbonate. When the organic layer was separated and DCM evaporated under a stream of nitrogen, the purity substantially
Molecules 2018, 23, 1090
4 of 12
aqueous sodium bicarbonate. When Molecules 2018, 23, x FOR PEER REVIEW
the organic layer was separated and DCM evaporated under 4 of 11 a stream of nitrogen, the purity substantially increased to 88%. Individual compounds 6 possess four increased to 88%. Individual compounds 6 possess four points of diversification: m and n refer to the points of diversification: m and n refer to the ring sizes and R1 and R2 are the substituents at carbon 1 and R2 are the substituents at carbon Cα and amine nitrogen, respectively (Table 1). ring sizes and R Cα and amine nitrogen, respectively (Table 1).
Scheme 1. Polymer‐supported synthesis of fused bicyclic compounds 6. Reagents and conditions: (i) Scheme 1. Polymer-supported synthesis of fused bicyclic compounds 6. Reagents and conditions: ◦ C,min, trichloroacetonitrile, anhydrous DCM, 0 °C, DBU, DBU, DCM, DCM, rt, 1 h; and Fmoc‐ (i) trichloroacetonitrile, anhydrous DCM, 0 30 30 min, rt,resin 1 h; washed resin washed and Fmoc-amino alcohol, · Et O, anhydrous THF added, rt, 30 min; (ii) 50% piperidine/DMF, rt, 15 min; amino alcohol, BF 3∙EtBF 2O, anhydrous THF added, rt, 30 min; (ii) 50% piperidine/DMF, rt, 15 min; (iii) 3 2 (iii) α-bromocarboxylic DIC, DCM, 5 min, then DIUfiltered, filtered,DIEA DIEAadded addedand and the the solution solution α‐bromocarboxylic acid, acid, DIC, DCM, 5 min, then DIU transferred to the reaction vessel with resin, rt, 1 h; (iv) amino-aldehyde dialkyl acetal, DIEA, DMF, transferred to the reaction vessel with resin, rt, 1 h; (iv) amino‐aldehyde dialkyl acetal, DIEA, DMF, rt, 2 h; (v) Ns-Cl or Tos-Cl, DIEA, DCM, rt, 2 h or 4-fluoro-3-nitrobenzotrifluoride, DIEA, DMSO, rt, rt, 2 h; (v) Ns‐Cl or Tos‐Cl, DIEA, DCM, rt, 2 h or 4‐fluoro‐3‐nitrobenzotrifluoride, DIEA, DMSO, rt, overnight; (vi) TFA/DCM 1:1, rt, 1 h; (vii) MeOH, rt, overnight. overnight; (vi) TFA/DCM 1:1, rt, 1 h; (vii) MeOH, rt, overnight. Table 1. Synthesized fused heterocyclic compounds 6. Table 1. Synthesized fused heterocyclic compounds 6. a RR22 Rings Purity Purity [%] Yield [%] b Entry * Entry Compound m m n n R1 R1 * Compound Rings [%] a Yield [%] b 6{1,2,1,1} 1 1 2 H Ns 7 + 5 94 70 1 6{1,2,1,1} 1 2 H Ns 7+5 94 70 6{1,2,1,2} 2 Tos 70 2 6{1,2,1,2} 1 1 2 2 H H Tos 7 +7 + 5 5 93 93 70 36{1,2,1,3} 6{1,2,1,3} 1 1 2 2 H H 2-NO -C66H 5 68 68 64 2 -4-CF33‐C 2‐4‐CF H3 3 7 +7 + 5 64 3 2‐NO 3 H Ns 8+5 86 66 4 6{1,3,1,1} 1 4 Ns 66 56{1,3,1,1} 6{1,3,1,2} 1 1 3 3 H H Tos 8 +8 + 5 5 53 86 21 66{1,3,1,2} 6{2,1,1,3} 1 2 3 1 H H 2-NO2Tos -4-CF3 -C6 H3 6 +8 + 5 6 97 53 62 5 21 7 6{2,2,1,1} Ns 7+6 98 83 6{2,1,1,3} 3‐C6H3 6 + 6 97 62 6 2 2 1 2 H H 2‐NO2‐4‐CF 8 6{2,2,1,2} 2 2 H Tos 7+6 97 90 7 Ns 3 -C6 H3 83 7 +7 + 6 6 95 98 80 96{2,2,1,1} 6{2,2,1,3} 2 2 2 2 H H 2-NO2 -4-CF 106{2,2,1,2} 6{3,1,1,3} 2 3 2 1 H H 2-NO2Tos -4-CF3 -C6 H3 6 +7 + 6 7 86 97 84 8 90 11 6{3,2,1,2} 3 2 H Tos 7+7 72 54 6{2,2,1,3} 9 2 2 H 2‐NO 2‐4‐CF3‐C6H3 7 + 6 95 80 12 6{3,2,1,3} 3 2 H 2-NO2 -4-CF3 -C6 H3 7+7 88 44 6{3,1,1,3} 6 + 7 86 84 10 3 1 H a 2‐NO2‐4‐CF3‐C6H3 * Entry 6 has previously been reported [34]. The overall purity of crude final compounds 6 after a five-step 6{3,2,1,2} 11 3 2 H Tos 7 + 7 72 54 b synthesis; Yield of the HPLC-purified compounds. 6{3,2,1,3} 12 3 2 H 2‐NO2‐4‐CF3‐C6H3 7 + 7 88 44
Analysis of the results indicated that all combination of cyclic N-acyliminium 6- and 7-membered a The overall purity of crude final compounds 6 after a * Entry 6 has previously been reported [34]. ringsfive‐step synthesis. (prepared using protected aldehydes with one and two-carbon spacers, n = 1, 2) fused with 5-, 6b Yield of the HPLC‐purified compounds. and 7-membered rings and provided the target molecular scaffolds with good to excellent purities and Analysis of the results indicated that all combination of cyclic N‐acyliminium 6‐ and 7‐ yields (Figure 3). membered rings (prepared using protected aldehydes with one and two‐carbon spacers, n = 1, 2) fused with 5‐, 6‐ and 7‐membered rings and provided the target molecular scaffolds with good to excellent purities and yields (Figure 3).
Molecules 2018, 23, 1090 Molecules 2018, 23, x FOR PEER REVIEW
5 of 12 5 of 11
Figure 3. Color‐coded structures of fused scaffolds. Green: good to excellent purity of product; blue: Figure 3. Color-coded structures of fused scaffolds. Green: good to excellent purity of product; blue: only specific combination was formed; red: no cyclic product detected only specific combination was formed; red: no cyclic product detected.
The protected aldehyde with a three‐carbon spacer used for the synthesis of [8 + 5,6,7]‐fused protectedcyclized aldehyde with a [8 three-carbon spacer used for the synthesis of [8(6{1,3,1,1} + 5,6,7]-fused rings The successfully only for + 5] scaffolds derivatized as sulfonamides and rings successfully cyclized only for [8 + 5] scaffolds derivatized as sulfonamides (6{1,3,1,1} and 6{1,3,1,2}) not for those derivatized as aryl compounds. Attempts to prepare [8 + 6,7] fused rings were 6{1,3,1,2}) not for those derivatized as aryl compounds. Attempts to prepare [8 + 6,7] fused rings were unsuccessful; LC/MS indicated the presence of a compound corresponding to cyclic hemiaminal 5. unsuccessful; LC/MS cyclization indicated the presence ofwe a compound corresponding to cyclic5{2,3,1,3} hemiaminal To evaluate different procedures, released hemiaminals 5{1,3,1,3}, and 5. To evaluate different cyclization procedures, we released hemiaminals 5{1,3,1,3}, 5{2,3,1,3} and 5{3,3,1,3} 5{3,3,1,3} from resins 4 using TFA/DCM 1:1. After the evaporation of TFA/DCM under a stream of from resins using TFA/DCM After the evaporation of left TFA/DCM under stream nitrogen, nitrogen, the 4 compounds were 1:1. dissolved in methanol and overnight at rt a to yield of7{1,3,1,3}, the compounds were dissolved in methanol and left overnight at rt7{1,3,1,3} to yieldand 7{1,3,1,3}, 7{2,3,1,3} 7{2,3,1,3} and 7{3,3,1,3} in more than 90% crude purity. Compounds 7{2,3,1,3} were and 7{3,3,1,3} in more than 90% crude purity. Compounds 7{1,3,1,3} and 7{2,3,1,3} were isolated, isolated, purified by reversed‐phase HPLC and characterized by LCMS and NMR (Supplementary purified by Then, reversed-phase HPLC several and characterized by LCMS and NMR (Supplementary Materials). Materials). we screened Lewis acids, including stannous chloride dihydrate Then,2∙2H we2O) screened Lewis acids, including stannous chloride dihydrate (SnCl2 ·2Hconditions 2 O) [36,37] (SnCl [36,37] several and scandium triflate (Sc(OTf) 3) [38] as well as different reaction and scandium triflate (Sc(OTf) ) [38] as well as different reaction conditions (temperature, concentration 3 (temperature, concentration of Sc(OTf) 3, reaction time from 1 to 10 days) and solvents (DCM, MeCN, of Sc(OTf)3 , reaction time from 1 to 10 days) and solvents (DCM, MeCN, dioxane) for the cyclization dioxane) for the cyclization of 7{1,3,1,3}, 7{2,3,1,3} and 7{2,3,1,3} in solution. Disappointingly, no cyclic of 7{1,3,1,3}, 7{2,3,1,3} and 7{2,3,1,3} in solution. Disappointingly, no cyclic product, not even a trace product, not even a trace amount, was detected. amount, was detected. To address the stereoselectivity of this modular solid‐phase synthesis, [7 + 5]‐membered fused To address the stereoselectivity of this modular solid-phase synthesis, [7 + 5]-membered fused compound 6{1,2,2,2} was synthesized using (S)‐α‐bromopropionic acid. The fused bicyclic compound compound 6{1,2,2,2} was synthesized using (S)-α-bromopropionic acid. The fused bicyclic compound featuring two asymmetric carbons was obtained as a mixture of diastereoisomers with a 5:1 ratio. featuring two asymmetric carbons was obtained as a mixture of diastereoisomers with a 5:1 ratio. Both diastereoisomers were isolated by reversed‐phase HPLC, and their structures were confirmed Both diastereoisomers were isolated by reversed-phase HPLC, and their structures were confirmed from their 1D and 2D NMR spectra. The assignment of the configuration of the new stereogenic center from their 1D and 2D NMR spectra. The assignment of the configuration of the new stereogenic center in both diastereomers of fused bicycle 6{1,2,2,2} was based on the diagnostic NOE effect observed in in both diastereomers of fused bicycle 6{1,2,2,2} was based on the diagnostic NOE effect observed the NOESY 1D and ROESY 2D spectra of the compounds in deuterated DMSO. Whereas the NOE in the NOESY 1D and 2D spectra of the compounds in deuterated DMSO. Whereas the correlation between 6‐H ROESY and CH 3 was exhibited in both diastereoisomers, the NOE correlation between CH3 and 9a‐H was exhibited only in the minor isomer and thus was decisive in confirming its configuration as (R,R)‐6{1,2,2,2} and that of the major isomer as (R,S)‐6{1,2,2,2} (Figure 4). Using
Molecules 2018, 23, 1090
6 of 12
NOE correlation between 6-H and CH3 was exhibited in both diastereoisomers, the NOE correlation 6 of 11 only in the minor isomer and thus was decisive in confirming its configuration as (R,R)-6{1,2,2,2} and that of the major isomer as (R,S)-6{1,2,2,2} (Figure 4). Using the the same experimental procedure and 1D and 2D NMR experiments, we identified the structures of same experimental procedure and 1D and 2D NMR experiments, we identified the structures of the the major and minor isomers of the enantiomeric compound afforded using (R)‐α‐bromopropionic major and minor isomers of the enantiomeric compound afforded using (R)-α-bromopropionic acid acid as (S,S)‐6{1,2,3,2} and (S,R)‐6{1,2,3,2}, respectively. Both diastereoisomers were isolated and fully as (S,S)-6{1,2,3,2} and (S,R)-6{1,2,3,2}, respectively. Both diastereoisomers were isolated and fully characterized (Table 2). characterized (Table 2). Molecules 2018, 23, x FOR PEER REVIEW between CH3 and 9a-H was exhibited
Figure 4. Configurational assignment of compounds (a) (R,R)- and (R,S)-6{1,2,2,2} and (b) (S,S)- and Figure 4. Configurational assignment of compounds (a) (R,R)‐ and (R,S)‐6{1,2,2,2} and (b) (S,S)‐ and (S,R)-6{2,2,2,2}. (S,R)‐6{2,2,2,2}. Table 2. Synthesized chiral fused heterocycles 6. The reactions with (R)‐α‐bromopropionic acid provided [7 + 6]‐fused ring model compounds 6{2,2,3,2} as a 55:45 mixture of diastereomers (Table 2). Both isomers were isolated by reversed‐phase Entry Compound m n Rings Purity [%] a R1 Yield [%] b HPLC, and their structures were confirmed by 1D and 2D NMR spectroscopy. The NOESY 1 (R,R)-6{1,2,2,2} 1 2 (R)-Me 7+5 85 49 experiments were performed in CDCl 3 because of the overlap of the resonances of protons H6 and 2 (R,S)-6{1,2,2,2} 1 2 (R)-Me 7+5 15 6 H10a in DMSO‐d 6. The results indicated that the major isomer prepared using (R)‐α‐bromopropionic 3 (S,S)-6{1,2,3,2} 1 2 (S)-Me 7+5 85 44 4 (S,R)-6{1,2,3,2} 1 2 (S)-Me 7+5 15 8 acid was (S,S)‐6{2,2,2,2} and the minor isomer was (S,R)‐6{2,2,2,2} (Table 2). 5 6
a
(S,S)-6{2,2,3,2} (S,R)-6{2,2,3,2}
2 2
2 2
(S)-Me (S)-Me
7+6 7+6
Table 2. Synthesized chiral fused heterocycles 6.
55 45
32 17
The overall purity of crude final compounds (R and S)-6 after a five-step synthesis. b Yield of the HPLC-purified compounds. Entry Compound m n R1 Rings Purity [%] a Yield [%] b
(R,R)‐6{1,2,2,2} 1 1 2 acid(R)‐Me 7 + 5 85 model compounds 49 The reactions with (R)-α-bromopropionic provided [7 + 6]-fused ring (R,S)‐6{1,2,2,2} 2 1 2 (R)‐Me 7 + 5 15 6 6{2,2,3,2} as a 55:45 mixture of diastereomers (Table 2). Both isomers were isolated by reversed-phase (S,S)‐6{1,2,3,2} 3 1 2 (S)‐Me 7 + 5 85 44 HPLC, and their structures were confirmed by 1D and 2D NMR spectroscopy. The NOESY experiments (S,R)‐6{1,2,3,2} 4 1 2 (S)‐Me 7 + 5 15 8 were performed in CDCl3 because of the overlap of the resonances of protons H6 and H10a in (S,S)‐6{2,2,3,2} 5 2 2 (S)‐Me 7 + 6 55 32 DMSO-d . The results indicated that the major2 isomer prepared using (R)-α-bromopropionic acid was (S,R)‐6{2,2,3,2} 6 6 2 (S)‐Me 7 + 6 45 17
(S,S)-6{2,2,2,2} and the minor isomer was (S,R)-6{2,2,2,2} (Table 2). a The overall purity of crude final compounds (R and S)‐6 after a five‐step synthesis. b Yield of the NMR spectra of the compounds in CDCl provided interesting information regarding the 3 HPLC‐purified compounds. stability of the diastereomers. We observed that the purified isomer (S,S)-6{2,2,3,2} was not NMR spectra solution of the compounds in CDCl 3 provided interesting information regarding the stable in CDCl and was partially converted into the (S,R)-6{2,2,3,2} isomer at room 3 stability of the diastereomers. We observed that the purified isomer (S,S)‐6{2,2,3,2} was not stable in temperature. The same effect was also observed for isomer (S,R)-6{2,2,3,2}, which was converted CDCl 3 solution and was partially converted into the (S,R)‐6{2,2,3,2} isomer at room temperature. The into (S,S)-6{2,2,3,2}. After 8 days in CDCl3 solution, the two isomers (S,S)-6{2,2,3,2} and (S,R)-6{2,2,3,2} same effect was also observed for isomer (S,R)‐6{2,2,3,2}, which was converted into (S,S)‐6{2,2,3,2}. formed an equilibrium mixture of diastereomers with the same ratio (55:45), likely due to the opening After days in CDClfused 3 solution, the two isomers (S,S)‐6{2,2,3,2} and (S,R)‐6{2,2,3,2} formed an of the8 acid-mediated ring and the formation of the cyclic iminium followed by ring closure. equilibrium mixture of diastereomers with the same ratio (55:45), likely due to the opening of the The same ratio of diastereomers was observed in the crude product after its TFA-mediated cleavage acid‐mediated fused ring and the formation of the cyclic iminium followed by ring closure. The same from the resin. ratio of diastereomers was observed in the crude product after its TFA‐mediated cleavage from the The experimental results indicated that 8-membered cyclic N-acyliminiums formed fused systems resin. with 5-membered rings but not with 6 and 7-membered rings. To address the effect of introducing The experimental results indicated that component 8‐membered cyclic N‐acyliminiums formed fused a conformational constraint into the acyclic (3-aminopropanol and 4-aminobutanol) systems with 5‐membered rings but not with 6 and 7‐membered rings. To address the effect of on cyclization, we prepared resin-bound intermediates using 2-(aminomethyl)phenol (8) and introducing a conformational constraint into the acyclic component (3‐aminopropanol and 4‐ (2-(aminomethyl)phenyl)methanol (9) (Figure 5). The beneficial effect of conformational constraints on aminobutanol) on cyclization, we prepared resin‐bound intermediates using 2‐(aminomethyl)phenol cyclization was reported in unrelated cyclizations [39,40]. N-Fmoc derivatives of amino alcohols 8 and (8) and (2‐(aminomethyl)phenyl)methanol (9) spacers, (Figure bromoacetic 5). The beneficial of were conformational 9, amino aldehydes with three different carbon acid andeffect Tos-Cl used for the constraints on cyclization was reported in unrelated cyclizations [39,40]. N‐Fmoc derivatives of amino alcohols 8 and 9, amino aldehydes with three different carbon spacers, bromoacetic acid and Tos‐Cl were used for the synthesis of model compounds 10 according to Scheme 1. Cleaving compounds 10{1,1} and 10{1,2} was accomplished using a mild cleavage cocktail of 10% TFA in DCM. For compounds 10{2,1} and 10{2,2}, the typical cleavage cocktail (50% TFA in DCM) was employed.
Molecules 2018, 23, 1090
7 of 12
synthesis of model compounds 10 according to Scheme 1. Cleaving compounds 10{1,1} and 10{1,2} cocktail of 10% TFA in DCM. For compounds 10{2,1}7 of 11 and 10{2,2}, the typical cleavage cocktail (50% TFA in DCM) was employed. The results indicated that good to excellent purities and yields of 10 were successfully obtained The results indicated that good to excellent purities and yields of 10 were successfully obtained via traceless solid‐phase synthesis of combinations of cyclic N‐acyliminium 6‐ and 7‐membered rings via traceless solid-phase synthesis of Molecules 2018, 23, x FOR PEER REVIEW combinations of cyclic N-acyliminium 6- and 7-membered 7 of 11 rings (prepared from protected aldehydes with one and two‐carbon spacers, n = 1, 2) with 6‐ and 7‐ (prepared from protected aldehydes with one and two-carbon spacers, n = 1, 2) with 6- and 7-membered membered rings (Table 3). Neither of the constrained amino alcohol promoted cyclization to form [8 The results indicated that good to excellent purities and yields of 10 were successfully obtained rings (Table 3). Neither of the constrained amino alcohol promoted cyclization to form [8 + 6]- or + 6]‐ or [8 + 7]‐fused rings. via traceless solid‐phase synthesis of combinations of cyclic N‐acyliminium 6‐ and 7‐membered rings [8 + 7]-fused rings. Molecules 2018, 23, x FOR PEER REVIEW was accomplished using a mild cleavage
(prepared from protected aldehydes with one and two‐carbon spacers, n = 1, 2) with 6‐ and 7‐ membered rings (Table 3). Neither of the constrained amino alcohol promoted cyclization to form [8 + 6]‐ or [8 + 7]‐fused rings.
Figure 5. 5. Compounds Compounds10 10prepared preparedusing using the constrained amino alcohols 8 and 9 according to Figure the constrained amino alcohols 8 and 9 according to the the Scheme 1. Scheme 1. Figure 5. Compounds 10 prepared using the constrained amino alcohols 8 and 9 according to the Scheme 1. Table 3. The synthesized fused heterocyclic compounds 10. Table 3. The synthesized fused heterocyclic compounds 10.
a
a Entry Compound Purity [%] Yield [%] b Entry Table 3. The synthesized fused heterocyclic compounds 10. Compound m m n n Purity [%] a Yield [%] b 10{1,1} 1 1 0 1 76 41 10{1,1} 0 m 1n 76 41 a b Entry Compound Purity [%] Yield [%] 10{1,2} 2 2 43 10{1,2} 0 0 2 2 81 81 43 10{1,1} 1 0 76 10{2,1} 1 1 11 97 9041 90 10{2,1} 3 3 1 97 10{1,2} 2 4 81 10{2,2} 1 0 22 96 8143 10{2,2} 4 3 1 10{2,1} 1 1 2 97 96 90 81 b
The overall purity of the crude final compounds 10 after a five-step synthesis process; Yield of the b Yield of the The overall purity of the crude final compounds 10 after a five‐step synthesis process. 10{2,2} 4 1 2 96 81 HPLC-purified compounds. a The overall purity of the crude final compounds 10 after a five‐step synthesis process. b Yield of the HPLC‐purified compounds. To demonstrate the second traceless strategy of the synthetic route for molecular scaffolds, HPLC‐purified compounds. a
To demonstrate the second traceless strategy of the synthetic route for molecular scaffolds, we we synthesized model compounds containing aromatic C-nucleophiles to effect the fused ring closure. To demonstrate the second traceless strategy of the synthetic route for molecular scaffolds, we synthesized model compounds containing aromatic C‐nucleophiles to effect the fused ring closure. In the previous synthesis, we used oxygen nucleophiles from amino alcohols for immobilization, synthesized model compounds containing aromatic C‐nucleophiles to effect the fused ring closure. In the previous synthesis, we used oxygen nucleophiles from amino alcohols for immobilization, a a scenario that is not applicable for C-nucleophiles. To change our synthetic strategy, we attached In the previous synthesis, we used oxygen nucleophiles from amino alcohols for immobilization, a scenario that is not applicable for C‐nucleophiles. To change our synthetic strategy, we attached the the acyclic precursors to the resin via amide nitrogen. This strategy also allowed for the traceless scenario that is not applicable for C‐nucleophiles. To change our synthetic strategy, we attached the acyclic precursors to compounds. the resin via This strategy was also allowed the traceless synthesis of the target Asamide shownnitrogen. in Scheme 2, synthesis carried outfor on aminomethyl acyclic precursors to the resin via amide nitrogen. This strategy also allowed for the traceless synthesis of the target compounds. As shown in Scheme 2, synthesis was carried out on aminomethyl polystyrene resin 11 equipped with an acid-labile BAL (4-(4-formyl-3-methoxyphenoxy)butyric acid) synthesis of the target compounds. As shown in Scheme 2, synthesis was carried out on aminomethyl polystyrene resin 11 equipped with an acid‐labile BAL (4‐(4‐formyl‐3‐methoxyphenoxy)butyric acid) polystyrene resin 11 equipped with an acid‐labile BAL (4‐(4‐formyl‐3‐methoxyphenoxy)butyric acid) linker [41]. linker [41]. linker [41].
Scheme 2. Polymer‐supported synthesis of fused bicyclic compounds 15 16. and Reagents 16. Reagents and Scheme 2. Polymer-supported synthesis of fused bicyclic compounds 15 and and conditions: (i) 0.5 M 2‐(3‐methoxyphenyl)ethan‐1‐amine in 10% AcOH/DMF, rt, overnight, then (i) 0.5conditions: M 2. 2-(3-methoxyphenyl)ethan-1-amine 10% AcOH/DMF, rt, overnight, then 3, Scheme Polymer‐supported synthesis of in fused bicyclic compounds 15 and 16. NaBH(AcO) Reagents and 3, rt, 5 h, then a second portion of NaBH(AcO)3, rt, 3 h; (ii) bromoacetic acid, DIC, DCM, NaBH(AcO) rt, 5 h, then a second portion of NaBH(AcO) , rt, 3 h; (ii) bromoacetic acid, DIC, DCM, rt, 1 h; 3 conditions: (i) 0.5 M 2‐(3‐methoxyphenyl)ethan‐1‐amine in 10% AcOH/DMF, rt, overnight, then rt, 1 h; (iii) amino‐aldehyde dialkyl acetal, DIEA, DMF, rt, 2 h; (iv) Tos‐Cl, DIEA, DCM, rt, 2 h; (v) (iii) amino-aldehyde dialkyl acetal, DIEA, DMF, rt, 2 h; (iv) Tos-Cl, DIEA, DCM, rt, 2 h; (v) TFA/DCM 3, rt, 5 h, then a second portion of NaBH(AcO)3, rt, 3 h; (ii) bromoacetic acid, DIC, DCM, NaBH(AcO) TFA/DCM (1:1), rt, 1 h. (1:1), rt, 1 h. rt, 1 h; (iii) amino‐aldehyde dialkyl acetal, DIEA, DMF, rt, 2 h; (iv) Tos‐Cl, DIEA, DCM, rt, 2 h; (v) TFA/DCM (1:1), rt, 1 h. After immobilization of the primary amines via reductive amination using 2‐(3‐ methoxyphenyl)ethan‐1‐amine, secondary amines resin 12 was acylated with bromopropionic acid to give immobilization resin 13. Then, the with the amines amino‐aldehyde dialkyl acetal, followed using by N‐ 2‐(3‐ After of reaction the primary via reductive amination methoxyphenyl)ethan‐1‐amine, secondary amines resin 12 was acylated with bromopropionic acid to give resin 13. Then, the reaction with the amino‐aldehyde dialkyl acetal, followed by N‐
Molecules 2018, 23, 1090
8 of 12
After immobilization of the primary amines via reductive amination using 2-(3-methoxyphenyl) ethan-1-amine, secondary amines resin 12 was acylated with bromopropionic acid to give resin 13. Then, the reaction with the amino-aldehyde dialkyl acetal, followed by N-derivatization with Tos-Cl, were carried out to yield intermediates 14. Finally, TFA exposure triggered the release of the polymer-supported acyclic precursors from the resin, unmasking the protecting groups and forming the six- and seven-membered N-acyliminium ions. Internal nucleophilic attack then provided target molecules 15{1} and isomeric 16{1} or, in the case of the protected aldehyde with the n = 2 carbon spacer, only compound 15{2} (Table 4). These compounds were isolated by reversed-phase HPLC, and NMR spectra confirmed their structures. Disappointingly, formation of the [8 + 6] combination was not detected. Table 4. Synthesis of fused heterocyclic compounds 19 and 20.
a
Entry
Compound
n
Purity [%] a
Yield [%] b
1 2 3
15{1} 16{1} 15{2}
1 1 2
83 17 77
42 6 22
The overall purity of the crude products after a five-step synthesis.
b
Yield of the HPLC-purified compounds.
3. Materials and Methods 3.1. General The solid-phase syntheses were carried out in plastic reaction vessels (syringes equipped with porous discs) using a manually operated synthesizer [42]. The volume of the wash solvent was 10 mL per 1 g of resin. The resin slurry was washed by shaking it with fresh solvent for at least 1 min before the solvent was changed. All reactions were carried out at ambient temperature unless stated otherwise. Commercially available Wang (100–200 mesh, 1.0 mmol/g) and aminomethyl resins (100-200 mesh, 1.26 mmol/g) were used. Commercial chemicals and solvents were ACS reagent grade. The yields of the crude products were calculated with respect to the loading of the first building block. 3.2. Reaction of Wang Resin 1 with Trichloroacetonitrile and Fmoc-amino Alcohol (Resin 2) Wang resin 1 (1 mmol/g loading, 1 g) was suspended in 10 mL of anhydrous DCM, then 1.5 mL of trichloroacetonitrile was added, and the resin was left in a freezer for 30 min. Next, a solution of 100 µL of DBU in 2 mL of anhydrous DCM was added, and the slurry was shaken for 1 h. The resin was washed with anhydrous DCM (3×) and anhydrous THF (3×). The solution of Fmoc-amino alcohol (3 mmol) in 10 mL of anhydrous THF was added to the resin, followed by the dropwise addition of 63 µL of a solution of BF3 ·Et2 O, and the slurry was shaken for 30 min. Resin 2 was washed THF (3×), MeOH (3×), and DCM (5×). A sample of the resin was washed with MeOH (3×) and dried, and 10 mg was cleaved with 50% TFA for 30 min, and the product quantified (Fmoc absorbance at 300 nm). Typical loading was between 0.35–0.50 mmol/g. 3.3. Acylation with α-Bromocarboxylic Acids (Resins 3 And 13) Resin 2 (1 g) was washed with DCM and DMF, Fmoc was deprotected with 50% piperidine in DMF for 15 min, and the resin was washed with DMF (3×) and DCM (5×). Reaction of the resin with a solution of α-bromocarboxylic acid (5 mmol) in 10 mL of DCM was carried out in a syringe with a frit, and DIC (2.5 mmol, 386 µL) was added. After 5 min, DIU was filtered, and the solution was transferred to the syringe with resin 1 and shaken for 1 h. When bromoacetic acid was used, DIEA (2.5 mmol, 436 µL) was also added. Then, the resin was washed with DCM (3×).
Molecules 2018, 23, 1090
9 of 12
3.4. Reaction with Amino-aldehyde Dialkyl Acetal and N-Derivatization (Resin 4) Resin 3 (1 g) was washed with DMF (3×), a solution of amino-aldehyde dialkyl acetal (10 mmol) and DIEA (10 mmol, 1.74 mL) in 10 mL of DMF was added, and the slurry was shaken for 2 h. An analytical sample was taken for analysis, reacted with 0.5 M Fmoc-OSu in DCM for 1 h and cleaved with 50% TFA for 30 min. Resin from the previous step was split into four 250 mg portions, which were washed with reaction solvent: DCM for sulfonamides (Ns-Cl, Tos-Cl) (1 mmol) and DMSO for 4-fluoro-3nitrobenzotrifluoride (1 mmol). Then, 3 mL of the reaction solvent and DIEA (1.8 mmol, 174 µL) was added, and the slurry was shaken for 2 h (in the case of sulfonamides) or overnight (in the case of 4-trifluoromethyl-2-nitroaryl). Resin 4 was washed with the reaction solvent (3×) and then DCM (3×). A sample for analysis was reacted with 0.5 M Fmoc-OSu in DCM for 1 h and cleaved with 50% TFA for 30 min. 3.5. Acid-Mediated Cleavage, Cyclization and Isolation (Compounds 6 and 7) Resin 4 was treated with 50% TFA in DCM for 1 h. The TFA solution was collected, the resin was washed with 50% TFA in DCM (3×), and the extracts were combined and evaporated under a stream of nitrogen. When TFA exposure yielded cyclic hemiaminals (5), the oily residues were dissolved in MeOH and left at rt overnight to yield 7. Compounds 6 and 7 were dissolved in acetonitrile or methanol and purified by semi-preparative reversed-phase HPLC. 3.6. Reductive Amination of the BAL Resin (Resin 12) A 20 mL syringe was charged with BAL resin 11 (500 mg) and washed with anhydrous DMF (3×). Then, 0.5 M 2-(3-methoxyphenyl)ethan-1-amine and 10% AcOH in anhydrous DMF (5 mL) was added into the syringe and shaken overnight. Next, sodium triacetoxyborohydride NaBH(AcO)3 (2.5 mmol, 528 mg) was added to the syringe. Because of the evolution of hydrogen gas, the reaction vessel was vented to relieve pressure by piercing it on the top with a needle on the top, placing it in a horizontal position and shaking it for 5 h. A second portion of NaBH(AcO)3 (2.5 mmol, 528 mg) was added and shaken for 3 h. After the reaction, the resin was washed with 5% AcOH in DMF (3×) and DMF (3×), neutralized with 5% piperidine in DMF (10 mL) and washed with DMF (5×) and DCM (5×) to yield resin 12. 3.7. Reaction with Amino-aldehyde Dialkyl Acetal and N-Derivatization (Resin 14) Acid-mediated Cleavage, Cyclization and Isolation (Compounds 15 and 16) The protocol was the same as that described for the synthesis of compound 6. 4. Conclusions We developed two synthetic strategies for traceless polymer-supported synthesis of natural product-like molecular scaffolds comprising two fused rings. The linear precursors were cyclized via acid-mediated tandem N-acylium ion formation followed by nucleophilic addition of O- and C-nucleophiles. Synthetic scheme facilitated the preparation of any combination of acyclic ring precursors for the synthesis of different ring sizes and assessed ring-forming limitations with respect to ring sizes. [6,7 + 5,6,7]-Fused cyclic combinations were obtained with excellent crude purity and yield. The 8-membered cyclic iminium was successfully fused only with the 5-membered cycle, whereas larger fused rings were not formed. The syntheses proceeded under mild conditions and used commercially available building blocks. Further work exploring the potential applicability of this strategy to larger ring sizes by using building blocks with the predisposition to adopt conformations favorable for large ring closure is in progress. Supplementary Materials: The following are available online. all compounds.
1 H, 13 C
NMR spectral data and figures of
Molecules 2018, 23, 1090
10 of 12
Author Contributions: V.K. conceived and designed the experiments; V.G.-N. performed the experiments. Acknowledgments: This research was supported by the Department of Chemistry and Biochemistry, University of Notre Dame, and by research grant 16-06446S from the Grant Agency of the Czech Republic (GACR). Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
13.
14.
15.
16.
17.
18. 19.
Newman, D.J.; Cragg, G.M.; Snader, K.M. The Influence of Natural Products Upon Drug Discovery. Nat. Prod. Rep. 2000, 17, 215–234. [CrossRef] [PubMed] Newman, D.J.; Cragg, G.M.; Snader, K.M. Natural Products as Sources of New Drugs Over the Period 1981–2002. J. Nat. Prod. 2003, 66, 1022–1037. [CrossRef] [PubMed] Lopez-Vallejo, F.; Giulianotti, M.A.; Houghten, R.A.; Medina-Franco, J.L. Expanding the Medicinally Relevant Chemical Space With Compound Libraries. Drug Discov. Today 2012, 17, 718–726. [CrossRef] [PubMed] Lovering, F.; Bikker, J.; Humblet, C. Escape From Flatland: Increasing Saturation As an Approach to Improving Clinical Success. J. Med. Chem. 2009, 52, 6752–6756. [CrossRef] [PubMed] Wetzel, S.; Bon, R.S.; Kumar, K.; Waldmann, H. Biology-Oriented Synthesis. Angew. Chem. Int. Ed. 2011, 50, 10800–10826. [CrossRef] [PubMed] Kaiser, M.; Wetzel, S.; Kumar, K.; Waldmann, H. Biology-Inspired Synthesis of Compound Libraries. Cell. Mol. Life Sci. 2008, 65, 1186–1201. [CrossRef] [PubMed] O’Connor, C.J.; Beckmann, H.S.G.; Spring, D.R. Diversity-Oriented Synthesis: Producing Chemical Tools for Dissecting Biology. Chem. Soc. Rev. 2012, 41, 4444–4456. [CrossRef] [PubMed] La Venia, A.; Ventosa-Andrés, P.; Krchnak, V. Peptidomimetics via Iminium Chemistry on Solid Phase: Single, Fused, and Bridged Heterocycles. Peptidomimetics II Top. Heterocycl. Chem. 2017, 49, 105–126. Maryanoff, B.E.; Zhang, H.C.; Cohen, J.H.; Turchi, I.J.; Maryanoff, C.A. Cyclizations of N-Acyliminium Ions. Chem. Rev. 2004, 104, 1431–1628. [CrossRef] [PubMed] Yazici, A.; Pyne, S.G. Intermolecular Addition Reactions of N-Acyliminium Ions (Part I). Synthesis 2009, 3, 339–368. Yazici, A.; Pyne, S.G. Intermolecular Addition Reactions of N-Acyliminium Ions (Part II). Synthesis 2009, 4, 513–541. Le Quement, S.T.; Nielsen, T.E.; Meldal, M. Scaffold Diversity Through Intramolecular Cascade Reactions of Solid-Supported Cyclic N-Acyliminium Intermediates. J. Comb. Chem. 2007, 9, 1060–1072. [CrossRef] [PubMed] Ventosa-Andres, P.; La-Venia, A.; Ripoll, C.A.B.; Hradilova, L.; Krchnak, V. Synthesis of Nature-Inspired Medium-Sized Fused Heterocycles From Amino Acids. Chem. Eur. J. 2015, 21, 13112–13119. [CrossRef] [PubMed] Hanessian, S.; McNaughton-Smith, G.; Lombart, H.G.; Lubell, W.D. Design and Synthesis of Conformationally Constrained Amino Acids As Versatile Scaffolds and Peptide Mimetics. Tetrahedron 1997, 53, 12789–12854. [CrossRef] Nielsen, T.E.; Le Quement, S.; Meldal, M. Solid-Phase Synthesis of Bicyclic Dipeptide Mimetics by Intramolecular Cyclization of Alcohols, Thiols, Amines, and Amides with N-Acyliminium Intermediates. Org. Lett. 2005, 7, 3601–3604. [CrossRef] [PubMed] Airiau, E.; Spangenberg, T.; Girard, N.; Schoenfelder, A.; Salvadori, J.; Taddei, M.; Mann, A. A General Approach to Aza-Heterocycles by Means of Domino Sequences Driven by Hydroformylation. Chem. Eur. J. 2008, 14, 10938–10948. [CrossRef] [PubMed] Bencsik, J.R.; Kercher, T.; O’Sulliva, M.; Josey, J.A. Efficient, Stereoselective Synthesis of Oxazolo[3,2-a] Pyrazin-5-Ones: Novel Bicyclic Lactam Scaffolds From the Bicyclocondensation of 3-Aza-1,5-Ketoacids and Amino Alcohols. Org. Lett. 2003, 5, 2727–2730. [CrossRef] [PubMed] Ciofi, L.; Morvillo, M.; Sladojevich, F.; Guarna, A.; Trabocchi, A. Skeletal Diversity by Sequential One-Pot and Stepwise Routes Using Morpholine Ester Scaffolds. Tetrahedron Lett. 2010, 51, 6282–6285. [CrossRef] Estiarte, M.A.; Rubiralta, M.; Diez, A.; Thormann, M.; Giralt, E. Oxazolopiperidin-2-Ones As Type II’ Beta-Turn Mimetics: Synthesis and Conformational Analysis. J. Org. Chem. 2000, 65, 6992–6999. [CrossRef] [PubMed]
Molecules 2018, 23, 1090
20. 21. 22.
23.
24.
25.
26. 27. 28. 29.
30.
31. 32. 33.
34.
35. 36.
37. 38. 39.
40.
11 of 12
Wirt, U.; Frohlich, R.; Wunsch, B. Asymmetric Synthesis of 1-Substituted Tetrahydro-3-Benzoazepines. Tetrahedron: Asymmetry 2005, 16, 2199–2202. [CrossRef] Vagner, J.; Qu, H.; Hruby, V.J. Peptidomimetics, a Synthetic Tool of Drug Discovery. Curr. Opin. Chem. Biol. 2008, 12, 292–296. [CrossRef] [PubMed] Lawandi, J.; Toumieux, S.; Seyer, V.; Campbell, P.; Thielges, S.; Juillerat-Jeanneret, L.; Moitessier, N. Constrained Peptidomimetics Reveal Detailed Geometric Requirements of Covalent Prolyl Oligopeptidase Inhibitors. J. Med. Chem. 2009, 52, 6672–6684. [CrossRef] [PubMed] Wang, W.; Yang, J.; Ying, J.; Xiong, C.; Zhang, J.; Cai, C.; Hruby, V.J. Stereoselective Synthesis of Dipeptide Beta-Turn Mimetics: 7-Benzyl and 8-Phenyl Substituted Azabicyclo[4.3.0]Nonane Amino Acid Esters. J. Org. Chem. 2002, 67, 6353–6360. [PubMed] Tong, Y.; Fobian, Y.; Wu, M.; Boyd, N.D.; Moeller, K.D. Conformationally Constrained Substance P Analogues: The Total Synthesis of a Constrained Peptidomimetic for the Phe7-Phe8 Region. J. Org. Chem. 2000, 65, 2484–2493. [CrossRef] [PubMed] Hanessian, S.; Buckle, R.; Bayrakdarian, M. Design and Synthesis of a Novel Class of Constrained Tricyclic Pyrrolizidinone Carboxylic Acids As Carbapenem Mimics. J. Org. Chem. 2002, 67, 3387–3397. [CrossRef] [PubMed] Cox, E.D.; Cook, J.M. The Pictet-Spengler Condensation: A New Direction for an Old Reaction. Chem. Rev. 1995, 95, 1797–1842. [CrossRef] Nielsen, T.E.; Meldal, M. Solid-Phase Synthesis of Pyrroloisoquinolines Via the Intramolecular N-Acyliminium Pictet-Spengler Reaction. J. Comb. Chem. 2005, 7, 599–610. [CrossRef] [PubMed] Nielsen, T.E.; Meldal, M. Solid-Phase Intramolecular N-Acyliminium Pictet-Spengler Reactions As Crossroads to Scaffold Diversity. J. Org. Chem. 2004, 69, 3765–3773. [CrossRef] [PubMed] Diness, F.; Beyer, J.; Meldal, M. Solid-Phase Synthesis of Tetrahydro-Beta-Carbolines and Tetrahydroisoquinolines by Stereoselective Intramolecular N-Carbamyliminium Pictet-Spengler Reactions. Chem. Eur. J. 2006, 12, 8056–8066. [CrossRef] [PubMed] Petersen, R.; LeQuement, S.T.; Nielsen, T.E. Synthesis of a Natural Product-Like Compound Collection Through Oxidative Cleavage and Cyclization of Linear Peptides. Angew. Chem. Int. Ed. 2014, 53, 11778–11782. [CrossRef] [PubMed] Schutznerova, E.; Oliver, A.G.; Zajicek, J.; Krchnak, V. Polymer-Supported Stereoselective Synthesis of (1S,5S)-6-Oxa-3,8-Diazabicyclo[3.2.1]Octanes. Eur. J. Org. Chem. 2013, 3158–3165. [CrossRef] La Venia, A.; Lemrova, B.; Krchnak, V. Regioselective Incorporation of Backbone Constraints Compatible With Traditional Solid-Phase Peptide Synthesis. ACS Comb. Sci. 2013, 15, 59–72. [CrossRef] [PubMed] Ventosa-Andres, P.; Barea Ripoll, C.A.; La-Venia, A.; Krchnak, V. Solid-Phase Synthesis of Fused 1,4-Diazepanone Peptidomimetics Via Tandem N-Iminium Ion Cyclization-Nucleophilic Addition. Tetrahedron Lett. 2015, 56, 5424–5428. [CrossRef] La Venia, A.; Dolensky, B.; Krchnak, V. Polymer-Supported Stereoselective Synthesis of Tetrahydro-2HOxazolo[3,2-a]Pyrazin-5(3H)-Ones From N-(2-Oxo-Ethyl)-Derivatized Dipeptides Via Eastbound Iminiums. ACS Comb. Sci. 2013, 15, 162–167. [CrossRef] [PubMed] Hanessian, S.; Xie, F. Polymer-Bound P-Alkoxybenzyl Trichloracetimidates: Reagents for the Protection of Alcohols as Benzyl Ethers on Solid-Phase. Tetrahedron Lett. 1998, 39, 733–736. [CrossRef] Cayley, A.N.; Gallagher, K.A.; Menard-Moyon, C.; Schmidt, J.P.; Diorazio, L.J.; Taylor, R.J.K. Preparation of Novel Polycyclic Heterocycles Using a Tin(II) Chloride Dihydrate-Mediated Deacetalisation-Bicyclisation Sequence. Synthesis 2008, 3846–3856. Reid, M.; Taylor, R.J.K. A Stannous Chloride-Induced Deacetalisation–Cyclisation Process to Prepare the ABC Ring System of ‘Upenamide. Tetrahedron Lett. 2004, 45, 4181–4183. [CrossRef] Ladziat, V. Recent Applications of Rare-Earth Metal(III) Triflates in Cycloaddition and Cyclization Reactions. Arkivoc 2014, 2014, 307–336. [CrossRef] Marti-Centelles, V.; Pandey, M.D.; Burguete, M.I.; Luis, S.V. Macrocyclization Reactions: The Importance of Conformational, Configurational, and Template-Induced Preorganization. Chem. Rev. 2015, 115, 8736–8834. [CrossRef] [PubMed] Vassilikogiannakis, G.; Margaros, I.; Tofi, M. Olefin Metathesis: Remote Substituents Governing the Stereoselectivity of 11-Membered-Ring Formation. Org. Lett. 2004, 6, 205–208. [CrossRef] [PubMed]
Molecules 2018, 23, 1090
41.
42.
12 of 12
Jensen, K.J.; Alsina, J.; Songster, M.F.; Vagner, J.; Albericio, F.; Barany, G. Backbone Amide Linker (BAL) Strategy for Solid-Phase Synthesis of C-Terminal-Modified and Cyclic Peptides 1, 2, 3. J. Am. Chem. Soc. 1998, 120, 5441–5452. [CrossRef] Krchnak, V.; Padera, V. The Domino Blocks: A Simple Solution for Parallel Solid Phase Organic Synthesis. Bioorg. Med. Chem. Lett. 1998, 22, 3261–3264. [CrossRef]
Sample Availability: Not available. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
