Exploration of the interrupted Fischer indolization reaction

Abstract

A convergent method to access the fused indoline ring system present in a multitude of bioactive molecules has been developed. The strategy involves the condensation of hydrazines with latent aldehydes to ultimately deliver indoline-containing products by way of an interrupted Fischer indolization sequence. The method is convergent, mild, operationally simple, broad in scope, and can be used to access enantioenriched products. In addition, our approach is amenable to the synthesis of furoindoline and pyrrolidinoindoline natural products as demonstrated by the concise formal total syntheses of physovenine and debromoflustramine B. The strategy will likely enable the synthesis of more complex targets such as the communesin alkaloids.

1. Introduction

The discovery of efficient methods to synthesize complex bioactive molecules continues to be a vital area of research.ⁱ A subset of compounds that have received substantial interest due to their medicinal properties and impressive structures are those that possess a fused indoline motif, of the type 1 (Figure 1 and Scheme 1). The simplest of these compounds are the acetylcholinesterase inhibitors physovenine (2) and physostigmine (3),^ii,iii which are composed of basic furo- and pyrrolidinoindoline motifs, respectively (Figure 1). Numerous relatives of pyrrolidinoindoline 3 have been isolated, including bis(prenylated) derivatives,^iv dimeric structures,^v and compounds possessing a heteroatom substituent at C3 (e.g., 4– 7, respectively).^vi,vii Beyond these compounds, a variety of more architecturally complex indoline containing natural products are known, such as the akuammiline alkaloids (e.g., 8–11),^viii,ix perophoramidine (12),^x the communesins (e.g., 13),^xi diazonamide A (14),^xii and bipleiophylline (15).^xiii Many of these molecules possess interesting biological properties, which further enhance their appeal as targets for total synthesis.

Figure 1.

Parent indoline 1 and representative natural products 2–15

Scheme 1.

Approach to indoline 1.

The importance of indoline-containing compounds has prompted the development of a number of methods to access such motifs, with numerous studies particularly in the area of pyrrolidinoindoline synthesis. In most cases, the fused indoline ring systems 1 are constructed by cyclization of precursors of the type 16, which in turn are derived from substituted indole^xiv or oxindole^xv intermediates (Scheme 1). Herein, we report the development of a powerful cascade reaction that allows direct access to 1 (via 16) from the coupling of two simple fragments.^xvi The transformation is convergent, broad in scope, proceeds under mild reaction conditions, and can be used to synthesize a variety of natural product scaffolds.

Our approach to the indoline scaffold 1 of compounds 2– 15 is inspired by the classic Fischer indole synthesis,^xvii,xviii and is presented in Scheme 2. We envisioned that an arylhydrazine 17 and an α-disubstituted aldehyde 18 would react in the presence of acid to afford enamine intermediate 19. Subsequent [3,3]-sigmatropic rearrangement and re-aromatization would provide aniline 20, which in turn would cyclize with loss of NH₃ to furnish transient indolenine 16. Intramolecular attack by a proximal heteroatom substituent (X=NR or O) would deliver the desired product 1. This interrupted Fischer indolization process would allow for the formation of three new bonds, two heterocyclic rings and two stereogenic centers, one of which is quaternary (C3).

Scheme 2.

Proposed fragment coupling / cyclization cascade to access indoline 1.

A key feature of our approach to 1 is the ready availability of starting materials 17 and 18. The arylhydrazine coupling partners 17 could be easily prepared or accessed from commercial sources.^xix Although the required α-branched aldehyde fragments 18 would not be obtainable commercially, isomeric lactols and hemiaminals 21 could likely serve as suitable aldehyde surrogates in the desired transformation (Scheme 3).^xx,xxi In turn, lactols and hemiaminals 21 could be accessed by reduction of readily available lactones or lactams 22.^xxii

Scheme 3.

Lactols and hemiaminals as latent aldehydes.

Only scattered examples of the interrupted Fischer indolization process have been reported over the past fifty years.^{xxiii,xxiv,xxv} Most notable are the seminal studies by Grandberg summarized in Figure 2.^xxiii In 1967, C2 substituted pyrrolidinoindoline 25 was prepared by reacting phenylhydrazine (23) and 5-chloro-3-methylpentan-2-one (24).^xxiiia However, this method is not applicable to the synthesis of furoindolines, or to the more complex ring systems encountered in numerous natural products. It was later demonstrated that furoindolines could be accessed by the acid-promoted reaction of phenylhydrazines with α-disubstituted lactones.^xxiiib For example, reaction of hydrazine 26 and lactone 27 in HCl/iPrOH afforded furoindoline 28 in 22% yield. This method bears limitations, such as the modest yields of products, the use of strongly acidic conditions, and the constraint to furoindoline ring systems. Despite these laudable efforts, and those of others,^xxiv,xxv a general and mild method to access 1 using the interrupted Fischer indolization strategy outlined in Scheme 2 has remained elusive. Moreover, with the exception of our studies,^xvi the notion that such a method could be used to prepare the indoline scaffold present in a multitude of complex biologically important compounds has not been realized.

Figure 2.

Grandberg’s syntheses of pyrrolidinoindoline 25 and furoindoline 28

2.Results and discussion

2.1. Synthesis of furoindolines

The feasibility of the proposed cascade reaction sequence of Scheme 2 was established in the context of furoindoline synthesis. Thus, the reaction between commercially available phenylhydrazine (23) and latent aldehyde 29 (1 equiv) was carried out under a variety of acidic conditions (Table 1). Lewis acids were examined and found to be ineffective (entries 1 and 2). However, use of p-toluenesulfonic acid, trifluoroacetic acid, or HCl each afforded the desired product 30 in modest yield (entries 3–5). Sulfuric acid-mediated reaction conditions were also explored, and ultimately provided the desired product in 87% yield (entry 6). Recognizing that a milder acid source would be more generally useful, acetic acid was examined. Although the use of glacial acetic acid afforded modest product yields (entry 7), employment of a 1:1 mixture of acetic acid and water at 60 °C furnished indoline 30 in 89% isolated yield (entry 8).^xxvi

Table 1.

Survey of acids to promote furoindoline formation.

entry	acid source	conditions	yielda
1	PCl3	benzene, 60 °C	< 5%
2	ZnCl2	EtOH, 100 °C	< 5%
3	TsOH	EtOH, H2O, 60 °C	51%
4	TFA	CH3CN, 60 °C	64%
5	5% HCl	CH3CN, 60 °C	70%
6	4% H2SO4	CH3CN, 60 °C	87%
7	AcOH	AcOH, 60 °C	52%
8	AcOH	1:1 AcOH/H2O, 60 °C	89%b

^aUnless otherwise noted, yields determined by ¹H NMR analysis.

^bIsolated yield.

As shown in Table 2, a number of arylhydrazines bearing N-substitution were examined in the interrupted Fischer indolization reaction. In addition to parent arylhydrazine 23 (entry 1), N-methyl,^xxvii N-benzyl, and N-allyl substituted hydrazines were deemed competent coupling partners (entries 2–4). Interestingly, the use of N-acetyl and N-Boc phenylhydrazines (entries 5 and 6) led predominantly to the recovery of unreacted starting materials.^xxviii,xxix

Table 2.

Variation of the hydrazine N-substituent.

entrya	hydrazine	product	yieldc
1			89%
2b			70%
3			59%
4			60%
5			<5%
6			<5%

^aConditions unless otherwise noted: lactol 29 (1 equiv), 1:1 AcOH/H₂O, 60 °C.

^bAcOH as solvent.

^cIsolated yield.

Substitution on the aryl ring of the hydrazine component was also investigated (Table 3). It was found that para, meta, and ortho substituents were tolerated under the reaction conditions (entries 1–6). Importantly, use of chlorohydrazines furnished haloindolines (entries 4 and 5), which could be further functionalized by transition metal-catalyzed cross-coupling chemistry. The transformation proceeded smoothly with p-methoxyphenylhydrazine as a substrate, thus affording C5-oxygenated products in good yields (entry 6). However, use of p-(trifluoromethyl)phenylhydrazine as a substrate led to low yields of product (entry 7).

Table 3.

Variation of the hydrazine aryl substituents.

entrya	hydrazine	product	yieldb
1			60%
2			75% (4 : 3)
3			62%
4			67%
5			60%
6			X = OMe; 75%
7			X = CF3; < 5%

^aConditions unless otherwise noted: lactol 29 (1 equiv), 1:1 AcOH/H₂O, 60 °C.

^bIsolated yield.

The scope of the lactol component for furoindoline synthesis was examined in the interrupted Fischer indolization process (Table 4). Allyl and phenyl substituents were tolerated, thus providing fused indolines with alternate C3 substitution (entries 1 and 2). It should be noted, however, that substrates bearing either a t-butyl or a methyl ester substituent led only to trace amounts of product formation (entries 3 and 4). Nonetheless, a 6-membered homologue of the furoindoline framework was accessible using this methodology using our standard reaction conditions (entry 5).

Table 4.

Variation of the lactol component.

entrya	lactol	product	yieldb
1			89%
2			75%
3			<5%
4			<5%
5			65%

^aConditions: hydrazine 23 (1 equiv), 1:1 AcOH/H₂O, 60 °C.

^bIsolated yield.

2.2. Synthesis of pyrrolidinoindolines

Having established the viability of the interrupted Fischer indolization approach for the synthesis of furoindolines, we sought to develop the corresponding transformation that would enable the synthesis of pyrrolidinoindolines and related derivatives. Thus, hemiaminal 31 was prepared from the corresponding lactam following a known procedure, and then subjected to phenylhydrazine (23) in the presence of 1:1 H₂O/AcOH (Scheme 4). To our delight, the interrupted Fischer indolization reaction proceeded smoothly at 100 °C and delivered the desired indoline 32 in 88% yield.

Scheme 4.

Synthesis of pyrrolidinoindoline 32.

Analogous to our studies in the area of furoindoline synthesis, the interrupted Fischer indolization reaction was found to be an effective means to access a range of pyrrolidinoindolines. As shown in Table 5, a variety of arylhydrazines were tolerated in the transformation. Reactions of N-substituted arylhydrazines furnished the desired indoline products in good yield (entries 1–3), whereas a range of arylhydrazines bearing benzenoid substitution were deemed competent coupling partners (entries 4–9). Similar to the results obtained in our furoindoline studies, use of p-(trifluoromethyl)phenylhydrazine as a substrate led to low yields of product (entry 10).

Table 5.

Variation of the hydrazine component.

entrya	hydrazine	product	yieldd
1b			81%
2			83%
3			70%
4			71%
5			55% (4 : 3)
6			73%
7			77%
8			84%
9c			X = OMe; 70%
10			X = CF3; < 5%

^aConditions unless otherwise noted: hemiaminal 31 (1 equiv), 1:1 AcOH/H₂O, 100 °C.

^b23 °C, AcOH as solvent.

^c75 °C.

^dIsolated yield.

The scope of the hemiaminal component was also investigated (Table 6). C3-allylated and -phenylated pyrrolidinoindolines could be accessed without difficulty (entries 1 and 2). Of note, these pyrrolidinoindoline motifs are present in an array of medicinally important compounds, such as debromoflustramine B (5, Figure 1)^iva and the hodgkinsine alkaloids.^xxx Furthermore, a 6-membered homologue was prepared in 81% yield (entry 3) reminiscent of the communesin and perophoramidine core structures. Finally, it was determined that a carbamylated hemiaminal could be employed in place of a sulfonamide (entry 4).

Table 6.

Variation of the hemiaminal component.

entrya	hemiaminal	product	yieldb
1			68%
2			70%
3			81%
4			88%

^aConditions: hydrazine 23 (1 equiv), 1:1 AcOH/H₂O, 100 °C.

^bIsolated yield.

As shown in Figure 3, the N-substituents of our pyrrolidinoindoline products can easily be manipulated. The sulfonamide group of 33 was removed upon treatment with Mg and NH₄Cl in MeOH^xxxi to provide pyrrolidinoindoline 34 in 79% yield.^xxxii Additionally, carbamate 35 was converted to the corresponding N-methylated product 36 when reacted with Red-Al. The latter of these results is particularly notable given that many pyrrolidinoindoline natural products possess this N-methylated substitution pattern (e.g, Figure 1, 3–7).

Figure 3.

Manipulation of the N-substituent.

2.3. Formal total syntheses of physovenine and debromoflustramine B, and assembly of the communesin indoline scaffold

Having developed a powerful means to synthesize fused indoline ring systems, we examined the scope and limitations of our methodology in more complex settings. As shown in Scheme 5, the newly discovered transformation has been utilized to achieve a concise formal total synthesis of the furoindoline natural product physovenine (2).^xxxiii Reaction of hydrazine 37^xxxiv with lactol 29 in AcOH furnished furoindoline 38 in 77% yield, which has previously been converted to physovenine (2) in two additional steps.^xva Although asymmetric routes to intermediate 38 have previously been reported, our single step route to (±)-38 is substantially shorter (one step compared to 7,^xva or 18^xxxiiig steps). Furthermore physovenine (2) can be optically resolved, on preparative scale, using column chromatography with cellulose triacetate.^xxxiiio

Scheme 5.

Formal total synthesis of physovenine (2).

The interrupted Fischer indolization reaction could also be used to complete a formal total synthesis of the pyrrolidinoindoline natural product debromoflustramine B (5).^xxxv Pyrrolidinone 39^xxxvi was elaborated to hemiaminal 40 using a standard two-step sequence. Treatment of 40 with 1-allyl-1-phenylhydrazine in H₂O/AcOH at 100 °C facilitated the key condensation/sigmatropic rearrangement to deliver bis(allylated)pyrrolidinoindoline 41. In turn, 41 was reacted with 2-methyl-2-butene in the presence of Grubbs’ second generation catalyst to afford bis(prenylated) derivative 42,^xxxvii which was converted to 5 by reduction with Red-Al.

Finally, we explored the scope and limitations of our methodology in the context of the communesin natural products (Scheme 7).^xxxviii Known sulfonamide 43^xxxix was reacted with 1-ethoxypropene in the presence of Cs₂CO₃ to afford hetero-Diels–Alder product 44, following the general procedure described by Corey.^xxxix Exposure of 44 to N-methyl phenylhydrazine (26) in 1:1 AcOH/H₂O delivered indoline 45, which possesses the tetracyclic 6,5,6,6-ring system of the communesin alkaloids.^xl As noted earlier, the previously described [3,3]-sigmatropic rearrangement strategies for the synthesis of fused indoline ring systems are not amenable to this complex scaffold.

Scheme 7.

Synthesis of communesin indoline scaffold 45.

2.4. Access to enantioenriched indoline products

Having demonstrated that the interrupted Fischer indolization reaction provides an effective means to access indoline scaffolds, we hoped to uncover a variant that would give access to enantioenriched indoline products. The most appealing scenario to achieve this goal would involve asymmetric catalysis. Thus, efforts were put forth to carry out the interrupted Fischer indolization reaction in the presence of chiral non-racemic phosphoric acids.^xli

As shown in Scheme 8, this asymmetric transformation proved challenging. Despite an extensive survey of reaction conditions (e.g., variations in substrates, phosphoric acid promoter, stoichiometry, solvent, and temperature), only modest levels of enantioselectivity could be obtained. For example, reaction of hydrazine 23 and lactol 29 in the presence of 1.2 equivalents of phosphoric acid 46 (prepared from (R)-BINOL)^xlii in benzene at 40 °C provided furoindoline 30 in 62% yield and 28% ee.^xliii Similar results were obtained when hemiaminal substrates were employed in place of lactols.

Scheme 8.

Furoindoline synthesis using phosphoric acid promoter 46.

Given the difficulty in achieving a reagent or catalyst-controlled asymmetric interrupted Fischer indolization, we turned to an auxiliary-based approach.^xliv Thus, Nishida’s enantioenriched arylhydrazine 50 was prepared using the sequence shown in Scheme 9.^{xxve,xlv,xlvi} Bromobenzene (47) was coupled with commercially available enantioenriched amine (–)-48 under Pd catalysis to provide aniline 49. Using a standard protocol, aniline 49 was converted to the targeted hydrazine 50 in 81% yield over two steps.

Scheme 9.

Synthesis of arylhydrazine 50.

The utility of arylhydrazine 50 in our interrupted Fischer indolization process was evaluated in the context of furoindoline synthesis (Figure 4). Gratifyingly, the reaction of 50 and lactol 29 proceeded smoothly under a variety of acidic conditions. When the reaction was carried out in the presence of 3 equivalents of chloroacetic acid in benzene at 40 °C, an 80% yield of diastereomeric indoline products 51 and 52 was obtained (d.r.=2.4:1).^xlvii The isomers were easily separable using conventional flash column chromatography on silica gel. The major isomer 51 was treated with Pd(OH)₂ and 1,4-cyclohexadiene in EtOH to remove the auxiliary and deliver optically enriched indoline 30.^xlviii,xlix The ee of 30 was found to be 97%,^xxii thus demonstrating that our methodology can be utilized to access enantioenriched products. The absolute configuration of 30 was determined based on correlation to known data,^xxxiiic and was found to be as depicted in Figure 4.

Figure 4.

Synthesis of enantioenriched 30.

3. Conclusions

In summary, we have developed an efficient method to access the fused indoline ring systems present in a variety of natural products. Our interrupted Fischer indolization strategy involves the condensation of readily available hydrazines with latent aldehydes to deliver indoline-containing products by way of a tandem [3,3]-sigmatropic rearrangement / cyclization cascade sequence. The method is convergent, mild, operationally simple, broad in scope, and can be used to access enantioenriched products. In addition, our approach is amenable to the synthesis of furoindoline and pyrrolidinoindoline natural products as demonstrated by the concise formal total syntheses of physovenine and debromoflustramine B. We expect that the interrupted Fischer indolization strategy will enable the synthesis of more complex targets such as the communesins and akuammiline alkaloids. Such studies in the realm of natural product synthesis are currently underway in our laboratory.

4. Experimental

4.1 Representative experimental procedure for furoindoline synthesis (Table 2, entry 1)

Lactol 29 (126 mg, 1.22 mmol) was dissolved in a 1:1 mixture of acetic acid and water (6 mL). Phenylhydrazine (23) (0.121 mL, 1.23 mmol) was added to the resulting mixture. The reaction was heated to 60 °C for 4.5 h, then cooled to 23 °C, and quenched with a solution of sat. aq. NaHCO₃ (15 mL). The resulting mixture was extracted with EtOAc (3 × 15 mL). The combined organic layers were dried over MgSO₄. Evaporation of the solvent under reduced pressure afforded the crude product. Purification by flash chromatography (7:1 hexanes:EtOAc) afforded indoline 30 (196 mg, 89% yield). Rf 0.7 (1:1 EtOAc:hexanes); ¹H NMR (300 MHz, CDCl₃): δ 7.08 (d, J = 7.2, 1H), 7.05 (t, J = 7.5, 1H), 6.76 (t, J = 7.5, 1H), 6.59 (d, J = 7.8, 1H), 5.28 (s, 1H), 3.96 (ddd, J = 8.4, 7.2, 1.8, 1H),3.56 (ddd, J = 10.8, 8.4, 5.1, 1H), 2.13 (ddd, J = 11.7, 5.4, 1.5, 1H), 2.07 (ddd, J = 11.7, 7.2, 4.2, 1H), 1.47 (s, 3H); ¹³C NMR (125 MHz, CDCl₃): δ 148.8, 133.9, 127.9, 122.9, 118.8, 108.8, 99.5, 67.3, 53.8, 41.4, 24.7; IR (film): 2967, 2845, 1611, 1486, 1265, 1055 cm⁻¹; HRMS-ESI (m/z) [M + H]⁺ calcd for C₁₁H₁₄NO, 176.1075; found 176.1078.

4.2. Representative experimental procedure for pyrrolidinoindoline synthesis (Table 6, entry 1)

3-Allyl-1-tosylpyrrolidin-2-ol^xxii (105 mg, 0.37 mmol) was dissolved in a 1:1 mixture of acetic acid and water (1.8 mL). Phenylhydrazine 23 (0.036 mL, 0.36 mmol) was added to the resulting mixture. The reaction was heated to 100 °C for 1 h 40 min, cooled to 23 °C, and then diluted with Et₂O (20 mL). The reaction mixture was then quenched with sat. aq. NaHCO₃ (20 mL), and the layers were separated. The aqueous layer was extracted with Et₂O (3 × 20 mL). The combined organic layers were washed with brine (20 mL), dried over MgSO₄, and concentrated in vacuo to afford the crude indoline product. Purification by flash chromatography (18:1:1 benzene:Et₂O:CH₂Cl₂) afforded the 3-allyl-indoline (Table 6, entry 1) as a yellow oil (88.0 mg, 68% yield). Rf 0.6 (8:1:1 benzene:Et₂O:CH₂Cl₂); ¹H NMR (500 MHz, CDCl₃): δ 7.75 (d, J = 8.0, 2H), 7.32 (d, J = 8.0, 2H), 7.08 (t, J = 7.5, 1H), 7.00 (d, J = 7.0, 1H), 6.75 (t, J = 7.5, 1H), 6.62 (d, J = 7.5, 1H), 5.55 (ddt, J = 16.8, 10.0, 7.5, 1H), 5.13 (s, 1H), 4.96–5.00 (m, 2H), 4.84 (s, 1H), 3.43 (ddd, J = 10.0, 8.0, 2.0, 1H), 3.13 (ddd, J = 10.5, 10.5, 6.0, 1H), 2.44 (s, 3H), 2.31 (ddd J = 19.5, 13.5, 7.5, 2H), 2.07 (ddd, J = 6.,5, 6.0, 2.0, 1H), 1.84 (ddd, J = 10.5, 8.0, 6.5, 1H); ¹³C NMR (125 MHz, CDCl₃): δ 149.1, 143.7, 136.4, 133.5, 131.4, 130.0 128.8, 127.3 123.2, 119.3, 118.8, 109.7, 82.7, 58.0, 47.6, 42.3, 36.2, 21.7; IR (neat): 3391, 3076, 1611, 1485, 1337, 1160 cm⁻¹; HRMS-ESI (m/z) [M+Na]⁺ calcd for C₂₀H₂₂N₂O₂SNa, 377.1300; found, 377.1298.

4.3 Synthesis of indoline diastereomers 51 and 52

To a mixture of aryl hydrazine 50 (52.4 mg, 0.20 mmol), lactol 29 (20.6 mg, 0.20 mmol), and benzene (1 mL) was added chloroacetic acid (56.7 mg, 0.60 mmol). The resulting mixture was heated at 40 °C for 24 h. The reaction mixture was cooled to 23 °C, diluted with CH₂Cl₂ (20 mL), washed with sat. aq. NaHCO₃ (5 mL) and extracted with CH₂Cl₂ (10 mL). The combined organic layers were dried over MgSO₄ and evaporated to dryness. Purification by flash chromatography (15:1 →10:1 hexanes:EtOAc) afforded a mixture of diastereomers as an orange solid (53.0 mg, 80% yield, 2.4:1 dr). To separate the diastereomers the mixture was repurified by flash chromatography, under the same conditions. The stereochemical configurations of 51 and 52 were inferred after the conversion of 51 to (+)-30. Indoline 51: Rf 0.5 (4:1 hexanes:EtOAc); ¹H NMR (500 MHz, CDCl₃) δ 8.23 (d, J = 8.0, 1H), 7.96 (d, J = 7.5, 1H), 7.82 (t, J = 8.0, 2H), 7.60 (t, J = 7.0, 1H), 7.56 (t, J = 7.5, 1H), 7.49 (t, J = 8.0, 1H), 7.11 (d, J = 7.5, 1H), 6.91 (t, J = 7.5, 1H), 6.03 (d, J = 7.5, 1H), 5.65 (s, 1H), 5.56 (q, J = 7.0, 1H), 4.03 (t, J = 8.0, 1H), 3.70–3.75 (m, 1H), 2.28 (dd, J = 11.5, 4.5, 1H), 1.86 (d, J = 6.5, 3H), 1.58 (s, 3H); ¹³C NMR (125 MHz, CDCl₃) δ149.1, 139.4, 134.5, 133.9, 130.8, 129.1, 127.8, 127.5, 125.9, 125.8, 125.2, 123.6, 122.5, 122.4, 117.3, 106.0, 101.4, 66.7, 52.3, 51.7, 41.9, 25.8, 19.0; IR (neat): 3046, 2963, 2852, 1606, 1594, 1487, 1459, 1395, 1298, 1236, 1013 cm⁻¹; HRMS-ESI (m/z) [M + Na]⁺ calcd for C₂₃H₂₃NONa, 352.1677; found 352.1686; [α]_D^24.4 + 125.4 (c 0.01, CHCl₃). Indoline 52: Rf 0.5 (4:1 hexanes:EtOAc); ¹H NMR (500 MHz, CDCl₃) δ 7.95 (d, J = 8.5, 1H), 7.90 (d, J = 8.0, 1H), 7.84 (d, J = 8.5, 1H), 7.74 (d, J = 7.5, 1H), 7.45–7.54 (m, 3H), 7.14 (t, J = 8.0, 1H), 7.10 (d, J = 7.0, 1H), 6.75 (d, J = 7.0, 1H), 6.51 (d, J = 7.5, 1H), 5.52 (q, J = 7.0, 1H), 4.69 (s, 1H), 3.93 (t, J = 7.5, 1H), 3.53 (ddd, J = 13.0, 8.5, 4.5, 1H), 2.18 (dd, J = 12.0, 4.5, 1H), 1.98 (ddd, J = 11.5, 11.5, 7.0, 1H), 1.90 (d, J = 6.5, 3H), 1.24 (s, 3H); ¹³C NMR (125 MHz, CDCl₃) δ 148.8, 136.3, 134.7, 133.8, 131.7, 128.7, 128.1, 128.0, 126.2, 125.5, 125.4, 124.2, 123.2, 122.7, 117.3, 105.0, 101.0, 67.0, 52.2, 49.3, 41.2, 24.7, 18.1; IR (film): 3681, 2973, 2845, 1605, 1487, 1215, 1059 cm⁻¹; HRMS-ESI (m/z) [M + Na]⁺ calcd for C₂₃H₂₃NONa, 352.1677; found 352.1680; [α]_D^24.2 –54.6 (c 0.01, CHCl₃).

4.4 Synthesis of indoline (+)-30

A mixture of indoline 51 (32.9 mg, 0.1 mmol), 1,4-cyclohexadiene (80.0 mg, 1.0 mmol), and palladium hydroxide (20% wt on carbon, 10.0 mg) in ethanol (1 mL) was heated at 80 °C for 6 h. The reaction mixture was cooled to 23 °C, filtered through celite, washed with CH₂Cl₂ (10 mL), and the solvent was removed under reduced pressure. Purification by flash chromatography (5:1 hexanes:EtOAc) furnished furoindoline (+)-30 (14.0 mg, 80% yield, 97% ee). [α]_D^24.3 +124.5 (c 0.01, CHCl₃), SFC (CHIRALPAK AS-H, CO₂/MeOH = 9/10, flow 1.5 mL/min, at 23 °C, detection at 254 nm) t_R 3.06 min (major) and t_R 4.43 min (minor). The absolute configuration of 30 was determined based on correlation to known data.^xxxiiic

Scheme 6.

Formal total synthesis of debromoflustramine B (5).