Synthesis and evaluation of analogs of tetracyclic Ganoderma alkaloids as inhibitors of the enzyme α-glucosidase

Abstract

A series of synthetic analogs of tetracyclic Ganoderma alkaloids were prepared via short routes in moderate overall yields by the reaction of 3-formylchromene derivatives with enamines prepared from cyclic 1,3-dicarbonyl compounds. Single-crystal X-ray analysis of one of the tetracyclic products confirmed that they embody original tetrahydro-6H-chromeno[4,3-b]quinoline (and tetrahydrochromeno[4,3-b] cyclopenta[e]pyridine) skeletons. When tested as α-glucosidase inhibitors, these analogs proved to be up to 40 times more potent than acarbose, used as the reference inhibitor.

---

Introduction

Type 2 diabetes mellitus is a chronic metabolic disease characterized by persistent hyperglycemia, resulting from insulin resistance and pancreatic β-cell dysfunction.^1a It is a major global health issue that affects over 500 million individuals and is the fourth leading cause of death in most developed countries.^1b Therefore, great efforts have been made to achieve a better understanding of the disease and improve its clinical management.

One of the pharmacological approaches devised to treat this condition entails the inhibition of key carbohydrate digestive enzymes, such as α-amylase^1c and α-glucosidase.^1d However, despite the high quality of the modern pharmacological arsenal,^1e the prevalence of the disease continues to grow and some of its complications remain challenging to control. Hence, the search for new inhibitors of these enzymes with novel structures is an active field of research.

Ganoderma is a genus of basidiomycete fungi that have been used for centuries in many Asian cultures for the treatment of several diseases.² Many species are known; among them, G. lucidum, G. sinense, G. capense, G. cochlear, and G. tsugae play important roles in folk medicine in various Southeast Asian countries.

For example, G. lucidum has long been used in traditional Chinese medicine to treat diabetes mellitus and recently it has been shown to enhance the blood glucose-lowering effects of modern drugs.³ In addition, some fungal polysaccharides^4a and proteoglucans have been demonstrated to possess activity in type 1 diabetic mice,^4b,c increasing blood insulin levels, stimulating the activity of liver enzymes related to carbohydrate metabolism, and thus accelerating the metabolism of glucose.^4d

The chemical constituents and biological activities of 25 species of Ganoderma have been recently reviewed.⁵ Different meroterpenoids (such as ganomycins)^6a,b have been shown to display aldose reductase⁷ or α-glucosidase^3c,8 inhibition, among other effects.^7b,9

These activities have also been found in some lanostane triterpenoids,^6c such as ganoderic acids (Fig. 1), ganodeweberiols, lucidenic acid esters, ganoleucins, and ganoleuconins, among others.

Fig. 1 Some anti-diabetic triterpenoids and meroterpenoids along with selected tetracyclic alkaloids isolated from different species of Ganoderma.

Despite their abundance in nature and wide range of biological and pharmacological activities, which often attract considerable attention, alkaloids remain among the least evaluated classes of compounds in the context of diabetes mellitus.¹⁰ Studies referring to Ganoderma alkaloids are relatively few; hence, this class of natural products and their bioactivities remain unknown or largely unexplored.

Ganoines I and II, isolated from G. capense in 1990, were the first alkaloids found in this genus,¹¹ whereas the tricyclic ganocochlearines A and B were the first alkaloids isolated from G. cochlear.¹² In 2017, additional ganocochlearines (C–I) were isolated from the latter species, with ganocochlearines H and I having skeletons that differ from those of their tetracyclic congeners.¹³ Most of these heterocycles were shown to lack toxicity toward normal cells and did not inhibit the proliferation of NRK-49F fibroblasts under TGF-β1-induction.

However, Chen and Lan recently synthesized the related lucidimines B and C (isolated from G. lucidum) and examined their bioactivity.¹⁴ They found that lucidimine B exhibited antiproliferative properties toward MCF-7 cells; in addition, it proved to be a better antioxidant than lucidimine C. On the other hand, ganocalicine A, from G. calidophilum, was shown to be an anti-allergic agent.¹⁵

Therefore, with a view to exploring the chemical space of structures inspired by this class of understudied compounds, and in continuation of our research work on the synthesis and biological evaluation of heterocyclic natural products and their analogs,¹⁶ herein we disclose our results related to the synthesis of analogs of tetracyclic Ganoderma alkaloids and their evaluation as α-glucosidase inhibitors.

Results and discussion

Retrosynthetic analysis

The heterocycles were synthesized according to the guidelines obtained from the retrosynthetic analysis displayed in Scheme 1. It was envisaged that the proposed analogs could be accessed through the late-stage construction of the central pyridine ring, followed by modifications of the functionalities attached to rings A and D. Accordingly, it was conjectured that the key heterocyclic ring C could be formed through a 6π-1-azaelectrocyclization.¹⁷

Scheme 1 Retrosynthetic analysis of the projected analogs of Ganoderma alkaloids.

On this basis, one of the C–N bonds of the prototypic target 1 was strategically disconnected, revealing 1-azatriene 2 as a potential advanced intermediate. A more in-depth analysis suggested disconnection of the triene at its central bond, assuming that the latter could be built by direct condensation between an enamine (3) and a 3-formylchromene (4).

In turn, enamine 3 could be derived from the corresponding cyclic 1,3-dicarbonyl precursor (5), where the additional carbonyl group could add a point of variation and mimic the oxygen functionalities found in the natural products. On the other hand, the required 3-formylchromene partner could be formed by the reaction of a conveniently functionalized salicylaldehyde derivative (6) and acrolein (7), through a tandem oxa-Michael addition/aldol condensation process.

Chemical synthesis

According to the retrosynthetic plan, which followed an A → AB → ABD → ABCD constructive approach, the synthesis commenced with the preparation of the starting aldehyde 9 (Scheme 2).

Scheme 2 Reagents and conditions: (a) Me₂SO₄, NaHCO₃, PhMe, 80 °C, 4 h (44%); (b) NaOH, H₂O, CHCl₃ (8 equiv.), 70 °C, 4 h (62%); (c) 7 (1.5 equiv.), dioxane, 100 °C, 6 h (9 → 12, 56%; 8 → 11, 80%); (d) MeI, K₂CO₃, DMF, rt, overnight (96%); (e) NH₄OAc, Si(OEt)₄, MeCN, reflux (100%); (f) FeCl₃, MeCN, MW (150 °C), 60 min (13, 17%, 16, 20%; 17, 19%; 18, 10%); (g) NaBH₄, MeOH : CH₂Cl₂ (1 : 10, v/v), rt, 18 h (75%); (h) I₂/Al–DMSO, MeCN rt, or NaS^tBu, DMF, 90 °C, 1.5 h (decomposition).

Initially, it was observed that the O-methylation of aldehyde 8 (Me₂SO₄, NaHCO₃, PhMe, reflux) proved to lack selectivity, whereas the Casnati–Skattebøl and Duff formylations of 4-methoxyphenol (10) gave unsatisfactory results (low yields, diformylation). In contrast, the Reimer–Tiemann reaction cleanly provided the expected product 9 in 62% yield when 8 equiv. of CHCl₃ were employed, and the reaction was carried out at 70 °C; not surprisingly, however, the yield dropped to 15% with the use of only 2 equiv. of CHCl₃.

Next, the transformation of 9 into formylchromene 12 was studied¹⁸ and the product was obtained in 56% yield when K₂CO₃ was used as a base in the presence of freshly prepared acrolein (7)¹⁹ in refluxing dioxane.

The reaction is likely to involve an initial conjugate addition of phenol to 7, a strong Michael acceptor, to furnish a 1,7-dicarbonyl system intermediate (i). In turn, the resulting acidic α-carbonyl methylene unit would cyclize by attacking the carbonyl of the benzaldehyde motif (ii), and this should ultimately dehydrate to furnish the formylchromene,²⁰ completing the tandem oxa-Michael addition–aldol condensation sequence.

Interestingly, the analogous heterocycle 11 was obtained in 80% yield when the reaction was carried out with the commercial aldehyde 8 at 100 °C (0.18 M level). It was observed that the performance of this transformation is sensitive to the concentration of the reagents, with its yields diminishing to 53% when conducted at the 0.45 M level. Methylation of 11 with MeI in DMF with K₂CO₃ as the base²¹ furnished 12 in 96% yield, turning the latter into an improved and more versatile route toward the 3-formylchromene fragment.

On the other hand, the formation of enamine 3 was first attempted mechanochemically by the reaction of 1,3-dione 5 with NH₄OAc under KHSO₄/SiO₂ promotion.²² This resulted in a 2 : 1 mixture of the product with the starting ketone, as per ¹H NMR integration of the diagnostic H-2 signals (δ = 5.29, s, 1H). For this reason, the crude enamine preparation was used in the reaction with 3-formylchromene 12. When it was carried out in MeCN using FeCl₃ as the promoter and oxidant,²³ it afforded a single isolated product in 17% yield, after heating under microwave irradiation at 150 °C for 1 h (Table 1, entry 1).

Table 1 Optimization of the cyclocondensation toward 13 ^a

Entry no.	Conditions		Yield (%)
	Enamine formation	Cyclization
a When the reactions were run in two steps, enamine formation was carried out first by the reaction between 5 and NH4OAc under the stated conditions, before running the cyclization stage with the crude enamine preparation. b The volatiles were removed in vacuo before running the cyclization stage.
1	KHSO4/SiO2, solventless, 0.25 h	FeCl3, MeCN, MW, 150 °C, 1 h	17
2	Si(OEt)4, MeCN, reflux, 1 h	FeCl3, MW, 80 °C, 1 h	18
3	Si(OEt)4, MeCN, MW, 150 °C, 1 h	FeCl3, O2, MW, 150 °C, 0.5 h	12
4	Si(OEt)4, MeCN, MW, 150 °C, 1 h	CuCl2, MW, 120 °C, 2 h	11
5	1. InCl3 (10 mol%), EtOH, rt, 2 h		16
	2. 80 °C, 36 h (dihydropyridine 13a, 84%)
	3. DDQ, THF, −30 °C → −5 °C
6	Si(OEt)4, PhMe, reflux, 48 h		26
7	Si(OEt)4, PhMe, reflux, 48 h (purif. on Al2O3)		40
8	Si(OEt)4, EtOH, reflux, 1 h^b	Zn(OTf)2, DMF, 75–100 °C, 48 h	45
9	DABCO, MeCN, rt, 1 h	70 °C, 3 h	20
10	Dioxane, rt, 0.5 h	DABCO, K2CO3, O2, reflux, 5 d	24
11	DCE, 80 °C, 2 d	K2CO3, I2, O2, reflux, 5 d	17

---

Suspecting that the observed result could be due to incomplete enamine formation, Si(OEt)₄ was employed as a H₂O scavenger, and the reaction was performed in refluxing MeCN,²⁴ which enabled a facile and efficient preparation of the required intermediate. However, when the latter was reacted with 12 under FeCl₃ promotion at 80 °C, the product was obtained in an analogous 18% yield (entry 2).

The product was subjected to a careful structural analysis. The conventional mechanism of a typical Stork's enamine alkylation of α,β-unsaturated carbonyls involves a Michael addition of the enamine to the β-carbon atom of the latter favored by the electrons of the nitrogen of the enamine,²⁵ resulting in the formation of a new C–C bond, which in the case of 12 and 3 would ultimately lead to the skeleton of 19, reminiscent of ganocochlearine I (Scheme 3).

Scheme 3 Proposed cyclization mode toward 19 based on the typical reactivity of the Stork's enamine alkylation of 12 with 3.

In light of previous results,²⁶ an analogous outcome would be expected from the conjugate addition of the 1,3-dicarbonyl precursor (5) to the enal moiety. Therefore, it was deemed necessary to carefully examine the reaction product by NMR to ensure unequivocal differentiation between tetracycle 19 and its isomer 13. In this case, the signals of the C_sp²–H motif of the pyridine ring (C7/H7 in 19) were considered the best diagnostic elements.

Examination of the ¹H NMR spectrum of the product revealed the presence of four hydrogens attached to sp² carbon atoms, from which it was possible to unequivocally assign three of them to the isocyclic aromatic ring, based on comparison with its precursor 12 and analysis of the 2D spectra (COSY, HSQC and HMBC). The remaining C_sp²–H signal [δ_H = 8.00 ppm (1H, s) and δ_C = 130.9 ppm] clearly indicated that 13 was the cyclization product (Scheme 2) because a more deprotected signal, such as δ_C7 ∼ 150 ppm, would be expected for 19, which has this C_sp²–H motif attached to the pyridine N-atom.

Considering that suitable crystals of the product were obtained from EtOAc, in order to get definitive structural evidence, single crystal X-ray diffraction analysis of the heterocycle was undertaken (Fig. 2). It was observed that the tetracycle crystallized in the monoclinic system, in the P2₁/n space group with four molecules per unit cell (Z = 4). In addition, at room temperature, the solid exhibited some disorder in the isocyclic ring (Fig. 2A), specifically at C2, C3, and C4 and in the heavy atoms of the OMe motif; however, the refinement data clearly pointed out to 13 as the structure of the tetracyclic reaction product.

Fig. 2 ORTEP diagrams of tetracycle 13 with the thermal ellipsoids drawn at their 50% level. (A) At room temperature, displaying distortion at C2–C4 and the attached OMe group (the C2–O20–C21 bonds are colored for the sake of clarity) and (B) at 80 K.

This conjecture was confirmed by performing the diffraction at 80 K (Fig. 2B), which improved accuracy, giving a definitive picture of the product structure and confirming that its room temperature disorder is thermal in nature (dynamic disorder). Under these conditions, it was observed that compound 13 displays an essentially planar structure, where both aromatic rings are slightly twisted, as evidenced by their small interplanar angle (10.31°). It was also noticed that the methyl moiety of the OMe group is oriented toward the C3 ortho side and is slightly out of plane [dihedral angle C3–C2–O20–C21 = −26.83(16)°]. In addition, C7, C13 and C21 are observed on the same side of the pyridine ring plane, whereas the carbonyl oxygen faces the opposite direction [C9–C10–C11–O19 = −6.87(16)°].

Having established the identity of the tetracyclic product 13, an optimization was undertaken (Table 1). It was observed that the use of aerobic conditions led to a diminished yield (12%; entry 3), while the performance did not improve when CuCl₂ was used as the promoter and oxidant²⁷ (11%, entry 4). On the other hand, when the reaction was performed as a one-pot process with InCl₃ as the promoter, the dihydropyridine intermediate 13a was formed in a good yield (84%);²⁸ however, further oxidation of the latter with DDQ in THF resulted in only 16% yield of the tetracyclic product 13 (entry 5), accompanied by a series of unidentified compounds.

In view of the above results, one-pot procedures were attempted (entries 6 and 7) in refluxing toluene. Unexpectedly, chromatographic product purification on an alumina column improved the yields up to 40% (entry 10). In addition, slightly better results (45% yield) were obtained when the enamine was prepared in refluxing EtOH and the cyclization was performed in DMF under Zn(OTf)₂ promotion,²⁹ after changing the solvent (entry 8).

Similarly, the use of bases (K₂CO₃, DABCO) to promote enamination (entry 9) and further cyclization (entries 9–11) failed to significantly improve the yield of the product, even when run under aerobic conditions³⁰ (entry 11).

Next, the reaction was tested with enamines 3a and 3b, derived from the related 5,5-dimethyl cyclohexane-1,3-dione (dimedone, 5a) and 4,4-dimethyl cyclohexane-1,3-dione (5b). It was observed that their performances were similar, since they afforded analogs 16 and 17 in 20% and 19% yields, respectively (Scheme 2). However, a diminished product yield was recorded (18, 10%) when 3c, the enamine derivative of cyclopentane-1,3-dione (5c), was employed instead.

The case of 5b is interesting, because only one of the two possible regioisomeric tetracycles (17) was isolated. This is a result of the selective formation of a single enamine (3b) through the reaction of the added NH₄AcO with the less hindered carbonyl moiety,³¹ as evidenced by the diagnostic signal of H-2 observed as a singlet in its ¹H NMR spectrum at δ = 5.14 ppm.

On the other hand, when the regioisomeric tetracycles 16 and 17 were compared, it was observed that the carbon atoms of the gem-dimethyl motif of 16 were less protected in their ¹³C NMR spectra than those of tetracycle 17 (δ = 28.5 vs. 24.3 ppm). However, the hydrogen nuclei of the methyl groups of the latter compound were more deshielded than those in 16 (δ = 1.24 vs. 1.14 ppm), as a result of their proximity to the carbonyl moiety.

The tetracyclic ketone 13 was uneventfully reduced with NaBH₄ in MeOH : CH₂Cl₂ (10 : 1, v/v) to afford alcohol 14. However, attempts to access phenol 15 by demethylation of 14 with the I₂/Al–DMSO system in MeCN at room temperature³² or with sodium mercaptides (NaSEt or NaS^tBu in DMF, 90 °C, 1.5 h) met with failure, leading to decomposition of the tetracycle into unidentifiable products. This was attributed to the presence of a competing ether moiety in the oxygen heterocyclic motif.

In addition, attempts at performing the cyclization with 3-formylchromene 11 bearing a free phenol resulted in the decomposition of the starting material. Therefore, an alternative protecting group was sought that could withstand the cyclization conditions and be more easily removable (Table 2).

Table 2 Protection of 11 and cyclization of the resulting 3-formylchromene

Entry no.	Protection			Cyclization
	Enal no.	R	Yield (%)	1,3-Diketone	n	Product no.	Yield (%)
a The reaction was carried out in DMF, in the presence of Zn(OTf)2, at 75–100 °C for 48 h.
1	20	Ac	93	5	1	Decomposition
2	21	TBS	99	5	1	24	27
3	22	Bn	85	5	1	Decomposition
4	23	MOM	91	5	1	25	30
5				5	1	25	30^a
6				5c	0	26	15

---

TBS, Bn and MOM ethers along with the acetate derivative of 11 were prepared in 85–99% yields.³³ However, both the acetate (entry 1) and the benzyl ether (entry 3) underwent decomposition when exposed to the cyclization conditions and the remaining protecting groups gave meagre yields of the products (entries 2, 4 and 6). Unfortunately, the yields did not improve under other conditions; for example, cyclization of the MOM derivative 23 in the presence of Zn(OTf)₂ (as in Table 1, entry 8) gave 25 in only 30% yield (Table 2, entry 5).

Next, the ketone moiety of tetracycle 25 was reduced with NaBH₄ in a MeOH/CH₂Cl₂ (10 : 1, v/v) medium, affording 27 in 95% yield (Scheme 4), while deprotection of the phenol by transacetalization with refluxing ⁱPrOH under Amberlyst-15 promotion for 18 h (ref. 34) provided 28 in 54% yield (62% overall yield when the sequence was performed as a one-pot process).

Scheme 4 Reagents and conditions: (a) MeOH, CH₂Cl₂ (10 : 1, v/v), 8 h (95%); (b) Amberlyst-15, ⁱPrOH, reflux, 18 h (54%); (c) Amberlyst-15, ⁱPrOH, reflux, 48 h (25 → 29, 88%; 26 → 30, 15%).

Similarly, when the MOM protecting group was removed from 25 and 26, 29 and 30 were afforded in 88 and 15% yields, respectively. The yield of the deprotection of 29 was unexpectedly low, compared with similar transformations (25 → 29, 88%; 27 → 28, 54%). Although the ultimate causes for this behavior remain unknown, it resembles the differences observed during the cyclization stage, which was more efficient with the six-membered 1,3-dicarbonyl compounds than with their five-membered congeners.

Mechanistic considerations

The low product yields and selectivity observed during the transformations, even at temperatures as low as 70 °C, prompted an investigation into their possible causes. Although no identifiable products besides the expected tetracycles were detected under the different conditions, the use of copper(II) catalysis and milder conditions enabled the isolation of a small amount of chromene 31, suggesting that deformylation may be one of the operating degradation routes for the starting material (Scheme 5). This event does not seem to be uncommon, in light of the fact that the deformylation of 3-formylchromenes under noble metal catalysis (Pd, Rh) has been repeatedly reported.³⁵

Scheme 5 Possible decomposition pathways of the starting 3-formyl chromene 12 under the cyclocondensation conditions.

On the other hand, it has been proposed that under thermal conditions, chromenes of this type can undergo a dearomatizative retro-6π-oxaelectrocyclization;³⁶ this would result in a highly reactive conjugated tetraene-dicarbonyl species (32), which would likely be prone to undergo further decomposition.

An additional explanation was derived from careful inspection of the possible reaction mechanism (Scheme 6, Fig. 3), related to the strategic disconnections made during the retrosynthetic analysis. Under the assistance of the pair of electrons of the nitrogen atom of the enamine (3), the direct attack of the latter on the carbonyl carbon atom of 12 and the corresponding [1,5]H-shift would result in intermediate 34 through TS33. Next, dehydration of the thus formed allylic alcohol moiety would afford the 1-azatriene intermediate.

Scheme 6 A possible reaction mechanism for the synthesis of tetracycle 13 from 12 and 3. Energies are in kcal mol⁻¹.

Fig. 3 Energy diagram for the formation of 34en route to the cyclized product 13a, and its competitive alternative 38. Energies are in kcal mol⁻¹. Bond forming/breaking distances shown in TS33, TS36 and TS37 are in angstroms (Å) and Wiberg indexes are shown in parentheses.

Interestingly, although dehydration could result in a pair of geometric isomers, only those with the E-configuration (35E) could then be able to undergo a 6π-1-azaelectrocyclization through TS36 to form a 1,2-dihydropyridine (13a); in contrast, the Z-configured congener is proposed to undergo decomposition, consistent with the fact that despite many efforts, intermediates of this type could not be isolated. Upon oxidation in the reaction medium, 13a would afford 13. Dihydropyridines have been oxidized with an ample variety of reagents.^37a In our case, however, the use of added oxidants such as Cu(II) and Fe(III) salts, as well as DDQ and O₂ did not result in improved yields (Table 1).

Although efforts have been made to identify the required “internal” oxidant, they have not been successful; however, it is likely that the starting materials or the product itself may be involved in this process, adding a new explanation for the observed diminished yields. This is not unlikely because 1,4-dihydropyridines formed from 1,5-diketones and primary amines are known to disproportionate in AcOH to the corresponding tetrahydropyridine and pyridine derivatives,^37b hydrogen transfer reactions among pyridine derivatives have been documented,^37c and it has been observed that certain α,β-unsaturated carbonyl derivatives can oxidize Hantzsch dihydropyridines into the corresponding pyridines in a reaction driven by aromatization.^37d

To better understand the reaction selectivity, a theoretical study was performed using DFT calculations at the m062x/6-311 g(d,p) theoretical level and employing toluene as the solvent. It was speculated that the carbon atom of 3-aminocyclohexanone (3) can attack the 3-formylchromene (12) either at the carbonyl moiety (1,2-addition) or at C4 of the chromene moiety (1,4-addition). Therefore, two attacks were modelled (Fig. 3, Scheme 6), observing the involvement of proton transfers from the amine to the oxygen of the carbonyl moiety in both cases, and an activation energy difference of 3.21 kcal mol⁻¹ between TS33 (favored) and TS37 (disfavored).^37e

The alternative transfer of H-4 to the carbonyl oxygen was also modelled and considered unlikely because the process has a free energy difference of 10.81 kcal mol⁻¹. Additional calculations further demonstrated that the 1,2-adduct (34, ΔG = 11.82 kcal mol⁻¹) is thermodynamically favored over the 1,4-adduct (38, ΔG = 15.61 kcal mol⁻¹).

The molecular orbitals (MO) of the starting materials, along with natural bond orbitals and the thermodynamic parameters of the reaction were also examined. In the MO analysis, it was observed that both carbons can act as acceptors in 12, with coefficients of 0.1134 and 0.2072 for the carbonyl and β-carbon atom, respectively.

These partial results seemed to be somehow in disagreement with the experimental and calculated activation energies. However, further analysis of the natural population indicated that the charge difference on the carbonyl carbon in the starting material and TS is −0.1232 for the 1,2-addition pathway, while an analogous charge difference on the β-carbon is −0.004316 for the 1,4-addition pathway. This charge transfer difference is significant and sufficient to explain the observed preference for the 1,2-addition route.

In the 1,2-addition transition state, the NBO analysis of TS33 identified the C–C bond as partially formed, appearing as a BD(C–C) orbital interacting with its corresponding BD*(C2–C19) orbital. This indicates that electron donation from the nucleophile to the antibonding orbital of the nascent C–C bond stabilizes the transition state. The associated second-order perturbation energy is 4.28 kcal mol⁻¹, significantly higher than the interaction observed in TS37 (1.73 kcal mol⁻¹). Thus, although the frontier orbital coefficients initially suggested reactivity at the β-carbon atom, NBO analysis of the transition states revealed a stronger orbital interaction in the 1,2-addition pathway, justifying its lower activation barrier and the kinetic preference.

After ruling out the 1,4-addition pathway, the cyclization process toward the pyridine moiety was examined, beginning with the dehydration of 34. It was observed that the formation of the E-configured triene 35E, suited for subsequent electrocyclization, is endothermic (ΔG = 2.76 kcal mol⁻¹) and only slightly energetically preferred (ΔΔG = 0.50 kcal mol⁻¹) over the dehydration toward its Z-configured congener 35Z. This result provides an additional explanation for the observed moderate reaction yields, as the latter undergoes decomposition instead of the 6π-electrocyclization.

Regarding the electrocyclization step (Scheme 6), the pathway presents an activation energy of ΔG* = 16.82 kcal mol⁻¹ (TS36) and a thermodynamic value of ΔG = −18.13 kcal mol⁻¹ for the cyclized intermediate (13a). This result suggests that the formation of the tetracyclic motif takes place with a lower activation energy than that calculated for the formation of isoquinolines in aromatic systems.³⁸ The HOMO analysis showed orbitals located in the azatriene system, corresponding to a disrotatory electrocyclization.

Biological activity

The inhibitory activity of a small library of tetracycles displaying variability at the C2 functionality and the D-ring (including size, substitution pattern, and degree of oxidation) was studied against α-glucosidase.^39a In the assay, the enzymatic activity was quantified by measuring the release of p-nitrophenol from the enzymatic hydrolysis of p-nitrophenyl-α-O-D-glucopyranoside, used as a synthetic substrate.^39a α-Cyclodextrin was added to the reaction medium to enhance the assay sensitivity^39b and acarbose was employed as the reference inhibitor.

After normalization of the results obtained at different concentrations of the test compounds (initially evaluated with eight concentration points), it was observed (Table 3) that except for compounds 13 and 14 (entries 1 and 3), the tetracycles showed higher inhibitory activity than acarbose. Therefore, for a more accurate assessment, the experiment was repeated using 11 concentration points.

Table 3 Inhibition of α-glucosidase with acarbose as the reference inhibitor^a

Entry no.	Comp. no.	Enzyme inhibition (IC50, μM)	Potency vs. acarbose (×)^b
a Assay conditions: α-glucosidase and p-nitrophenyl-α-O-D-glucopyranoside (substrate) were employed in a medium containing 0.1 M phosphate buffer (pH 7) and α-cyclodextrin. Absorbance readings were performed at 405 nm. b Referred to the IC50 of acarbose in the 11-points test (entry 9).
1	13	761.0 ± 0.16	0.7
2	13a	266.1 ± 0.13	1.9
3	14	500.7 ± 0.06	1.0
4	16	105.5 ± 0.11	4.8
5	17	152.3 ± 0.17	3.3
6	18	31.0 ± 0.09	16.3
7	24	32.5 ± 0.05	15.6
8	28	93.3 ± 0.11	5.4
9	29	25.1 ± 0.12	20.2
10	30	12.3 ± 0.05	41.2
11	Acarbose	505.2 ± 0.05	1.0

---

The results of the second assay confirmed that compound 30 (entry 10) displayed the lowest IC₅₀ value, being over 40 times more potent than acarbose and approximately 2–2.5 times more potent than compounds 18, 24 and 29 (entries 6, 7 and 9), which showed similar potencies and also displayed promising activity. A more in-depth analysis revealed that the heterocycles derived from cyclopentane-1,3-dione (18 and 30; entries 6 and 10) and those containing a free phenol in their 2-position, such as 29 and 30 (entries 9 and 10), exhibited the highest potency. It was also found that the substituent on ring A modulates the activity of the tetracycles, with the methyl ether 13 (entry 1) being less active that the related TBS ether (24, entry 7) and the free phenol (29, entry 9).

In addition, it was observed that ketones 16 and 17 (entries 4 and 5), characterized by the presence of the gem-dimethyl moieties on the D-ring, were 4–6 times more potent than their related ketone 13, which lacks these substituents (entry 1). On the other hand, the dihydropyridine derivative 13a (entry 2) was 2.7 times more active than its unsaturated congener 13 (entry 1).

To evaluate promiscuity and selectivity, the inhibitory properties of the series were also assessed against two alternative hydrolases, namely acetylcholinesterase and β-glucosidase. However, no significant inhibition was observed for either enzyme within the experimental concentration range.

Taken together, these results provided a better understanding of the biological profile of the tested compounds and their enzyme selectivity. These findings offer valuable clues for the design of new α-glucosidase inhibitors as potential anti-diabetic drugs based on these previously unknown skeletons.

Conclusions

In conclusion, we have developed short routes toward tetracyclic analogs of Ganoderma alkaloids, which were based on the reaction between cyclic enamines derived from 1,3-dicarbonyl compounds and 3-formylchromene derivatives. Exhaustive optimization of the reaction conditions leading to the tetracyclic compounds was performed and DFT calculations provided sound explanation for the regioselectivity of the enamine attack on the α,β-unsaturated system. In this way, ten compounds were prepared in moderate overall yields, with original skeletons and several aspects of structural diversity. The tetracyclic analogs proved to be inactive as β-glucosidase and acetylcholinesterase inhibitors; however, they displayed selective inhibition of α-glucosidase, being up to 41 times more potent than acarbose.

Experimental

General information

The reactions were performed under a dry argon atmosphere using oven-dried glassware. Anhydrous dioxane was obtained by refluxing the analytical grade reagent over sodium (benzophenone as the indicator) and distilling from the deep blue solution of the sodium benzophenone ketyl. Anhydrous MeOH, EtOH and ⁱPrOH were obtained by treatment of the corresponding analytical grade solvent with clean magnesium turnings and a crystal of iodine, followed by a 4 h reflux and distillation from the resulting magnesium alkoxide suspension. Anhydrous MeCN and CH₂Cl₂ were prepared by refluxing the corresponding analytical solvent over P₂O₅ for 4 h, followed by distillation. Anhydrous DMF was obtained by heating the analytical grade product over BaO for 4 h, followed by distillation under reduced pressure. The anhydrous solvents were transferred to dry Young ampoules containing molecular sieves where they were kept for use. All other reagents were used as received.

The reactions were monitored by TLC (Merck's silica gel 60 GF₂₅₄) run in hexanes/EtOAc mixtures of different polarities. The chromatographic spots were revealed by exposure to UV light (254 and 365 nm) and spraying with phosphomolybdic acid solution, followed by careful heating to improve selectivity. For the workup procedure, in general, the reaction mixture was diluted with brine, and the products were extracted three times with EtOAc. The combined organic extracts were then washed once with brine, dried over MgSO₄, and concentrated under reduced pressure. The residue was purified by flash column chromatography (Merck's silica gel 60 H, particle size <55 μm or alumina), employing hexane/EtOAc polarity gradient techniques, under positive pressure of air.

Atom numbering and naming of the compounds were performed in agreement with the IUPAC guidelines.

Equipment

The melting points (uncorrected) were measured on an Ernst Leitz Wetzlar model 350 hot-stage microscope. The FT-IR spectra were recorded on a Shimadzu Prestige 21 spectrophotometer, with the samples prepared as solid dispersions in KBr disks. The NMR spectra were recorded with a Bruker Avance 300 FT-NMR spectrometer (300.13 MHz for ¹H NMR and 75.48 MHz for ¹³C NMR) in CDCl₃, MeOH-d₄ or DMSO-d₆, as required. The chemical shifts are reported in ppm in the δ scale, and TMS was used as the internal standard; the residual solvent peaks of CDCl₃ (δ_H = 7.26 ppm, δ_C = 77.16 ppm), MeOH-d₄ (δ_H = 3.31 ppm, δ_C = 49.00 ppm) and DMSO-d₆ (δ_H = 2.54 ppm, δ_C = 40.45 ppm) were used as internal references. The assignments of signals marked with an asterisk (*) may be interchanged. In special cases, NOE and 2D NMR experiments (COSY, HSQC, TOCSY, and HMBC) were also employed. The HRMS data were obtained with a Bruker MicroTOF-Q II instrument (Bruker Daltonics, Billerica, MA) from UMyMFOR (Buenos Aires, Argentina). Detection of the ions was performed by electrospray ionization in the positive ion mode.

The single crystal X-ray diffraction data were collected using a Bruker D8 QUEST ECO diffractometer (Bruker AXS GmbH, Karlsruhe, Germany) equipped with a PHOTON II CPAD area detector with MoK radiation (λ = 0.71073 Å) at room temperature. For the measurement at 80 K, an open-flow Oxford Cryosystems nitrogen gas chiller (Oxford Cryosystems Ltd, Long Hanborough, UK) was employed. The data collection strategy and data reduction followed standard procedures implemented in Bruker's APEX4 software. Structures were solved using SHELXS-97^40a and refined using the full-matrix least squares procedure with SHELXL2018/3.^40b All H atoms were initially found in a difference map; those that bonded to C were further idealized [C–H range = 0.93–0.99 Å] and refined using a conduction model. Anisotropic displacement parameters were employed for non-hydrogen atoms, and H atoms were treated isotropically with U_iso(H) = 1.5U_eq. (X) (methyl H) or 1.2U_eq. (X) (aromatic and other H) times the equivalent displacement parameter (U_eq.) of the parent atoms. Multi-scan absorption correction was performed employing SADABS 2016/2.^40c WinGX^40d and PLATON^40e were used to prepare the material for publication. The selection of suitable single crystals for measurement was carried out with a Leica stereoscopic magnifier model S9D (Leica, Wetzlar, Germany) with 1–5× objectives. The crystal structure data of 13 (Tables S1–S9) were deposited as cif files at the Cambridge Crystallographic Data Centre (CCDC).⁴¹

Computational methods

Conformational searches for the reactants, TS and products were run using the conformational search module of Hyperchem v.7.52 with the MM+ method. Selected structures were then successively optimized at the M062X/6-311G** level with toluene as the solvent.

Frequency calculations were made to confirm the nature of the stationary points and to evaluate their thermochemical properties. Transition states were confirmed by the presence of unique imaginary frequencies and by intrinsic reaction coordinate (IRC) calculations. The molecular orbitals of the reactants and natural bonding orbitals of the transition state were calculated to analyze the frontier orbital interactions at the M062X/6-311G** level of theory.

Biological activity

The assays were carried out in 96-well, flat-bottomed, clear acrylic microplates. Once initiated by addition of the substrate, the microplates were shaken for 2 seconds and, while being maintained at 37 °C, the increase in absorbance at 405 nm was recorded every 30 seconds for 10–15 minutes using a VERSAMax microplate reader (Molecular Devices, Sunnyvale, USA). Wells containing DMSO, without an inhibitor, were used as a negative assay control and served as a reference for maximum enzyme activity.

Each reaction was performed in triplicate, and the mean values with their corresponding standard deviations were used for data analysis. For IC₅₀ determination, 8 or 11 serial dilutions of each test compound were prepared as equally spaced points on a natural logarithmic (Neperian) scale. The IC₅₀ values were obtained by non-linear regression fitting of the log [inhibitor] versus normalized response data.

Enzymes. Electric eel acetylcholinesterase (EC 3.1.1.7), almond β-glucosidase (EC 3.2.1.21), and yeast α-glucosidase (EC 3.2.1.20) were purchased from Sigma-Aldrich (St Louis, USA).

Reagents. p-Nitrophenyl-β-O-D-glucopyranoside, p-nitrophenyl-α-O-D-glucopyranoside, α-cyclodextrin, Ellman's reagent [5,5-dithio-bis-(2-nitrobenzoic acid)], acetylthiocholine iodide, eserine, and 1-deoxynojirimycin were purchased from Sigma-Aldrich (St Louis, USA). Acarbose (produced by Cayman Chemical, Cambridge, UK) was purchased from Migliore Laclaustra SRL (Buenos Aires, Argentina).

α-Glucosidase inhibition assay^39a. Each well contained the test compound dissolved in DMSO (10 μL), yeast α-glucosidase (60 μL, 0.4 U mL⁻¹) in phosphate buffer (0.1 M, pH 7), and α-cyclodextrin (110 μL, 3 mM) in phosphate buffer (0.1 M, pH 7). After 10 min of incubation at 37 °C, the reaction was initiated by adding the substrate, p-nitrophenyl-α-O-D-glucopyranoside (90 μL, 4.88 mM) in phosphate buffer (0.1 M, pH 7). A DMSO solution of acarbose, subjected to the same conditions, was used as the reference inhibitor.

β-Glucosidase inhibition assay⁴². Each well contained the test compound dissolved in DMSO (10 μL), almond β-glucosidase (60 μL, 0.4 U mL⁻¹) in phosphate buffer (0.1 M, pH 7), and α-cyclodextrin (110 μL, 3 mM) in phosphate buffer (0.1 M, pH 7). After 10 min of incubation at 37 °C, a solution of the substrate, p-nitrophenyl-β-O-D-glucopyranoside (90 μL, 4.88 mM) in phosphate buffer (0.1 M, pH 7), was added. A DMSO solution of N_α,N_τ-di-4-fluoro arylbenzenesulfonyl histamine⁴² was also used as the positive control.

Acetylcholinesterase inhibition assay⁴². Each well contained the test compound dissolved in DMSO (10 μL), electric eel acetylcholinesterase (218 μL, 0.0042 U mL⁻¹) in phosphate buffer (0.1 M, pH 7.5), and Ellman's reagent (22 μL, 3.8 mM) in phosphate buffer (0.1 M, pH 7.5). After 10 min of incubation at 37 °C, the reaction was initiated by adding the substrate, acetylthiocholine iodide (20 μL, 6.21 mM) in phosphate buffer (0.1 M, pH 7.5). Eserine, dissolved in DMSO, was also used as the positive control.

Synthesis and characterization

2-Hydroxy-5-methoxybenzaldehyde (9). A mixture of 4-methoxyphenol (10, 125 mg, 1 mmol) and a freshly prepared NaOH solution (320 mg NaOH/400 µL H₂O) was heated to 70 °C under stirring and treated dropwise with CHCl₃ (955.2 µL, 8 mmol), added every 15 minutes. Stirring at 70 °C was continued for another 20 min, after which the mixture was diluted with hot water (400–500 µL) and neutralized with a 5 M H₂SO₄ solution (150–200 µL). The solvent was evaporated under reduced pressure, saturated NaCl solution (5 mL) was added, and the mixture was extracted with EtOAc (3 × 15 mL). The combined organic extracts were dried over MgSO₄ and concentrated under reduced pressure. Chromatography of the residue afforded 9 (94 mg, 62%) as a yellow oil. IR (film, ν): 2940, 2837, 1663, 1485, 1271, 1152, 1036, 928, 795 and 773 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 10.64 (1H, s, OH), 9.85 (1H, s, CHO), 7.14 (1H, dd, J = 9.0 and 3.1, H-4), 6.99 (1H, d, J = 3.1, H-6), 6.93 (1H, d, J = 9.0, H-3) and 3.81 (3H, s, OMe). ¹³C NMR (75 MHz, CDCl₃): δ = 196.3 (CHO), 156.2 (C-2), 152.9 (C-5), 125.4 (C-4), 120.2 (C-1), 118.9 (C-6), 115.3 (C-3) and 56.1 (OMe).

6-Hydroxy-2H-chromene-3-carbaldehyde (11)¹⁸. Under a nitrogen atmosphere, K₂CO₃ (525 mg, 3.8 mmol) was added to a stirred solution of benzaldehyde 8 (350 mg, 2.53 mmol) in 1,4-dioxane (2 mL). A solution of acrolein (7, 506 µL, 7.6 mmol) in anhydrous dioxane (2 mL) was added dropwise and the system was then heated at 100 °C for 24 h. The reaction products were extracted with EtOAc (3 × 15 mL), and the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue gave chromene 11 (338 mg, 88%) as a yellow solid; mp: 104–106 °C (sublimates). IR (film, ν): 3420, 2860, 1753, 1668, 1490, 1337, 1213, 908, 824 and 725 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.59 (1H, s, J = 8.6, CHO), 7.19 (1H, s, H-4), 6.79 (2H, d, J = 2.4, H-7 and H-8), 6.71 (1H, d, J = 2.3, H-5), 4.97 (2H, d, J = 1.1, H-2) and 4.67 (1H, s, OH). ¹³C NMR (75 MHz, CDCl₃): δ = 190.0 (CHO), 150.3 (C-8a), 150.3 (C-6), 141.2 (C-4), 132.8 (C-3), 121.4 (C-4a), 120.3 (C-7), 117.5 (C-8), 115.1 (C-5) and 63.3 (C-2).

6-Methoxy-2H-chromene-3-carbaldehyde (12). A mixture of chromene 11 (160 mg, 0.9 mmol) and K₂CO₃ (251 mg, 182 mmol) was dissolved in DMF (0.5 mL). MeI (62 µL, 0.99 mmol) was then added at rt under stirring and the reaction was allowed to continue for 20 h at rt. The reaction products were recovered by extraction with EtOAc (3 × 15 mL), and the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue afforded 12 (164.7 mg, 96%) as an orange solid; mp: 54–56 °C (lit.:^18c 52–53 °C). IR (film, ν): 2920, 2849, 1668, 1634, 1574, 1489, 1217, 1163, 1040 and 706 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.58 (1H, s, CHO), 7.21 (1H, s, H-4), 6.87 (1H, dd, J = 8.9 and 2.8, H-7), 6.81 (1H, d, J = 8.8, H-5), 6.74 (1H, d, J = 2.8, H-8), 4.97 (2H, s, H-2) and 3.78 (3H, s, OMe). ¹³C NMR (75 MHz, CDCl₃): δ = 189.89 (CHO), 154.6 (C-6), 150.3 (C-8a), 141.4 (C-4), 132.6 (C-3), 121.2 (C-4a), 119.4 (C-8), 117.4 (C-5), 113.3 (C-7), 63.3 (C-2) and 56.0 (OMe).

6-(Methoxymethoxy)-2H-chromene-3-carbaldehyde (23). A mixture of chromene 11 (300 mg, 1.7 mmol), DMAP (10.3 mg, 0.85 mmol), and DIPEA (410 mg, 5.1 mmol) was dissolved in CH₂Cl₂ (2 mL) and placed in an ice bath. The stirred solution was treated with a solution of MOMCl in PhMe (1.68 M, 3 mL, 5.5 mmol), and the system was kept stirring for 18 h at a temperature below 25 °C. The reaction mixture was diluted with brine (5 mL), the products were extracted with EtOAc (3 × 15 mL), and the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue furnished 23 (300 mg, 80%) as a yellowish oil. IR (film, ν): 3440, 2920, 2851, 1670, 1574, 1487, 1335, 1219, 1153 and 1007 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.58 (1H, s, CHO), 7.21 (1H, s, H-4), 6.99 (1H, dd, J = 8.8 and 2.9, H-7), 6.93 (1H, d, J = 2.9, H-5), 6.81 (1H, d, J = 8.8, H-8), 5.12 (2H, s, H-2), 4.98 (2H, d, J = 1.1, OCH₂O) and 3.49 (3H, s, OMe). ¹³C NMR (75 MHz, CDCl₃): δ = 189.9 (CHO), 152.2 (C-6), 151.2 (C-8a), 141.3 (C-4), 132.6 (C-3), 122.0 (C-7), 121.3 (C-4a), 117.5 (C-5), 116.4 (C-8), 95.3 (OCH₂O), 63.4 (C-2) and 56.1 (OMe). HRMS (ESI-TOF): m/z calcd for C₁₂H₁₂O₄ [(M − H)⁻]: 219.0661; m/z found: 219.0652.

6-((tert-Butyldimethylsilyl)oxy)-2H-chromene-3-carbaldehyde (21). A stirred solution of chromene 11 (50 mg, 0.284 mmol) and imidazole (23.2 mg, 0.34 mmol) in CH₂Cl₂ (1 mL) was treated with TBDMSCl (51.3 mg, 0.34 mmol). The reaction was further stirred at rt for 18 h, after which it was diluted with brine (5 mL) and the reaction products were extracted with EtOAc (3 × 15 mL). The combined organic extracts were dried (MgSO₄), concentrated under reduced pressure, and the residue was chromatographed to give 21 (81.6 mg, 99%) as a yellow solid; mp: 39–41 °C. IR (film, ν): 2930, 2857, 1674, 1570, 1485, 1333, 1271, 1111, 1026, 974, 872 and 781 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.58 (1H, s, CHO), 7.18 (1H, s, H-4), 6.82–6.73 (2H, m, H-8, H-7), 6.69 (1H, d, J = 2.4, H-5), 4.97 (2H, d, J = 1.1, H-2), 0.98 (9H, s, SiCMe₃) and 0.18 (6H, s, SiMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 189.9 (CHO), 150.6 (C-8a), 150.3 (C-3), 141.5 (C-4), 132.4 (C-6), 125.0 (C-7), 121.3 (C-4a), 119.8 (C-5), 117.3 (C-8), 63.3 (C-2), 25.8 (SiCMe₃), 18.3 (SiCMe₃) and −4.3 (2C, SiMe₂). HRMS (ESI-TOF): m/z calcd for C₁₆H₂₂O₃Si [(M + H)⁺]: 291.1411; m/z found: 291.1416.

3-Formyl-2H-chromen-6-yl acetate (20). A stirred solution of phenol 11 (100 mg, 0.57 mmol) and DMAP (58 mg, 0.52 mmol) in CH₂Cl₂ (4 mL) was successively treated with Et₃N (119 µL, 0.85 mmol) and Ac₂O (80 µL, 0.85 mmol), and the system was heated at 80 °C for 1 h. The reaction mixture was diluted with brine (5 mL) and the products were extracted with EtOAc (3 × 15 mL). The combined organic extracts were dried (MgSO₄) and concentrated under reduced pressure. Chromatography of the residue yielded 20 (96.4 mg, 93%) as a yellow solid; mp: 63–65 °C. IR (film, ν): 2860, 1753, 1668, 1489, 1337, 1213, 1144, 1011, 909, 824 and 725 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.58 (1H, s, CHO), 7.19 (1H, s, H-4), 7.00 (1H, dd, J = 8.6 and 2.7, H-7), 6.97 (1H, d, J = 2.5, H-5), 6.86 (1H, d, J = 8.6, H-8), 5.02 (2H, d, J = 1.1, H-2) and 2.29 (3H, s, OCOMe). ¹³C NMR (75 MHz, CDCl₃): δ = 189.8 (CHO), 169.7 (OCOMe), 153.7 (C-8a), 145.0 (C-6), 140.4 (C-3), 132.6 (C-4), 126.1 (C-7), 121.8 (C-5), 121.1 (C-4a), 117.4 (C-8), 63.5 (C-2) and 21.1 (OCOMe). HRMS (ESI-TOF): m/z calcd for C₁₂H₁₀O₄ [(M + H)⁺]: 219.0661; m/z found: 219.0652.

6-(Benzyloxy)-2H-chromene-3-carbaldehyde (22). A stirred mixture of chromene 11 (40 mg, 0.23 mmol) and K₂CO₃ (63 mg, 0.45 mmol) in DMF (0.75 mL) at 0 °C was treated with benzyl bromide (27 µL, 0.23 mmol) and further stirred at rt for 18 h. The reaction was diluted with brine (5 mL) and the products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue afforded 22 (51.6 mg, 85%) as an orange solid; mp: 72–74 °C. IR (film, ν): 3445, 2959, 1640, 1495, 1454, 1348, 1269, 1215, 1180 and 739 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 9.58 (1H, s, CHO), 7.45–7.30 (5H, m, H–Ar′), 7.19 (1H, s, H-4), 6.95 (1H, dd, J = 8.9 and 3.0, H-7), 6.84–6.80 (2H, m, H-5, H-8), 5.04 (2H, s, OCH₂Ar′) and 4.98 (2H, d, J = 1.2, H-2). ¹³C NMR (75 MHz, CDCl₃): δ = 189.9 (CHO), 153.7 (C-6), 150.5 (C-8a), 141.4 (C-4), 136.9 (C-1′), 132.6 (C-3), 128.8 (C-3′, C-5′), 128.2 (C-4′), 127.6 (C-2′, C-6′), 121.2 (C-4a), 120.4 (C-7), 117.5 (C-5), 114.6 (C-8), 70.9 (OCH₂Ar′) and 63.3 (C-2).

General procedure for the synthesis of enaminones

A stirred mixture of the given cyclic 1,3-dione (0.21 mmol) and NH₄OAc (20 mg, 0.26 mmol) in MeCN (1 mL) was treated with Si(OEt)₄ (57.3 µL, 0.26 mmol), and the mixture was heated under reflux for 20 h affording a yellow oil. Product formation was confirmed by TLC, and the products were identified by spectroscopy of the unstable crude products, which were used without further purification.

3-Aminocyclohex-2-en-1-one (3). IR (film, ν): 3420, 1651, 1539, 1416, 1267, 1194, 1146, 1018, 837 and 669 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.90 (1H, s, NH), 5.29 (1H, s, H-2), 5.23 (1H, s, NH), 2.35 (2H, t, J = 6.3, H-4), 2.28 (2H, t, J = 6.5, H-6) and 2.01–1.90 (2H, m, H-5). ¹³C NMR (75 MHz, CDCl₃): δ = 198.5 (C-1), 167.4 (C-3), 99.9 (C-2), 35.5 (C-6), 28.9 (C-4) and 21.8 (C-5).

3-Amino-5,5-dimethylcyclohex-2-en-1-one (3a). ¹H NMR (300 MHz, CDCl₃): δ = 9.82 (1H, s, NH), 5.40 (1H, s, NH), 5.31 (1H, s, H-2), 2.20 (2H, s, H-6), 2.15 (2H, s, H-4) and 1.05 (6H, s, CMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 198.1 (C-1), 166.4 (C-3), 98.3 (C-2), 49.3 (C-4), 42.7 (C-6), 33.0 (C-5) and 28.4 (2C, CMe₂).

3-Amino-6,6-dimethylcyclohex-2-en-1-one (3b). ¹H NMR (300 MHz, CDCl₃): δ = 6.97 (1H, s, NH₂), 5.14 (1H, s, H-2), 5.00 (1H, s, NH₂), 2.37 (2H, t, J = 6.3, H-6), 1.78 (2H, t, J = 6.4, H-5) and 1.09 (6H, s, CMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 203.4 (C-1), 164.7 (C-3), 98.7 (C-2), 39.2 (C-4), 35.7 (C-5), 26.0 (C-6) and 25.1 (2C, CMe₂).

6-Methoxy-2H-chromene (30). Isolated in a small amount via Cu(II)-promoted cyclization of 12. ¹H NMR (300 MHz, CDCl₃): δ = 8.72 (1H, d, J = 8.8, H-8), 6.65 (1H, dd, J = 3.0, 8.8, H-7), 6.54 (1H, d, J = 3.0, H-5), 6.39 (dt, 1H, J = 1.8, 9.9, H-4), 5.82 (1H, dt, J = 3.7, 9.9, H-3), 4.75 (2H, dd, J = 3.7, 3.7, H-2) and 3.75 (3H, s, OMe). ¹³C NMR (75 MHz, CDCl₃): δ = 154.1 (C-6), 148.0 (C-8a), 124.7 (C-4), 123.0 (C-3), 117.4 (C-3a), 116.2 (C-8), 114.1 (C-7), 111.8 (C-5), 65.4 (C-2) and 55.7 (OMe).

2-Methoxy-6,9,10,11-tetrahydro-8H-chromeno[4,3-b]quinolin-8-one (13). A stirred solution of chromene 12 (30 mg, 0.14 mmol) in DMF (0.5 mL) at rt was treated with Zn(OTf)₂ (68 mg, 0.19 mmol), followed by the addition of a solution of enamine 3 (40 mg, 0.36 mmol) in DMF (0.5 mL) via a cannula. The reaction was heated at 100 °C for 48 h, after which it was diluted with brine (5 mL) and the products were extracted with EtOAc (3 × 15 mL). The combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue afforded tetracycle 13 (17 mg, 44%) as a yellow solid; mp: 135–137 °C. IR (film, ν): 2941, 2830, 1680, 1593, 1493, 1456, 1275, 1217, 1043, 818, 797 and 698 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.00 (1H, s, H-7), 7.77 (1H, d, J = 2.8, H-1), 6.95 (1H, dd, J = 8.9 and 2.9, H-3), 6.90 (1H, d, J = 8.8, H-4), 5.17 (2H, s, H-6), 3.88 (3H, s, OMe), 3.18 (2H, t, J = 6.2, H-11), 2.76–2.62 (2H, m, H-9) and 2.27–2.13 (2H, m, H-10). ¹³C NMR (75 MHz, CDCl₃): δ = 197.7 (C-8), 163.9 (C-11a), 155.2 (C-2), 152.2 (C-4a), 151.7 (12a), 130.9 (C-7), 127.0 (C-6a), 125.1 (C-12b), 122.9 (C-7a), 120.2 (C-3), 118.4 (C-1), 108.5 (C-4), 67.8 (C-6), 56.0 (OMe), 38.6 (C-9), 32.8 (C-11) and 22.1 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₇H₁₅NO₃ [(M + H)⁺]: 282.1125; m/z found: 282.1126.

2-(Methoxymethoxy)-6,9,10,11-tetrahydro-8H-chromeno[4,3-b] quinolin-8-one (25). A stirred mixture of 1,3-cyclohexanedione (5, 31 mg, 0.27 mmol) and NH₄OAc (21 mg, 0.27 mmol) in PhMe (1 mL) was treated with Si(OEt)₄ (60 µL, 0.27 mmol) and the system was heated at 50 °C for 1 h. After cooling to rt, chromene 23 (30 mg, 0.14 mmol) was added, and the resulting system was heated at 110 °C for 48 h. The reaction was diluted with brine (5 mL) and the products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue on neutral alumina afforded tetracycle 25 (12 mg, 27%) as a yellow solid; mp: 105–107 °C. IR (film, ν): 3979, 2953, 1687, 1593, 1490, 1458, 1348, 1269, 1220, 1153, 993, 920, 823 and 736 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.00 (1H, s, H-7), 7.92 (1H, d, J = 3.0, H-1), 7.06 (1H, dd, J = 8.9 and 3.0, H-3), 6.91 (1H, d, J = 8.9, H-4), 5.21 (2H, s, OCH₂O), 5.17 (2H, s, H-6), 3.51 (3H, s, OMe), 3.18 (2H, t, J = 6.2, H-11), 2.73–2.65 (2H, m, H-9) and 2.27–2.13 (2H, m, H-10). ¹³C NMR (75 MHz, CDCl₃): δ = 197.7 (C-8), 163.9 (C-11a), 152.6 (C-4a), 152.5 (C-2), 152.0 (C-12a), 130.9 (C-7), 127.1 (C-7a), 125.0 (C-6a), 123.1 (C-12b), 121.8 (C-3), 118.3 (C-4), 112.4 (C-1), 95.1 (OCH₂O), 67.8 (C-6), 56.2 (OMe), 38.6 (C-9), 32.8 (C-11) and 22.0 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₈H₁₇NO₄ [(M + H)⁺]: 312.1230; m/z found: 312.1221.

2-((tert-Butyldimethylsilyl)oxy)-6,9,10,11-tetrahydro-8H-chromeno[4,3-b]quinolin-8-one (24). A stirred mixture of NH₄OAc (16 mg, 0.2 mmol), 1,3-diketone 5 (17.4 mg, 0.15 mmol), and Si(OEt)₄ (46.1 µL, 0.2 mmol) in PhMe (0.5 mL) was heated at 55 °C for 2 h. A solution of chromene 21 (30 mg, 0.1 mmol) in PhMe (0.5 mL) was then added, and the system was stirred at 60 °C for 18 h. Brine (5 mL) was added and the reaction products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue with neutral alumina afforded tetracycle 24 (11 mg, 27%) as a yellow solid; mp 98–100 °C. IR (film, ν): 2928, 2857, 1684, 1593, 1491, 1267, 1217, 1186, 918 and 841 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.00 (1H, s, H-7), 7.74–7.70 (1H, m, H-1), 6.85 (2H, d, J = 1.8, H-3 and H-4), 5.17 (2H, s, H-6), 3.19 (2H, t, J = 6.2, H-11), 2.73–2.66 (2H, m, H-9), 2.26–2.15 (2H, m, H-10), 1.01 (9H, s, SiCMe₃) and 0.23 (6H, s, SiMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 197.8 (C-8), 163.9 (C-11a), 152.3 (C-12a), 152.0 (C-2), 150.9 (C-2), 130.9 (C-7), 127.0 (C-7a), 125.0 (C-6a), 124.7 (C-3*), 123.2 (C-12b), 118.1 (C-4*), 116.1 (C-1), 67.8 (C-6), 38.7 (C-9), 32.8 (C-11), 25.9 (SiCMe₃), 22.1 (C-10), 18.4 (SiCMe₃) and −4.26 (2C, SiMe₂). HRMS (ESI-TOF): m/z calcd for C₂₂H₂₇NO₃Si [(M + H)⁺]: 382.1833; m/z found: 382.1839.

2-Methoxy-10,10-dimethyl-6,9,10,11-tetrahydro-8H-chromeno [4,3-b]quinolin-8-one (16). A stirred mixture of 5,5-dimethyl-1,3-cyclohexanedione (5a, 30 mg, 0.21 mmol), NH₄OAc (20 mg, 0.26 mmol), and Si(OEt)₄ (57.3 µL, 0.26 mmol) in MeCN (1 mL) was heated at 82 °C for 18 h to form enamine 3a, which was then used without purification for the next step. In a microwave tube, a solution of the previously formed enamine 3a in MeCN (1 mL) was added to a mixture of chromene 12 (20 mg, 0.11 mmol) and FeCl₃ (17.1 mg, 0.1 mmol) in MeCN (1 mL). The system was subjected to microwave irradiation (80 °C) for 2.5 h, after which the reaction was diluted with brine (5 mL) and the reaction products were extracted with EtOAc (3 × 15 mL). The combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue afforded tetracycle 16 (7 mg, 20%) as a yellow solid; mp: 114–116 °C. IR (film, ν): 2955, 2868, 2833, 1680, 1593, 1493, 1435, 1219, 1040 and 816 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.00 (1H, s, H-7), 7.78 (1H, d, J = 2.7, H-1), 7.02–6.88 (2H, m, H-3, H-4), 5.19 (2H, s, H-6), 3.89 (3H, s, OMe), 3.09 (2H, s, H-9), 2.56 (2H, s, H-11) and 1.14 (6H, s, CMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 197.8 (C-8), 162.5 (C-11a), 155.2 (C-2), 152.7 (C-12a), 151.6 (C-4a), 130.5 (C-7), 126.1 (C-7a), 125.0 (C-6a), 123.0 (C-12b), 120.3 (C-3), 118.4 (C-4), 108.5 (C-1), 67.8 (C-6), 56.1 (OMe), 52.2 (C-10), 46.6 (C-11), 33.2 (C-9), and 28.5 (CMe₂). HRMS (ESI-TOF): m/z calcd for C₁₉H₁₉NO₃ [(M + H)⁺]: 310.1438; m/z found: 310.1451.

2-Methoxy-9,9-dimethyl-6,9,10,11-tetrahydro-8H-chromene[4,3-b]quinolin-8-one (17). A solution of 4,4-dimethyl-1,3-cyclohexanedione (5b, 30 mg, 0.21 mmol) in MeCN (1 mL) was treated successively with NH₄OAc (20 mg, 0.26 mmol) and Si(OEt)₄ (57.3 µL, 0.26 mmol). The resulting solution was stirred and heated at 82 °C to form the corresponding enamine 3b (after TLC), which was then used without purification. MeCN (1 mL) and FeCl₃ (17.1 mg, 0.1 mmol) were added to a microwave tube containing chromene 12 (20 mg, 0.11 mmol). A solution of the previously formed enamine 3b in MeCN (1 mL) was then transferred to the tube, and the system was heated under microwave radiation at 80 °C for 3 h. The reaction products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue furnished the pyridine tetracycle 17 (7 mg, 20%) as a yellow solid; mp: 86–88 °C. IR (film, ν): 1645, 1493, 1383, 1273, 1204, 1130, 1044, 802, 756 and 667 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 8.03 (1H, s, H-7), 7.78 (1H, d, J = 2.6, H-1), 6.98–6.88 (2H, m, H-3 and H-4), 5.18 (2H, s, H-6), 3.89 (3H, s, OMe), 3.25–3.15 (2H, m, H-9), 2.11–1.98 (2H, m, H-10) and 1.24 (6H, s, CMe₂). ¹³C NMR (75 MHz, CDCl₃): δ = 202.2 (C-8), 162.7 (C-11a), 155.2 (C-2), 152.1 (C-12a), 151.7 (C-4a), 131.8 (C-7), 125.8 (C-7a), 125.1 (C-6a), 122.9 (C-12b), 120.2 (C-3), 118.4 (C-4), 108.5 (C-1), 67.9 (C-6), 56.1 (OMe), 41.6 (C-9), 35.5 (C-10), 29.1 (C-11), 24.3 (2C, CMe₂). HRMS (ESI-TOF): m/z calcd for C₁₉H₁₉NO₃ [(M + H)⁺]: 310.1438; m/z found: 310.1450.

2-Methoxy-9,10-dihydrochromeno[4,3-b]cyclopenta[e]pyridin-8 (6H)-one (18). A stirred mixture of 1,3-cyclopentanedione (5c, 350 mg, 3.6 mmol), NH₄OAc (550 mg, 7.14 mmol) and Si(OEt)₄ (1593.3 µL, 7.14 mmol) in EtOH (8 mL) was refluxed for 3 h to form the corresponding enamine of interest (3c), which was used without purification for the next step, after removing the solvent. A mixture of chromene 12 (300 mg, 1.5 mmol) and Zn(OTf)₂ (267 mg, 0.733 mmol) was dissolved in DMF (4 mL) and treated under magnetic stirring with a freshly prepared solution of enamine 3c in DMF (4 mL). The reaction was heated at 107 °C for 3 h, after which it was cooled to rt and quenched by addition of brine (5 mL). The organic products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue afforded tetracycle 18 (58 mg, 15%) as an orange solid; mp: 112–114 °C. IR (film, ν): 2924, 2848, 1709, 1595, 1497, 1433, 1271, 1207, 1093 and 819 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 7.82 (1H, d, J = 2.7, H-1), 7.75 (1H, s, H-7), 6.99–6.87 (2H, m, H-3 and H-4), 5.20 (2H, s, H-6), 3.90 (3H, s, OMe), 3.35–3.26 (2H, m, H-10) and 2.85–2.75 (2H, m, H-9). ¹³C NMR (75 MHz, CDCl₃): δ = 204.5 (C-8), 174.8 (C-10a), 155.3 (C-2, C-11a), 151.9 (C-4a), 129.2 (C-6a), 127.8 (C-7), 125.9 (C-7a), 123.1 (C-11b), 120.9 (C-4), 118.5 (C-3), 108.6 (C-1), 68.3 (C-6), 56.1 (OMe), 36.1 (C-9) and 28.9 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₆H₁₃NO₃ [(M + H)⁺]: 268.0968; m/z found: 268.0973.

2-(Methoxymethoxy)-9,10-dihydrochromeno[4,3-b]cyclopenta[e]pyridin-8(6H)-one (26). A stirred solution of 1,3-cyclopentanedione (5c, 280 mg, 2.9 mmol), NH₄OAc (440 mg, 5.71 mmol) and Si(OEt)₄ (1189.2 µL, 5.71 mmol) in EtOH (6.5 mL) was heated at 60 °C for 18 h to form the enamine of interest (3c). The reaction solvent was removed by distillation under reduced pressure and replaced with PhMe (4 mL). Zn(OTf)₂ (230 mg, 0.633 mmol) was added to the resulting solution under stirring, followed by a solution of chromene 12 (300 mg, 1.4 mmol) in PhMe (6 mL), and the system was heated at 107 °C for 20 h. The reaction products were extracted with EtOAc (3 × 15 mL), and the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue gave 26 (58 mg, 15%) as an orange solid; mp: 91–93 °C. IR (film, ν): 2924, 2852, 1709, 1595, 1494, 1454, 1215, 1182, 1152, 1076 and 1015 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 7.96 (1H, d, J = 3.0, H-1), 7.72 (1H, s, H-7), 7.09 (1H, dd, J = 8.9 and 3.0, H-3), 6.93 (1H, d, J = 8.9, H-8), 5.22 (2H, s, H-6), 5.19 (2H, s, OCH₂O), 3.52 (3H, s, OMe), 3.31–3.25 (2H, m, H-10) and 2.82–2.76 (2H, m, H-9). ¹³C NMR (75 MHz, CDCl₃): δ = 204.5 (C-8), 174.8 (C-10a), 154.9 (C-4a), 152.7 (C-2), 152.6 (C-11a), 129.2 (C-6a), 127.8 (C-7), 125.7 (C-7a), 123.2 (C-11b), 122.4 (C-3), 118.4 (C-4), 112.6 (C-1), 95.1 (C-6), 68.2 (OCH₂O), 56.2 (OMe), 36.1 (C-9) and 28.9 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₇H₁₅NO₄ [(M + H)⁺]: 298.1074; m/z found: 298.1083.

2-Methoxy-8,9,10,11-tetrahydro-6H-chromeno[4,3-b]quinolin-8-ol (14). A stirred solution of tetracycle 13 (20 mg, 0.07 mmol) in a mixture of MeOH and CH₂Cl₂ (1 : 10, v/v; 1.1 mL) was treated with NaBH₄ (15.4 mg, 0.28 mmol) and stirred again for 18 h at rt. The reaction was diluted with brine (5 mL) and the products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried (MgSO₄) and then concentrated under reduced pressure. Chromatography of the residue provided 14 (15 mg, 75%) as an orange solid; mp: 78–80 °C. IR (film, ν): 3417, 2938, 2851, 1647, 1564, 1493, 1273, 1204, 1040, 939, 870, 800 and 737 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 7.75–7.72 (1H, m, H-1), 7.50 (1H, s, H-7), 6.90–6.86 (2H, m, H-3 and H-4), 5.12 (2H, s, H-6), 4.81 (1H, t, J = 5.5, H-8), 3.87 (3H, s, OMe), 3.08–2.87 (2H, m, H-11), 2.15–2.02 (2H, m, H-9, H-10) and 1.94–1.80 (2H, m, H-9, H-10). ¹³C NMR (75 MHz, CDCl₃): δ = 157.0 (C-11a), 155.1 (C-2), 150.7 (C-4a), 147.6 (C-12a), 133.3 (C-7a), 132.3 (C-7), 124.6 (C-6a), 123.8 (C-12b), 118.4 (C-3*), 118.0 (C-4*), 108.0 (C-1), 68.2 (C-8), 68.0 (C-6), 56.1 (OMe), 32.6 (C-11), 32.5 (C-9) and 19.0 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₇H₁₇NO₃ [(M + H)⁺]: 284.1281; m/z found: 284.1287.

2-Hydroxy-6,9,10,11-tetrahydro-8H-chromeno[4,3-b]quinolin-8-one (29). A stirred solution of tetracycle 25 (30 mg, 0.1 mmol) in ⁱPrOH (2 mL) was treated with Amberlyst-15 (60 mg), and the system was heated under reflux for 18 h. The reaction mixture was filtered to separate the resin, which was washed with EtOAc (5 mL), and the combined organic phases were concentrated under reduced pressure. Chromatographic purification of the residue (hexane–EtOAc mixtures containing 1% Et₃N) gave phenol 29 (20 mg, 88%) as a brown solid; mp: >200 °C. IR (film, ν): 3437, 2922, 2851, 2357, 1645, 1555, 1452, 1416, 1310, 1184, 1024, 802, 784 and 667 cm⁻¹. ¹H NMR (300 MHz, DMSO-d₆): δ = 9.38 (1H, s, OH), 8.03 (1H, s, H-7), 7.56 (1H, dd, J = 2.3 and 0.9, H-1), 6.82 (2H, d, J = 2.3, H-4 and H-3), 5.19 (2H, s, H-6), 3.08 (2H, t, J = 6.0, H-11), 2.68–2.60 (2H, m, H-9) and 2.16–2.05 (2H, m, H-10). ¹³C NMR (75 MHz, DMSO-d₆): δ = 197.5 (C-8), 164.0 (C-11a), 153.0 (C-4a), 151.7 (C-12a), 150.4 (C-2), 131.3 (C-7), 127.1 (C-7a), 125.7 (C-6a), 123.0 (C-12b), 120.6 (C-3), 118.5 (C-4), 110.9 (C-1), 67.3 (C-6), 38.5 (C-9), 32.6 (C-11) and 21.9 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₆H₁₃NO₃ [(M + H)⁺]: 268.0968; m/z found: 268.0967.

8,9,10,11-Tetrahydro-6H-chromeno[4,3-b]quinoline-2,8-diol (28). A mixture of tetracycle 27 (20 mg, 0.06 mmol) and Amberlyst 15 (48.6 mg) in ⁱPrOH (1 mL) was heated at reflux with stirring for 18 h. The reaction mixture was filtered to remove the resin, which was then washed with CH₂Cl₂ (5 mL). The combined organic phases were concentrated under reduced pressure and the residue was chromatographed (CH₂Cl₂ : EtOH; 1% gradient), affording 28 (11 mg, 63%) as a brown solid; mp: 114–116 °C. IR (film, ν): 3418, 2928, 1645, 1537, 1470, 1452, 1337, 1200, 1065, 1014, 943 and 665 cm⁻¹. ¹H NMR (300 MHz, MeOH-d₄): δ = 7.89 (1H, s, OH-2), 7.60 (1H, s, H-7), 7.53 (1H, d, J = 2.3, H-1), 6.82–6.72 (2H, m, H-3, H-4), 5.08 (2H, s, H-6), 4.74 (1H, s, H-8), 4.58 (1H, s, OH-8), 2.98–2.86 (2H, m, H-11), 2.05 (2H, dd, J = 13.8 and 8.4, H-9 and H-10) and 1.84 (2H, dd, J = 8.4 and 4.2, H-9 and H-10). ¹³C NMR (75 MHz, MeOH-d₄): δ = 157.1 (C-11a), 152.6 (C-4a), 150.5 (C-2), 147.7 (C-12a), 134.4 (C-7a), 133.2 (C-7), 125.7 (C-6a), 124.0 (C-12b), 118.4 (C-3), 117.9 (C-4), 110.4 (C-1), 67.9 (C-6), 67.5 (C-8), 32.2 (C-11*), 32.1 (C-9*) and 19.0 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₆H₁₅NO₃ [(M + H)⁺]: 270.1125; m/z found: 270.1138.

2-Hydroxy-9,10-dihydrochromeno[4,3-b]cyclopenta[e]pyridin-8 (6H)-one (30). A stirred mixture of tetracycle 26 (50 mg, 0.17 mmol) and Amberlyst-15 (100 mg) in ⁱPrOH (4 mL) was heated under reflux for 20 h. The mixture was filtered to remove the resin, which was washed with EtOAc (5 mL), and the combined organic phases were concentrated under reduced pressure. Chromatography of the residue yielded 30 (20 mg, 47%) as an orange solid; mp: 120–122 °C. IR (film, ν): 3416, 2924, 1694, 1645, 1591, 1487, 1413, 1190, 1140, 1024, 821 and 806 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 7.84 (1H, d, J = 2.8, H-1), 7.74 (1H, s, H-7), 6.94 (2H, dd, J = 8.4 and 3.0, H-3 and H-4), 5.19 (2H, s, H-6), 5.18 (1H, s, OH), 3.35–3.24 (2H, m, H-10) and 2.85–2.76 (2H, m, H-9). ¹³C NMR (75 MHz, CDCl₃): δ = 204.5 (C-8), 174.8 (C-10a), 155.2 (C-11a), 153.5 (C-4a), 151.8 (C-2), 131.0 (C-6a), 127.8 (C-7), 125.8 (C-7a), 123.1 (C-11b), 122.8 (C-3), 118.4 (C-4), 111.8 (C-1), 68.3 (C-6), 36.1 (C-9) and 28.9 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₅H₁₁NO₃ [(M + H)⁺]: 254.0812; m/z found: 254.0817.

2-Methoxy-6,9,10,11,12,12a-hexahydro-8H-chromeno[4,3-b]quinolin-8-one (13a). A stirred solution of 1,3-cyclohexanedione (5, 24 mg, 0.21 mmol) and NH₄OAc (16.2 mg, 0.21 mmol) in ⁱPrOH (1 mL) at rt was treated with InCl₃ (2.3 mg, 0.0105 mmol). After 1 h, the system was cooled to 0 °C in an ice bath, chromene 12 (20 mg, 0.105 mmol) was added, and the system was left for 20 h at 5 °C. Brine (5 mL) was added to the reaction mixture, and the organic solvent was removed under reduced pressure. The reaction products were extracted with EtOAc (3 × 15 mL); the combined organic extracts were dried over MgSO₄ and then concentrated under reduced pressure. Chromatography of the residue gave 13a (25 mg, 84%) as a yellow solid; mp: 127–129 °C. IR (film, ν): 3447, 2953, 2849, 1678, 1593, 1493, 1350, 1302, 1217, 1042, 818 and 698 cm⁻¹. ¹H NMR (300 MHz, CDCl₃): δ = 6.71 (1H, d, J = 8.7, H-4), 6.62 (1H, dd, J = 8.7 and 2.9, H-3), 6.55 (1H, d, J = 2.8, H-1), 6.04 (1H, s, H-7), 4.85 (1H, s, H-12a), 4.40 (2H, s, H-6), 3.75 (3H, s, OMe), 2.59 (2H, dt, J = 17.9 and 4.2, H-9 and H-11), 2.47–2.30 (2H, m, H-9 and H-11), 2.09–1.87 (2H, m, H-10) and 1.25 (1H, s, NH). ¹³C NMR (75 MHz, CDCl₃): δ = 192.4 (C-8), 191.0 (C-11a), 154.4 (C-2), 146.9 (C-4a), 131.1 (C-6a), 123.8 (C-12b), 119.8 (C-7), 115.9 (C-4), 114.6 (C-7a), 113.9 (C-3), 111.5 (C-1), 67.1 (C-6), 55.9 (OMe), 33.4 (C-9*), 33.0 (C-11*), 32.8 (C-12a) and 20.2 (C-10). HRMS (ESI-TOF): m/z calcd for C₁₇H₁₇NO₃ [(M − H)⁻]: 282.1125; m/z found: 282.1117.

Author contributions

All authors have agreed to their individual contributions. L. Javier Cala Gomez: formal analysis, validation, visualization, writing (original draft; review & editing); Natalia L. Calvo: data curation, formal analysis, investigation, validation, visualization, writing (original draft; review & editing); Mario O. Salazar: data curation, formal analysis, investigation, validation, visualization, writing (original draft; review & editing); Sebastián O. Simonetti: data curation, formal analysis, investigation, validation, visualization, writing (original draft; review & editing); R. L. E. Furlan: data curation, formal analysis, investigation, validation, visualization, writing (original draft; review & editing); A. B. J. Bracca: conceptualization, data curation, formal analysis, funding acquisition, investigation, project administration, supervision, validation, visualization, writing (original draft; review & editing); and T. S. Kaufman: conceptualization, data curation, formal analysis, funding acquisition, investigation, project administration, supervision, validation, visualization, writing (original draft; review & editing).

Conflicts of interest

There are no conflicts to declare.

Data availability

All data supporting this article have been included in the supplementary information (SI). Supplementary information: spectral compound characterization data [¹H NMR, ¹³C NMR, 2D spectra (COSY, HSQC, HMBC)], details of DFT calculations and single-crystal XRD data. See DOI: https://doi.org/10.1039/d5ob01787d.

CCDC 2501681 and 2501682 contain the supplementary crystallographic data for this paper.^43a,b

Acknowledgements

The authors gratefully acknowledge Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET, PIP 0765) and Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT, PICT 2021-0162 and PICT 2019-3969) for financial support. L. J. C. G. also thanks CONICET for his doctoral fellowship. The authors are grateful to the FAIRE programme provided by the Cambridge Crystallographic Data Centre (CCDC) for the opportunity to use the Cambridge Structural Database (CSD) and associated software. The results of the mechanistic calculations have been obtained by using the facilities of the CCT-Rosario Computational Center, member of the High Performance Computing National System (SNCAD, MincyT-Argentina).

References

1. (a) J. Flannick, S. Johansson and P. R. Njølstad, Nat. Rev. Endocrinol., 2016, 12, 394–406; (b) International Diabetes Federation, IDF Diabetes Atlas, ed. IDF, 10th edn, 2021. https://www.diabetesatlas.org/ (accessed November 1, 2025); (c) M. Telagari and K. Hullatti, Indian J. Pharmacol., 2015, 47, 425–429; (d) S. Kumar, S. Narwal, V. Kumar and O. Prakash, Pharmacogn. Rev., 2011, 5, 19–29; (e) A. Chaudhury, C. Duvoor, V. S. R. Dendi, S. Kraleti, A. Chada, R. Ravilla, A. Marco, N. S. Shekhawat, M. T. Montales and K. Kuriakose, Front. Endocrinol., 2017, 8, 1–12.
2. (a) B. Boh, M. Berovic, J. Zhang and L. Zhi-Bin, Biotechnol. Annu. Rev., 2007, 13, 265–301; (b) S. Wachtel-Galor, J. Yuen, J. A. Buswell and I. F. F. Benzie, Ganoderma lucidum (Lingzhi or Reishi). A Medicinal Mushroom, in Herbal Medicine: Biomolecular and Clinical Aspects, 2nd edn, ed. I. F. F. Benzie and S. Wachtel-Galor, CRC Press, Boca Raton USA, 2011, ch. 9; (c) R. Yangzom and P. Wangchuk, Ganoderma lucidum (Lingzhi Mushroom): Its Medicinal Uses, Biomolecules and Therapeutic Applications, in Phytochemistry and Nutritional Composition of Significant Wild Medicinal and Edible Mushrooms. Traditional Uses and Pharmacology, ed. A. Sharma, G. Bhardwaj and G. A. Nayik, RSC, London, UK, 2023; (d) Therapeutic Mushrooms for Diabetes Mellitus. Current Evidences and Future Scope, ed. U. Azeem and K. R. Hakeem, Apple Academic Press, Inc., Palm Bay, USA, 2023.
3. (a) W. L. Li, H. C. Zheng, J. Bukuru and N. De Kimpe, J. Ethnopharmacol., 2004, 92, 1–21; (b) B. S. Teng, C. D. Wang, H. J. Yang, J. S. Wu, D. Zhang, M. Zheng, Z. H. Fan, D. Pan and P. Zhou, J. Agric. Food Chem., 2011, 59, 6492–6500; (c) K. Winska, W. Maczka, K. Gabryelska and M. Grabarczyk, Molecules, 2019, 24, 4075.
4. (a) C. Xiao, Q. Wu, J. Zhang, Y. Xie, W. Cai and J. Tan, J. Ethnopharmacol., 2017, 196, 47–57; (b) H. Hikino, C. Konno, Y. Mirin and T. Hayashi, Planta Med., 1985, 339–340; (c) M. Tomoda, R. Gonda, Y. Kasahara and H. Hikino, Phytochemistry, 1986, 25, 2817–2820; (d) H. Hikino, M. Ishiyama, Y. Suzuki and C. Konno, Planta Med., 1989, 55, 423–428.
5. (a) M. C. A. Galappaththi, N. M. Patabendige, B. M. Premarathne, K. K. Hapuarachchi, S. Tibpromma, D.-Q. Dai, N. Suwannarach, S. Rapior and S. C. Karunarathna, Biomolecules, 2023, 13, 24; (b) Q. Xia, H. Zhang, X. Sun, H. Zhao, L. Wu, D. Zhu, G. Yang, Y. Shao, X. Zhang, X. Mao, L. Zhang and G. She, Molecules, 2014, 19, 17478–17535.
6. (a) J. Zhang, K. Ma, H. Chen, K. Wang, W. Xiong, L. Bao and H. Liu, J. Antibiot., 2017, 70, 915–917; (b) K. Wang, L. Bao, K. Ma, J. Zhang, B. Chen, J. Han, J. Ren, H. Luo and H. Liu, Eur. J. Med. Chem., 2017, 127, 1035–1046; (c) H. T. Ma, J. F. Hsieh and S. T. Chen, Phytochemistry, 2015, 114, 109–113.
7. (a) T. Kubota, Y. Asaka, I. Miura and H. Mori, Helv. Chim. Acta, 1982, 65, 611–619; (b) S. Fatmawati, R. Kondo and K. Shimizu, Bioorg. Med. Chem. Lett., 2013, 23, 5900–5903; (c) S. Fatmawati, K. Shimizu and R. Kondo, Planta Med., 2010, 76, 1691–1693; (d) S. Fatmawati, K. Shimizu and R. Kondo, Fitoterapia, 2010, 81, 1033–1036.
8. (a) N. P. D. Nguyen, P. T. Quy, D. C. To, T. Q. Bui, N. V. Phu, T. T. A. My, P. H. Nguyen, N. H. Kien, N. T. T. Hai and N. T. A. Nhung, Trop. J. Nat. Prod. Res., 2023, 7, 3421–3432; (b) L. Yang, D.-X. Kong, N. Xiao, Q.-Y. Ma, Q.-Y. Xie, J.-C. Guo, C. Y. Deng, H.-X. Ma, Y. Hua, H.-F. Dai and Y.-X. Zhao, Bioorg. Chem., 2022, 127, 106025; (c) K. Wang, L. Bao, W. Xiong, K. Ma, J. Han, W. Wang, W. Yin and H. Liu, J. Nat. Prod., 2015, 78, 1977–1989.
9. (a) C. W. Phan, V. Sabaratnam and U. R. Kuppusamy, Antidiabetic Effects of Ganoderma Prospects and Challenges, in Ganoderma: Cultivation, Chemistry and Medicinal Applications, ed. K. Acharya and S. Khatua, CRC Press, Boca Raton, USA, 2024, vol, 1; (b) X.-R. Peng, S. B. Unsicker, J. Gershenzon and M.-H. Qiu, Nat. Prod. Rep., 2023, 40, 1354–1392; (c) S. Fatmawati, K. Shimizu and R. Kondo, Phytomedicine, 2011, 18, 1053–1055.
10. (a) H. Rasouli, R. Yarani, F. Pociot and J. Popović-Djordjević, Pharmacol. Res., 2020, 155, 104723; (b) S. Sukhikh, O. Babich, A. Prosekov, O. Kalashnikova, S. Noskova, A. Bakhtiyarova, O. Krol, E. Tsvetkova and S. Ivanova, Metabolites, 2023, 13, 513; (c) S. B. Gaikwad, G. K. Mohan and M. S. Rani, Pharm. Crops, 2014, 5(Supp. 1: M2), 11–28.
11. J. J. Yang and D. Q. Yu, Acta Pharm. Sin., 1990, 25, 555–559.
12. L. Tian, X.-L. Wang, Y.-Z. Wang and Y.-X. Cheng, Nat. Prod. Res. Dev., 2015, 27, 1325–1328.
13. X.-L. Wang, M. Dou, Q. Luo, L.-Z. Cheng, Y.-M. Yan, R.-T. Li and Y.-X. Cheng, Fitoterapia, 2017, 116, 93–98.
14. Y. Chen and P. Lan, ACS Omega, 2018, 3, 3471–3481.
15. S. Z. Huang, B. H. Cheng, Q. Y. Ma, Q. Wang, F. D. Kong, H. F. Dai, S. Q. Qiu, P. Y. Zheng, Z. Q. Liu and Y.-X. Zhao, RSC Adv., 2016, 6, 21139–21147.
16. (a) I. Cortés, E. Cordisco, T. S. Kaufman, M. A. Sortino, L. A. Svetaz and A. B. J. Bracca, RSC Adv., 2021, 11, 19587–19597; (b) L. Fernández, A. L. Reviglio, D. A. Heredia, G. M. Morales, M. Santo, L. Otero, F. Alustiza, A. C. Liaudat, P. Bosch, E. L. Larghi, A. B. J. Bracca and T. S. Kaufman, Heliyon, 2021, 7, e06436; (c) I. Cortés, C. M. Borini Etichetti, J. E. Girardini, A. B. J. Bracca and T. S. Kaufman, Synthesis, 2019, 433–440; (d) J. L. Pergomet, M. G. Di Liberto, M. G. Derita, A. B. J. Bracca and T. S. Kaufman, Fitoterapia, 2018, 125, 98–105; (e) E. L. Larghi, M. A. Operto, R. Torres and T. S. Kaufman, Eur. J. Med. Chem., 2012, 55, 74–84.
17. (a) D. F. Vargas, E. L. Larghi and T. S. Kaufman, Nat. Prod. Rep., 2019, 39, 354–401; (b) D. A. Heredia, E. L. Larghi and T. S. Kaufman, Eur. J. Org. Chem., 2016, 1397–1404; (c) C. C. Silveira, E. L. Larghi, S. R. Mendes, A. B. J. Bracca, F. Rinaldi and T. S. Kaufman, Eur. J. Org. Chem., 2009, 4637–4645.
18. (a) Y. Satoh, J. L. Stanton, A. J. Hutchison, A. H. Libby, T. J. Kowalski, W. H. Lee, D. H. White and E. F. Kimble, J. Med. Chem., 1993, 36, 3580–3594; (b) Y. Zhou, Y. Zhu, S. Yan and Y. Gong, Angew. Chem., Int. Ed., 2013, 52, 10265–10269; (c) S. Y. Shin, H. Jung, S. Ahn, D. Hwang, H. Yoon, J. Hyun, Y. Yong, H. J. Cho, D. Koh, Y. H. Lee and Y. Lim, Bioorg. Med. Chem., 2014, 22, 1809–1820.
19. (a) H. Adkins and W. H. Hartung, Org. Synth. Coll., 1941, 1, 15; (b) W. L. F. Armarego and D. D. Perrin, Purification of Laboratory Chemicals, 4th edn, Butterworth-Heinemann, Oxford, UK, 2017.
20. T. G. F. Marsh, A. Santambrogio, B. J. Murray, L. T. Boulton, J. A. Aguilar, D. S. Yufit, G. Sandford and W. D. G. Brittain, Eur. J. Org. Chem., 2023, e202300058.
21. Y. Satoh, J. L. Stanton, A. J. Hutchison, A. H. Libby, T. J. Kowalski, W. H. Lee, D. H. White and E. F. Kimble, J. Med. Chem., 1993, 36, 3580–3594.
22. S. L. Xu, C. P. Li and J. H. Li, Synlett, 2009, 0818–0822.
23. (a) S. Hirai, Y. Horikawa, H. Asahara and N. Nishiwaki, Chem. Commun., 2017, 53, 2390–2393; (b) M. Arita, S. Yokoyama, H. Asahara and N. Nishiwaki, Synthesis, 2019, 2007–2013.
24. Y. Zhao, J. Zhao, Y. Zhou, Z. Lei, L. Li and H. Zhang, New J. Chem., 2005, 29, 769–772.
25. S. Hirai, Y. Horikawa, H. Asahara and N. Nishiwaki, Chem. Commun., 2017, 53, 2390–2393.
26. (a) S. Rana, H. S. Hong, L. Barrigan, L. W. Jin and D. H. Hua, Bioorg. Med. Chem. Lett., 2009, 19, 670–674; (b) F. Liéby-Muller, C. Allais, T. Constantieux and J. Rodriguez, Chem. Commun., 2008, 4207–4209.
27. G. Bartoli, K. Babiuch, M. Bosco, A. Carlone, P. Galzerano, P. Melchiorre and L. Sambri, Synlett, 2007, 2897–2901.
28. (a) V. P. A. Raja, G. Tenti, S. Perumal and J. C. A. Menéndez, Chem. Commun., 2014, 50, 12270–12272; (b) M. Venkati and G. L. D. Krupadanam, Synth. Commun., 2001, 31, 2589–2598.
29. (a) J. Lombard, S. Romain, S. Dumas, J. Chauvin, M. N. Collomb, D. Daveloose, A. Deronzier and J. C. Leprêtre, Eur. J. Inorg. Chem., 2005, 3320–3330; (b) G. Serdaroğlu, N. Uludag and E. Üstün, J. Mol. Liq., 2023, 375, 121364.
30. (a) F. Zou, F. Pei, L. Wang, Z. Ren, X. Cheng, Y. Sun, J. Wu and P. He, Synlett, 2019, 717–720; (b) A. Kundu, B. Bhattacharyya, K. Dhara, S. Paul, I. Majumdere and R. Kundue, New J. Chem., 2020, 44, 4898–4906; (c) X. Chang, X. Yang, Z. Chen and W. Zhong, Synlett, 2019, 1431–1436.
31. (a) S. Kantevari, D. Addla and B. Sridhar, Synthesis, 2010, 3745–3754; (b) P. N. Ngubane, T. Papo, S. Zamisa and B. Omondi, J. Mol. Struct., 2025, 1322, 140567.
32. (a) D. Sang, C. Yi, C. He, J. Wang, J. Tian, M. Yao and H. Shi, Tetrahedron Lett., 2018, 59, 1469–1472; (b) J. Tian, H. Yue, P. Yang and D. Sang, ChemistrySelect, 2019, 4, 38–41; (c) T. S. Kaufman, J. Chem. Soc., Perkin Trans. 1, 1996, 2497–2505.
33. (a) Y. Ishihara, S. Azuma, T. Choshi, K. Kohno, K. Ono, H. Tsutsumi, T. Ishizu and S. Hibino, Tetrahedron, 2011, 67, 1320–1333; (b) R. S. Porto, M. L. A. A. Vasconcellos, E. Ventura and F. Coelho, Synthesis, 2005, 2297–2306; (c) M. Süsse, S. Johne and S. Hesse, Helv. Chim. Acta, 1992, 75, 457–470; (d) Y. Satoh, J. L. Stanton, A. J. Hutchison, A. H. Libby, T. J. Kowalski, W. H. Lee, D. H. White and E. F. Kimble, J. Med. Chem., 1993, 36, 3580–3594.
34. (a) T. S. Kaufman, R. P. Srivastava, R. D. Sindelar, S. M. Scesney and H. C. Marsh Jr., J. Med. Chem., 1996, 38, 1437–1445; (b) T. S. Kaufman, R. P. Srivastava, R. D. Sindelar, S. M. Scesney and H. C. Marsh Jr., Bioorg. Med. Chem. Lett., 1995, 5, 501–506.
35. (a) M. C. Bröhmer, N. Volz and S. Bräse, Synlett, 2009, 1383–1386; (b) A. Modak, A. Deb, T. Patra, S. Rana, S. Maity and D. Maiti, Chem. Commun., 2012, 48, 4253–4255; (c) Z. Selakovic, A. M. Nikolic, V. Ajdacic and I. M. Opsenica, Eur. J. Org. Chem., 2022, e202101265; (d) B. Gutmann, P. Elsner, T. Glasnov, D. M. Roberge and C. O. Kappe, Angew. Chem., Int. Ed., 2014, 53, 11557–11561.
36. (a) M. C. Bröhmer, N. Volz and S. Bräse, Synlett, 2009, 1383–1386; (b) G. Hall, A. Samat, R. Guglielmetti, L. Van Parys, W. Saeyens, D. De Keukeleir, K. Lorenz and A. Mannscbreck, Helv. Chim. Acta, 1997, 80, 1122–1132.
37. (a) J. Xia and G.-W. Wang, Synthesis, 2005, 2379–2383; (b) K. V. Maslov, T. I. Akimova and V. A. Kaminski, Chem. Heterocycl. Compd., 2002, 38, 417–421; (c) W. Shang, S. Li, Y. Ding, Z. Pan, X. Tong and C. Xia, Org. Chem. Front., 2017, 4, 1789–1793; (d) S. J. Garden, C. R. Werneck Guimaraes, M. B. Corréa, C. A. Fernandes de Oliveira, A. da Cunha Pinto and R. Bicca de Alencastro, J. Org. Chem., 2003, 68, 8815–8822; (e) S. Bahmanyar and K. N. Houk, J. Am. Chem. Soc., 2001, 123, 11273–11283.
38. Y. Cao, J. S. M. Perry, E. Zhang, A. Trinh, A. Kacker, S. Cruz, H. Ceballos, A. Pan, W. Huang and K. G. M. Kou, JACS Au, 2025, 5, 1429–1438.
39. (a) M. I. Osella, M. O. Salazar, M. D. Gamarra, D. M. Moreno, F. Lambertucci, D. E. Frances and R. L. E. Furlan, RSC Med. Chem., 2020, 11, 518–527; (b) M. R. Eftink and J. C. Harrison, Bioorg. Chem., 1981, 10, 388–398.
40. (a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112–122; (b) G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3–8; (c) L. Krause, R. Herbst-Irmer, G. M. Sheldrick and D. Stalk, J. Appl. Crystallogr., 2015, 48, 3–10; (d) L. J. Farrugia, J. Appl. Crystallogr., 2012, 45, 849–854; (e) A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2009, 65, 148–155.
41. C. R. Groom, I. J. Bruno, M. P. Lightfoot and S. C. Ward, The Cambridge Structural Database, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 2016, 72, 171–179.
42. M. O. Salazar, M. I. Osella, I. A. Ramallo and R. L. E. Furlan, RSC Adv., 2018, 8, 36209–36218.
43. (a) CCDC 2501681: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2pz6d9; (b) CCDC 2501682: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2pz6fb.

---