Synthesis and biological evaluation of 10-benzyloxy-Narciclasine

Abstract

A chemoenzymatic convergent synthesis of 10-benzyloxy narciclasine from bromobenzene was accomplished in 16 steps. The key transformations included toluene dioxygenase-mediated hydroxylation, nitroso Diels-Alder reaction and intramolecular Heck cyclization. The unnatural derivative of narciclasine was subjected to biological evaluation and its activity was compared to other C-10 and C-7 compounds prepared previously.

1. Introduction

Narciclasine (1), discovered by Ceriotti [1] in 1968, belongs to the Amaryllidaceae alkaloid family, whose other major constituents are shown in Fig. 1.

Fig. 1.

Structures of some Amaryllidaceae constituents.

In the NCI 60 human cancer cell line screen, narciclasine and pancratistatin registered mean GI50 values of 0.016 μM and 0.091 μM, making them among the most potent small molecule Amaryllidaceae anticancer constituents [2], and it was discovered that narciclasine is capable of inducing apoptosis through the activation of death receptors FAS and DR4 [3]. It was also demonstrated that narciclasine inhibits protein synthesis at the ribosomal 60S unit [4], by forming hydrogen bonds with the rRNA of the peptidyl transferase center (PTC) in the A-site (Fig. 2) [5]. In this manner, the molecule prevents the entrance of the aminoacyl-tRNA, thus impeding the synthesis of proteins.

Fig. 2.

Binding of narciclasine to the 60S ribosomal subunit.

The capacity of the Amaryllidaceae alkaloids to selectively target cancer cells has triggered the interest of the synthetic community, directed mainly toward the synthesis of narciclasine (1) [6] and pancratistatin (2) [7]. Activity in total synthesis and biological evaluation has been extensively reviewed [8]. The earliest total syntheses of narciclasine were accomplished in Rigby by 1997 [6a] and in 1999 by Keck [6b] and Hudlicky [6c]. Design and synthesis of natural and unnatural derivatives of some of the isocarbostyril alkaloids, in particular pancratistatin [9], led to the identification of the so called “bay area” of pancratistatin and narciclasine analogs, where structural modifications can be made without loss of activity (Fig. 3).

Fig. 3.

Suggested minimum pharmacophore of narciclasine (1) and pancratistatin (2) [10].

While any changes to the areas marked with the solid red line (Fig. 3) significantly affect, and usually diminish, the biological activities, the northwestern space of the pharmacophore, highlighted with the blue dotted line and comprising the C-1 and C-10 positions (Fig. 3), allows modifications with no detriment to activity. This feature has been exploited in the investigations of unnatural derivatives of pancratistatin, mostly at C-1 but also at C-7 and the alcohols at C-3 and C-4. Most of the reports of unnatural Amaryllidaceae derivatives are related to modifications of pancratistatin, as described in our 2016 review [8f]. In some cases these modifications lead to compounds with augmented biological activity (viz C-1 benzoate and C-1 benzoyl methyl derivatives of pancratistatin [8f]). We have made and tested several derivatives of narciclasine: C-7 aza-compound [11], its N-oxide (inactive) and C-10 aza-compound 7 (Fig. 4) that was active but slightly less potent than narciclasine [12].

Fig. 4.

Synthesized C-10 derivatives of narciclasine.

In order to explore the biological activity of new derivatives in the C-10 space, we designed the de novo total synthesis of 10-benzyloxy-narciclasine (8), which is reported in this paper. In fact, analyzing the mode of interaction of narciclasine with its target on the A-site of the ribosomal peptidyl transferase [4], it was proposed that the introduction of a lipophilic substituent at the C10 position could increase the affinity to the binding site by making a π-stacking interaction with the rRNA. The linear synthetic pathway makes use of an inexpensive, highly oxygenated aromatic starting material to construct ring A by sequential functionalization, later coupled with the precursor to ring C, obtained by chemoenzymatic methods. The in vitro biological evaluation of the novel compound on a tumor cell line is also reported, and its activity will be compared to the activity of other natural and unnatural congeners.

2. Results and discussion

2.1. Retrosynthetic analysis

The retrosynthetic approach to 10-benzyloxy-narciclasine is outlined in Scheme 1.

Scheme 1.

Retrosynthetic analysis.

Access to the target compound 8 is envisaged through the intramolecular Heck-type cyclization of intermediate 9, obtained after N–O bond reductive cleavage of oxazine 10. The oxazine intermediate contains the fully functionalized ring A and the ring C masked in a bicyclic system, which are predicted to be joined through a nitroso-Diels-Alder reaction between hydroxamic acid 11 and diene diol 12. The A-ring coupling partner derives from the sequential functionalization of commercially available 2,3,4-trihydroxy benzaldehyde 13, while compound 12 is accessed by enzymatic oxidative dearomatization of bromobenzene 14. This chemoenzymatic vicinal dihydroxylation of aromatic substrates [13], allows for the large-scale production of valuable stereoselective synthons.

2.2. Synthesis of ring A

Scheme 2 summarizes the synthesis of aromatic ring A of the target compound. The first challenge was the selective methylation of the C-4-phenolic hydroxyl in 2,3,4-trihydroxy benzaldehyde 13. To perform this transformation, a 2,3-borate ester was formed in situ by mixing the starting material and borax in water, applying a protocol developed by Pettit in the synthesis of combretastatins A-1 and B-1 [14]. The reaction was monitored by observing a significant increase of the solubility of the organic compound that dissolved in the aqueous media resulting in a yellow solution. Addition of dimethyl sulfate afforded the desired monomethylated product 15 after several hours and an acidic work-up, accompanied by negligible amounts of the 2-methoxy regioisomer. The next step required the alkylation of the catechol in 15 to allow the installation of the methylene dioxy bridge; however, this reaction was plagued by poor yields never exceeding 35–40% after purification. As the only other product found in the reaction mixture was the starting material, its recycling through the same reaction conditions allowed the access to the required compound in suitable amounts. After alkylation of the phenolic groups already present in the molecule, Dakin oxidation [15] of the aldehyde moiety in 16 proceeded in good yields and led to phenol 17, which has an important role in controlling the regioselectivity of the next steps. Selective bromination at the position ortho to the phenol was accomplished by treatment with 0.5 eq. of N,N′-dibromo-2,2-dimethylhydantoin (DBDMH), added portion-wise [16]. The reaction proceeded to completion and, owing to its high selectivity, did not require chromatographic purification. With Bn-protected bromide 19 in hand, the last step in the full functionalization of the phenyl ring entailed the introduction of a carbonyl at the remaining aromatic position, by means of a Rieche formylation. A subsequent Pinnick oxidation furnished carboxylic acid 20, in turn converted into acyl chloride 21. Compound 21, upon treatment with hydroxylamine hydrochloride under basic conditions, yielded hydroxamic acid 11 in moderate-to-good yields. Hydroxamic acid 11 represents the building block for the aromatic half of the target compound. The preparation of ring C, its coupling to ring A and the synthetic endgame are reported in the next sections.

Scheme 2.

Synthesis of ring A.

2.3. Synthesis of ring C and Diels-Alder reaction

To access compound 12, the oxidative dearomatization of bromobenzene 14 (Scheme 3) was carried out by means of the whole-cell fermentation of the substrate with the recombinant strain of E. Coli JM109 (pDTG601) [17], which over-expresses toluene dioxygenase and is capable of dihydroxylating a wide array of aromatic starting materials [18]. This method, discovered and developed by Gibson [19], provides highly useful synthons containing a vicinal cis-diol in large quantities and in a stereoselective fashion.

Scheme 3.

Nitroso-Diels-Alder reaction.

A one-pot protocol for the diol protection and coupling with the A-ring precursor was developed, as shown in Scheme 3. Treatment of compound 12 with 2,2-dimethoxy propane (DMP) in the presence of toluene sulphonic acid afforded acetonide 22. Hydroxamic acid 11 was oxidized in situ with Bu₄NIO₄ to its nitrosyl derivative, and a one-pot nitroso-Diels-Alder reaction with compound 22 furnished oxazine 10 in good yields.

In addition to its regioselectivity, this transformation proceeds with high stereoselectivity [20] as the bulky acetonide group directs the cycloaddition to the α-face of the molecule.

2.4. Oxazine reduction and the completion of the synthesis

Compound 10, a crucial intermediate in the retrosynthetic scheme above, contains the A and C rings coupled and concealed in a tricyclic system. It was necessary to reduce the oxazine portion of the molecule to reveal the substituted conduramine portion of the target compound, which would then lead to the final intramolecular closure (Scheme 4).

Scheme 4.

Completion of the synthesis.

In the next step the N–O bond reduction was attempted with excess aluminum–mercury amalgam. When the reaction was conducted following literature conditions [21], using 30 equiv. of finely cut aluminum foil treated with 1% aq. KOH, then 0.5% aq. HgCl₂ (Table 1, entry 1), the only product that was isolated was the debrominated amide 24, instead of the desired intermediate 23. It was clear that the excess of the reducing agent led to overreduction of the product and its dehalogenation. When the reaction was repeated with 10 equiv of aluminum foil (Table 1, entry 2), a low yield of the desired intermediate 23 accompanied the formation of the over-reduced by-product. In the last iteration (Table 1, entry 3), the very thin and reactive foil was replaced by the same quantity (10 equiv.) of thicker and less reactive turnings, which were still treated as before. This modified procedure led to the desired compound 23 obtained as the only product and in an average 50% yield, accompanied by some starting material, which was recycled through the same reaction conditions. The debrominated compound was produced in trace amounts or not at all.

Table 1.

Optimization of the oxazine reduction.

Entry	Aluminum source	Equivalents	Yield of 23	Yield of 24
1	Foil	30	traces	87%
2	Foil	10	12%	61%
3	Turnings	10	54%	traces

Oxazine reduction

TBS-protection of the free C2 alcohol of compound 23 yielded compound 9, which, after protection of the amide as a Boccarbamate, was subjected to the key Heck intramolecular cyclization to establish the molecular backbone. This type of abnormal Heck reaction, which requires an anti-hydride elimination instead of the conventional syn-hydride elimination to regain the double bond, has already been used in many syntheses of Amaryllidaceae alkaloids and derivatives, although with inconsistent results [22]. When the transformation was conducted with the key substrate 9, the desired product was obtained in yields never exceeding 15%. The reaction was plagued by problems of low reproducibility, with isolation of unreacted starting material or the debrominated cyclization precursor on various occasions. The testing of different catalysts such as Pd(PPh₃)₄ or Pd₂(dba)₃, and the screening of other inorganic bases already used in the same kind of transformation [11] (Ag₃PO₄, Cs₂CO₃) did not bring any improvement to the reaction. In the same manner, when the ring closure was attempted by using various organic bases such as Et₃N, tetramethylethylenediamine (TMEDA), 1,2-diazabiciclo[2.2.2.]octane (DABCO) and 1,8-diazabiciclo[5.4.0]undec-7-ene (DBU), previously employed to perform similar intramolecular reactions [23], no significant enhancement was recorded. As a consequence, the reaction had to be repeated several times to accumulate enough material to carry forward in the synthesis and produce enough material for biological testing.

To obtain 10-benzyloxy-narciclasine 8, the protected intermediate 25 was treated with TMSCl and KI in order to demethylate the anisole moiety at the C7 position. Acidic work-up and extraction afforded, after chromatographic purification, 10-benzyloxy-narciclasine in 15 steps from 2,3,4-trihydroxy benzaldehyde.

2.5. Biological activity

The novel compound 10-benzyloxy-narciclasine 8 was tested in vitro against BE(2)-human neuroblastoma cells. A panel of natural and unnatural congeners, previously synthesized in the Hudlicky group [24,25], were also tested. The results are shown in Table 2 for comparison.

Table 2.

In vitro activity of a panel of Amaryllidaceae congeners.

Compound	1	28	29	30	31	32	8
IC50 (μM)	0.011	0.865	0.092	0.007	0.007	>500	20.753
Average ± SD	± 0.001	± 0.070	± 0.005	± 0.001	± 0.001		± 2.048

Results of biological evaluation

These preliminary results show that 10-benzyloxynarciclasine displays biological activity significantly lower than that of any other molecule tested, except for 32, where the B-ring is not closed. It is instructive to draw a comparison between the synthesized C10-analogue 8 and previously reported 31, also containing a large aromatic residue in the “bay area” of the molecule. In contrast to the poorly active new analogue 8, compound 31 shows potency that exceeds that of narciclasine itself (Table 2). It is important to note, however, that 31, as well as 30, are esters that can undergo intra-cellular hydrolysis and lose the large hydrophobic group from the “bay area”, while 8 is an ether that cannot easily hydrolyze in cell culture.

2.5.1. Conclusions

The design of the C-10 benzyloxy derivative was based on the promising results obtained with analogs containing substituents in the “bay area” of the molecule. We reasoned that the introduction of a lipophilic moiety at C-10 may increase the binding affinity of the compound to its target. It is disappointing, however, to see the weak activity in 8 while the activity of the C-1 ester or the 10-aza derivatives were in the same range as that of narciclasine. In addition, we hoped to prepare a C-10 hydroxy derivative that would be further oxidized to a quinone. The hydroquinine might retain activity while the quinone may not, as it has been demonstrated that the presence of the C-7 hydroxyl is essential for activity in both narciclasine and pancratistatin. Unfortunately, the insufficient amount of material prevented any further conversion to the two compounds.

We will next pursue the synthesis of C-1 derivatives to test the alterations in the bay region of narciclasine. Given the challenges presented by synthesis of these densely functionalized C-10 derivatives, further studies will leverage a semi-synthetic route from narciclasine extracted from daffodil bulbs. We will report on the outcome of this research in due course.

3. Experimental section

All solvents were distilled and kept dry before usage. All reactions were done in inert atmosphere (Ar or N₂) and at 25 °C, unless otherwise stated. All reagents were obtained from commercial sources. Nuclear magnetic resonance (NMR) analyses performed on Bruker Avance AV 300, Bruker Avance III HD 400 Bruker Avance AV 600 digital NMR spectrometers, running Topspin 2.1 and 3.5 software. The probes are furnished with VT (variable temperature) and gradient equipment. Chemical shifts are given in δ, coupling constants (J) in Hz. Melting points (mp) were measured using a capillary apparatus. Mass spectra (HRMS) measurements were determined using a LTQ Orbitrap XL. The molecular mass-associated ion was measured by electron ionization, electrospray ionization or fast atom bombardment. Infrared (IR) spectra were recorded on an FT-IR spectrophotometer as neat and are reported in wave numbers (cm⁻¹). Column chromatography performed by using flash grade 60 silica gel. Thin layer chromatography (TLC) was performed on silica gel 60 F254-coated aluminum sheets. TLC plates were visualized using UV and stained using iodine, cerium ammonium molybdate (CAM), KMnO₄ solutions, FeCl₃ solutions, ninhydrin solutions or 2,4-dinitrophenylhydrazine (2,4-DNP) solutions.

3.1. 2,3-Dihydroxy-4-methoxybenzaldehyde (15)

2,3,4-Trihydroxybenzaldehyde (10 g, 65 mmol) was suspended in distilled water (100 mL), then borax (97.3 mmol, 19.6 g, 1.5 equiv) was added and the mixture stirred for 10 min. To the orange solution NaOH (71 mmol, 2.8 g, 1.1 equiv) was added, and the mixture was stirred for 10 min. After dimethyl sulfate (71 mmol, 6.7 mL, 1.1 equiv) was added dropwise over 5 min, the reaction mixture was stirred for 15 h. The reaction was quenched by adding 6 M HCl until the mixture became acidic (pH ~ 2). The mixture was extracted with EtOAc (3 × 20 mL) and the combined organic fractions dried over MgSO₄. After gravity filtration to remove the drying agent, the crude material was absorbed on silica gel and purified by flash column chromatography [hexanes/EtOAc (3:1 v/v)]. The desired product is a pale -yellow solid. Yield 6.1 g (56%).

15: R_f = 0.26 [Hex/EtOAc (2:1 v/v)]; mp = 115–117 °C (Hex/EtOAc 1:1); ¹H NMR (300 MHz, CDCl₃) δ 11.10 (bs, 1H), 9.74 (s, 1H), 7.13 (d, J = 8.7 Hz, 1H), 6.61 (d, J = 8.7 Hz, 1H), 5.53 (bs, 1H), 3.97 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 128 195.2, 153.0, 149.0, 133.0, 1126.1, 116.1, 103.6, 56.3. The data of the product match those in the literature [14].

3.2. 4-Methoxy-2,3-methylenedioxybenzaldehyde (16)

The mixture of 2,3-dihydroxy-4-methoxybenzaldehyde (15, 26.3 g, 156.4 mmol) and potassium carbonate (oven-dried, 45.4 g, 329 mmol, 2.1 equiv) in DMF (105 mL) was stirred for 15 min at rt and dibromomethane (43.5 g, 17.6 mL, 251 mmol, 1.6 equiv) was added. The mixture was heated at 95 °C (bath) for 4 h, at which time TLC showed complete disappearance of starting material. The mixture was cooled in an ice bath, and cold water (total 300 mL) was slowly added with stirring to separate solid product. After the mixture had been stirred for 1 h, the solid was filtered off and washed with aqueous methanol (40%), (2 × 30 mL). The product was dried on a hot plate at 35 °C. Yield 23.4 g (45%).

16: R_f = 0.36 [Hex/EtOAc (4:1 v/v)]; mp = 82–84 °C (DCM); ¹H NMR (300 MHz, CDCl₃) δ 9.98 (s, 1H), 7.30 (d, J = 8.7 Hz, 1H), 6.62 (d, J = 8.8 Hz, 1H), 6.13 (s, 2H), 3.98 (s, 3H). The data of the product match those in the literature [26].

3.3. 4-Methoxy-2,3-methylenedioxy-phenol (17)

4-Methoxy-2,3-methylenedioxybenzaldehyde (16, 3 g, 17 mmol) was dissolved in MeOH (100 mL) and concentrated H₂SO₄ (2.5 mL). The mixture was cooled in an ice bath and 30% aq H₂O₂ v/v (12 mL, 99 mmol, 6 equiv), diluted with MeOH (50 mL) was added dropwise to the stirring solution over 30 min. The reaction then proceeded at room temperature for 8 h in an open vessel (250 mL RBF, 24/40 neck). The reaction was quenched by adding sat. aq. NaHSO₃ (20 mL), then MeOH was removed under vacuum, and 2 M KOH was added until basicity (pH ~ 12). The basic aqueous solution was extracted with DCM (2 × 15 mL) to remove any impurities. The aqueous layer was acidified with until neutrality 6 M HCl and extracted with EtOAc (3 × 20 mL), the combined organic fractions were dried over MgSO₄, filtered and concentrated under vacuum to obtain the product as an off-white solid. Yield 2.6 g (92%).

17: R_f = 0.23 [Hex/EtOAc (4:1 v/v)]; mp = 102–105 °C (Hex/EtOAc); ¹H NMR (300 MHz, CDCl₃) δ 6.43 (d, J =9.0 Hz, 1H), 6.40 (d, J = 9.1 Hz, 1H), 5.97 (s, 2H), 4.76 (bs, 1 H), 3.85 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 138.4, 136.3, 135.4, 134.2, 110.1, 107.9, 101.8, 57.1. The data of the product match those in the literature [27].

3.4. 5-Bromo-7-methoxybenzo[d][1,3]dioxol-4-ol (18)

To compound 17 (2.2 g, 13 mmol) suspended in DCM (70 mL) was added DBDMH (1.9 g, 6.5 mmol, 0.5 equiv) in six equal portions over 20 min, to give an orange mixture, which was allowed to react at room temperature for 16 h. The reaction was quenched by adding sat aq NaHSO₃ (15 mL), then the mixture was extracted with DCM (3 × 10 mL). The collected organic fractions were then dried over MgSO₄ and, after filtration and concentration, the product was obtained as an off-white solid. The crude material was used directly for the next step. Yield 3 g (92%).

18: R_f = 0.35 [Hex/EtOAc (2:1 v/v)]; mp = 124–127 °C (Hex/EtOAc); ¹H NMR (300 MHz, CDCl₃) δ 6.67 (s, 1H), 6.05 (s, 2H), 5.04 (bs, 1 H), 3.86 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 150.6, 148.5, 135.7, 123.8, 114.6, 108.0, 103.0, 56.8. The product matches the literature data [28].

3.5. 4-(Benzyloxy)-5-bromo-7-methoxybenzo[d][1,3]dioxole (19)

To compound 18 (1.5 g, 6 mmol) in dry THF (40 mL) was added NaH directly (60% in mineral oil, 1.7 g, 12 mmol, 2 equiv) to yield a pale yellow solution, which was stirred for 10 min before BnBr (1.4 mL, 12 mmol, 2 equiv) was added dropwise over 5 min. The mixture was then heated at reflux for 15 h. The reaction mixture was cooled to 0 °C and quenched by adding water (10 mL), then the mixture was extracted with EtOAc (3 × 15 mL). The collected organic fractions were dried over MgSO₄, then filtered and concentrated under vacuum. The crude material was adsorbed on silica and purified by flash column chromatography [hexanes/EtOAc (1:1 v/v)], and the product was obtained as a yellow solid. Yield 1.6 g (76%).

19: R_f = 0.66 [Hex/EtOAc (1:1 v/v)]; mp = 56–58 °C (Hex/EtOAc); ¹H NMR (300 MHz, CDCl3) δ 7.52−7.41 (m, 5 H), 6.98 (s, 1H), 6.02 (s, 2H), 5.09 (s, 2H), 3.82 (s, 3H); ¹³C NMR (75 MHz, CDCl3) δ 140.9, 140.5, 139.3, 136.0, 134.3, 130.5, 128.6, 128.0, 127.2, 123.4, 106.3, 60.5; IR (neat) 2941, 1597, 1428, 1335, 1252, 1037 cm⁻¹; HRMS (EI) calcd. for C₁₆H₁₃BrO₆: 335.9997. Found: 335.9972.

3.6. 3-Benzyloxy-2-bromo-6-methoxy-4,5-methylenedioxybenzoic acid (20)

4-(Benzyloxy)-5-bromo-7-methoxybenzo[d][1,3]dioxole (19) (1.5 g, 4.5 mmol) dissolved in anhydrous DCM (30 mL) was cooled to −78 °C, then TiCl₄ (0.53 mL, 4.9 mmol, 1.1 equiv) was added dropwise over 1 min. The resulting solution was stirred for 5 min, then dichloromethyl methyl ether (0.44 mL, 4.9 mmol, 1.1 equiv) was added dropwise, and the temperature was allowed to rise to room temperature. After 1.5 h, the reaction was quenched by adding 2 M HCl (5 mL), the mixture was then extracted with EtOAc (3 × 10 mL). The collected organic fractions were dried over MgSO₄. After filtration of the drying agent and concentration, the crude product (0.74 g, 2.0 mmol) was dissolved in t-BuOH (20 mL) to gain a clear colorless solution. Sulfamic acid (5.8 g, 60 mmol, 30 equiv) was added and stirred for 10 min, followed by a 5% aq NaH₂PO₄ w/v (10 mL). NaClO₂ (0.54 g, 6 mmol, 3 equiv) dissolved in 5% aq NaH₂PO₄ w/v (5 mL) was then added to the reaction mixture. The resulting bright yellow mixture was stirred for 12 h at room temperature protected from light. The reaction was quenched by adding a saturated aq NaHSO₃ (5 mL) and extracted with 5% aq NaHCO₃ w/v (3 × 10 mL). The aqueous phase was acidified with 2 M HCl until pH ~2 and extracted with EtOAc (4 × 15 mL). The collected organic fractions were then dried over MgSO₄, filtered and concentrated to obtain the product as a pale pink crystalline solid: 1.18 g, 69% yield over two steps.

20: mp = 103–105 °C (EtOAc).¹H NMR (300 MHz, CD₃OD) δ 7.48−7.45 (m, 2H), 7.35−7.32 (m, 3H), 6.04 (s, 2H), 5.16 (s, 2H), 3.91 (s, 3H); ¹³C NMR (75 MHz, CD₃OD) δ 169.0, 142.0, 139.6, 138.0, 137.7, 136.3, 129.5, 129.4, 129.3, 126.1, 105.8, 104.0, 75.6, 61.1. IR (neat) 3313, 3011, 2963, 2854, 1702, 1682, 1430, 802 cm⁻¹; HRMS (EI) calcd. for C₁₆H₁₃BrO₆: 379.9869. Found: 379.9852.

3.7. 3-Benzyloxy-2-bromo-6-methoxy-4,5-methylenedioxybenzoic acid chloride (21)

The reaction was conducted under anhydrous conditions. The carboxylic acid 20 (2.84 g, 0.0075 mol) was dissolved in DCM (25 mL), then oxalyl chloride (1 mL, 0.011 mol) was added dropwise over 15 min to give an orange suspension. After the addition of oxalyl chloride, 0.2 mL of anhydrous DMF were added to the mixture, which caused the suspension to turn a deep orange color with a diffuse evolution of bubbles that lasted for 5 min. After 1 h, the reaction mixture was subjected to vacuum concentration. The product was obtained as a brown solid (2.63 g, 91%).

21: R_f = 0.49 [Hex/EtOAc (2:1 v/v)]; mp = 121–124 °C (DCM); ¹H NMR (300 MHz, CDCl₃) δ 7.49−7.34 (m, 5H), 6.03 (s, 2H), 5.16 (s, 2H), 4.00 (s, 3H); ¹³C NMR (75 MHz, CD₃OD) δ 165.6, 141.7, 137.7, 136.1, 135.9, 134.9, 128.5, 28.4, 128.2, 127.3, 104.2, 102.6, 74.8, 60.5; IR (neat) 3006–3092, 2902–2966, 1788, 1621, 1482 cm⁻¹; HRMS (EI) calcd for C₁₆H₁₂BrClO₅: 397.9557. Found: 397.9545.

3.8. 7-(Benzyloxy)-6-bromo-N-hydroxy-4-methoxybenzo[d][1,3] dioxole-5-carboxamide (11)

Hydroxylamine hydrochloride (0.542 g, 7.8 mmol) and lithium carbonate (0.72 g, 9.7 mmol) were stirred in a mixture of H₂O (2 mL) and Et₂O (10 mL). After about 10 min, the temperature was brought to 0 °C, and the acyl chloride 21 (2.5 g, 6.5 mmol) was added in three equal portions over 5 min. The solution turned from colorless to brown-orange, and, as the temperature neared room temperature, diffuse bubbling was observed. After 1 h, 15 mL of Et₂O was added to dilute the thick brown semisolid, then the mixture was extracted with EtOAc (4 × 15 mL), dried over Na₂SO₄, filtered and concentrated under vacuum to obtain the product as a brown solid. Yield 1.8 g (70%).

11: R_f = 0.19 [Hex/EtOAc (2:1 v/v)]; mp = 125–127 °C (i-PrOH); ¹H NMR (300 MHz, CDCl₃) δ 8.32 (bs 1H), 7.48−7.32 (m, 5H), 7.10 (bs 1H), 6.01 (s, 2H), 5.14 (s, 2H), 3.95 (s, 3H); ¹³C NMR (150 MHz, acetone-d6) δ 206.3, 162.6, 141.7, 139.3, 138.4, 137.8, 135.9, 129.1, 107.8, 103.6, 75.1, 61.1; IR (neat) 3305, 3215, 2953–3031, 2874, 1636, 1486, 1477 cm⁻¹; HRMS (CI) calcd for C₁₆H₁₄BrNO₆: 395.0004. Found: 395.0064; Anal. Calcd for C₁₆H₁₄BrNO₆: C, 48.50; H, 3.56. Found C, 48.32; H, 3.87.

3.9. (7-(Benzyloxy)-6-bromo-4-methoxybenzo[d][1,3]dioxol-5yl)((3aS,4R,7R,7aS)-4-bromo-2,2-dimethyl-3a,4,7,7a-tetrahydro-4,7(epoxyimino)benzo[d][1,3]dioxol-8-yl)methanone(7-(benzyloxy)-6-bromo4-methoxybenzo[d][1,3]dioxol-5-yl)((3aS,4R,7R,7aS)-4-bromo-2,2dimethyl-3a,4,7,7a-tetrahydro-4,7-(epoxyimino)benzo[d][1,3]dioxol-8yl)methanone (10)

Diene diol 12 (0.154 g, 0.663 mmol) was dissolved in 2,2-DMP (10 mL) in presence of p-TsOH (0.012 g, 0.0663 mmol) and stirred at room temperature for 20 min. The pale yellow solution was then concentrated under reduced pressure, and the residual brown oil was dissolved in THF (4 mL). After the solution was washed with sat aq NaHCO₃ (1 × 2 mL), the organic phase was dried over MgSO₄ and filtered by gravity. Bu₄NIO₄ was then added to afford a yellow solution. A solution of hydroxamic acid 11 (0.126 g, 0.363 mmol) in THF (1 mL) was then added dropwise over 1 min at room temperature. The resulting solution was stirred at room temperature in air for 16 h (100 mL RBF, 24/40 neck). The completion of the reaction was confirmed by the disappearance of the starting material on TLC. The solvent was removed by evaporation, and the crude product was purified by flash column chromatography [hexanes/EtOAc (1:1 v/v)] to afford the title compound (0.197 g, 87%) as a white crystalline solid.

10: R_f = 0.32 [hexanes:EtOAc (1:1 v/v)]; mp = 68–70 °C (benzene/pentane 1:1); ${\left[\alpha \right]}_{\text{D}}^{24}=-2.1$ (c = 1.06, CHCl₃); ¹H NMR (600 MHz, acetone-d6, mixture of rotamers) δ 7.51−7.35 (m, 5H), 6.65−6.44 (m, 2H), 6.19−6.12 (m, 2H), 5.59 (s, 1H), 5.18 (s, 2H), 4.79−4.63 (m, 2H), 3.93−3.86 (m, 3H), 1.35−1.30 (m, 6H); ¹³C NMR (150 MHz, THF-d8, mixture of rotamers) δ 165.7, 141.6, 138.7, 138.1, 137.5, 136.0, 135.4, 135.0, 132.9, 132.2, 129.0, 128.8, 128.7, 125.1, 111.8, 103.6, 103.5, 89.9, 82.2, 75.2, 75.0, 60.8, 60.4, 50.5, 50.3, 25.8, 25.6, 25.5; IR (CHCl₃) 3008, 1658, 1610, 1448, 1421, 1377, 1262, 1208, 1085, 1053, 1030, 985, 950 cm⁻¹; MS (EI) m/z (%) 625 (1). 366 (15), 363 (100), 274 (8), 269 (13); HRMS (EI) calcd for C₂₅H₂₃NO₈Br₂: 624.9770, found: 624.9774; Anal. Calcd for C₂₅H₂₃NO₈Br₂: C, 48.02, H 3.71, Found C 48.23, H 3.83.

3.10. 7-(Benzyloxy)-6-bromo-N-((3aS,4R,7S,7aR)-7-hydroxy-2,2-dimethyl3a,4,7,7a-tetrahydrobenzo[d][1,3]dioxol-4-yl)-4methoxybenzo[d][1,3]dioxole-5-carboxamide (23)

The oxazine 10 (0.05 g, 0.08 mmol) was dissolved in THF/H₂O 10:1 (3.3 mL), affording a clear solution. Aluminum turnings (Flinn Scientific, 5 mesh, 0.022 g, 0.8 mmol) were dipped, in sequence, in a 1% aq. KOH and 0.5% aq. HgCl₂ solution, and were added immediately to a stirring solution of 10. The reaction mixture was stirred at room temperature for 15 h. The resulting viscous grey mixture was then filtered through a pad of Celite rinsed with MeOH, and the eluent was subsequently evaporated under vacuum. The resulting pink oil was purified by flash column chromatography [hexanes/EtOAc (1:1 v/v)] to give 23 as a white solid (0.024 g, 54%)

23: R_f = 0.28 [Hex/EtOAc (1:1 v/v)]; m.p. = 103–105 °C (i-PrOH); ${\left[\alpha \right]}_{\text{D}}^{24}-38.838$ (c = 0.485, CHCl₃); IR (neat) 3349, 3065−3032, 2977−2852, 1638, 1516, 1445, 1421, 1379, 1357, 1047 cm⁻¹; ¹H NMR (300 MHz, CDCl₃) δ 7.54−7.31 (m, 5H), 6.13 (s, 2H), 5.88 (s, 2H), 5.18 (s, 2H), 4.66 (s, 1H), 4.49 (d, J = 6.7 Hz, 1H), 4.38 (d, J = 6.5 Hz, 2H), 4.20 (m, 2H), 3.96 (s, 3H), 1.46 (s, 3H), 1.38 (s, 3H); ¹³C NMR (600 MHz, CDCl₃) δ 165.2, 140.8, 138.3, 137.0, 136.6, 131.2, 129.8, 128.6, 128.4, 109.1, 106.6, 102.4, 78.9, 76.5, 74.8, 68.0, 61.1, 60.5, 49.6, 27.0, 24.9, 14.3; HRMS (EI) calcd for C₂₅H₂₆BrNO₈: 547.0842, found 547.0824; Anal. Calcd for C₂₅H₂₆BrNO₈: C, 54.76; H, 4.78; Found C, 55.01; H, 4.78.

3.11. 7-(Benzyloxy)-6-bromo-N-((3aS,4R,7S,7aS)-7-((tertbutyldimethylsilyl)oxy)-2,2-dimethyl-3a,4,7,7atetrahydrobenzo[d][1,3]dioxol-4-yl)-4-methoxybenzo[d] [1,3]dioxole-5carboxamide (9)

The free alcohol 23 (0.153 g, 0.28 mmol) was dissolved in dry DCM (5 mL), affording a clear solution. After the temperature was brought to 0 °C, Et₃N (0.07 mL, 0.31 mmol) and TBSOTf (0.08 mL, 0.56 mmol) were added to the stirring solution. The ice bath was then removed, and the reaction mixture was stirred for 16 h at room temperature. The mixture was then diluted with DCM (2 mL), washed with 10% aqueous citric acid (1 × 3 mL), then the organic phase was dried over MgSO₄ and concentrated under reduced pressure. The crude product was purified by flash column chromatography [hexanes/EtOAc (4:1 v/v)] to afford the product as a colorless semisolid (0.15 g, 81%).

9: R_f = 0.67 [Hex:EtOAc (2:1 v/v)]; ${\left[\alpha \right]}_{\text{D}}^{24}-10.244$ (c = 2.4, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 7.50−7.33 (m, 5H), 6.57 (d, J = 9.6 Hz, 1H), 6.20−6.15 (m, 2H), 5.97−5.95 (m, 2H), 5.09 (s, 2H), 4.95−4.91 (m, 1H), 4.57 (dd, J = 6.9, 2.2 Hz, 1H), 4.36 (dd, J = 6.9, 1.7 Hz, 1H), 4.19 (dd, J = 5.1, 2.3 Hz, 1H), 3.91 (s, 3H), 1.37 (s, 3H), 1.32 (s, 3H), 0.68 (s, 9H), 0.03 (s, 3H), 0.02 (s, 3H); ¹³C NMR (600 MHz, CDCl₃) δ 164.6, 140.6, 138.0, 136.9, 136.6, 135.3, 132.5, 130.9, 128.6, 128.4, 128.3, 126.4, 108.4, 106.5, 102.2, 78.5, 76.8, 74.8, 66.4, 60.8, 45.8, 26.4−24.5, 17.7; IR (neat) 3254, 3065, 3033, 2984, 2856, 1648, 1602, 1444, 1420, 1358, 1250, 1049, 751, 695 cm⁻¹; HRMS (EI) calcd for C₃₁H₄₀BrNO₈Si: 661.1707; found 661.1698. anal. calcd for C₃₁H₄₀BrNO₈Si: C, 56.19; H, 6.08; found C, 55.96; H, 6.09.

3.12. (3aS,3bR,12S,12aS)-benzyl 10-(benzyloxy)-12-((tertbutyldimethylsilyl)oxy)-6-methoxy-2,2-dimethyl-5-oxo-3b,5,12,12atetrahydrobis([1,3]dioxolo)[4,5-c:4′,5′-j]phenanthridine-4(3aH)carboxylate (25)

The reaction was under anhydrous conditions. The substrate 9 (30 mg, 0.05 mmol) was dissolved in dry THF (2 mL) to gain an off-white solution. The mixture was placed in an ice bath and NaH (60% in mineral oil, 2 mg, 1.1 eq.) was added. The mixture was stirred for 10 min, then Boc₂O (16 mg, 1.5 eq.) was added, the ice bath was removed and the reaction was refluxed for 14 h. The mixture was then allowed to cool down to room temperature, H₂O (1 mL) was added and the mixture was extracted with EtOAc (3 × 1 mL). The collected organic fractions were dried over MgSO⁴ and, after filtration and concentration, the substrate (25 mg, 0.031 mmol) was dissolved in anhydrous anisole (1 mL), then Pd(OAc)₂ (1.9 mg, 0.0062 mmol, 0.2 mol%) was added together with Tl(OAc) (16 mg, 0.062 mmol, 2 eq.), and the mixture was stirred for 5 min. The yellow solution was degassed with Ar for 5 min, then dppe (4.8 mg, 0.012 mmol, 0.4 mol%) was added and the mixture was heated with a condenser at 130 °C for 12 h. The reaction mixture was then cooled and filtered through Celite with EtOAc as solvent, the filtrate was washed with HCl 2 M (2 × 2 mL), and then subjected to pre-parative TLC using hexanes/EtOAc (2:1) as eluent. The product was isolated as a yellow solid with a 10% yield (2.3 mg, 0.005 mmol).

25: R_f = 0.42 [Hex:EtOAc (2:1 v/v)]; mp = 131–133 °C (DCM/pentane 1:1); ${\left[\alpha \right]}_{\text{D}}^{24}62.378$ (c = 0.8, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 7.39−7.35 (m, 5H), 6.62 (d, J = 9.4 Hz, 1H), 6.12 (m, 1H), 6.02 (s, 2H), 5.28 (s, 2H), 4.81 (dd, J = 3.4, 1.3 Hz, 1H), 4.42 (m, 1H), 4.35 (dd, J = 6.9, 2.1 Hz, 1H), 4.26 (dd, J = 7.0, 3.3 Hz, 1H), 3.92 (s, 3H), 1.39 (s, 3H), 1.32 (s, 3H); ¹³C NMR (300 MHz, CDCl₃) δ 166.2, 148.8, 143.6, 138.3, 131.2, 130.4, 128.5, 108.9, 108.1, 102.2, 100.8, 78.7, 67.6, 56.7, 48.8, 26.7, 24.6; IR (neat) 3348, 3063, 2921, 2852, 1657, 1609, 1210, 1153, 686 cm⁻¹; HRMS (EI) calcd for C₂₆H₂₇NO₆: 449.1838; found 449.1812.

3.13. (2S,3R,4S,4aR)-11-(Benzyloxy)-2,3,4,7-tetrahydroxy-3,4,4a,5-tetrahydro[1,3]dioxolo[4,5-j]phenanthridin-6(2H)-one (8)

The reaction was performed under anhydrous conditions. The substrate 25 (15 mg, 0.02 mol) was dissolved in acetonitrile (1 mL) to gain a pale yellow solution. TMSCl (5% in acetonitrile, 1.1 eq) was added dropwise over 2 min followed by KI (4 mg, 0.022 mol, 1.1 eq) and the mixture was stirred at 60 °C for 2 h. The reaction was then quenched by adding HCl 6 M (2 mL) and an extraction was performed using EtOAc (8 × 1 ml). The collected organic fractions were dried over MgSO₄, filtered by gravity and concentrated under reduced pressure and purified by flash column chromatography [DCM/MeOH (100:1 v/v)] to obtain the product as a pale yellow solid (7 mg, 46%).

8: R_f = 0.23 [DCM/MeOH (70:1 v/v)]; ${\left[\alpha \right]}_{\text{D}}^{24}-126.6$ (c = 0.4, DMSO); mp = 87−89 °C (DCM); IR (neat) 3323, 3054, 2982,=2938, 2901, 1638, 1549, 1237, 1044, 708 cm⁻¹; ¹H NMR (300 MHz, DMSO-d₆) δ 13.25 (s, 1H), 7.89 (s, 1H), 7.52 7.28, (m, 5H), 6.16 (s, 1H), 6.08 (s, 2H), 5.49−5.24 (br s, 3H), 4.69 (s, 2H), 4.22 (d, J = 3.2 Hz, 1H), 4.04 (s, 1H), 3.82 (d, J = 3.1 Hz, 1H), 3.72 (s, 1H);¹³C NMR (150 MHz, DMSO-d₆) δ 169.4, 152.8, 145.3, 138.5, 133.9, 132.6, 129.6, 129.1, 128.8, 128.5, 125.2, 106.0, 102.6, 96.3, 72.8, 69.6, 69.3, 53.3; HRMS (EI) calcd for C₂₁H₁₉NO₈: 413.1111; found: 413.1027.

Cell Viability Assay:

Cell viability was measured by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay. Briefly, neuroblastoma cell line BE(2)-C cells (obtained from American Type Culture Collection) were maintained in Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F12) media supplemented with 10% fetal bovine serum (FBS). 1000 cells per well were plated into 96-wells plate, and cells were treated with a serial of dilutions (from 500 μM to 0.1 nM) of individual compounds in triplicate for five days. After five days, cells were replaced with MTT reagent (at 0.18 mg/mL in DMEM/F12) in each well and incubated with cells for 1 h at 37 °C. Culture media was removed after microplates were spin at 2000 rpm for 5 min. DMSO was then used to dissolve the crystals formed in each 96 well. Optical density values at wavelength 570 nm and 630 nm were measured using SpectraMax 190 (Molecular Devices, San Jose, CA), and the difference in the two optical density values was used to analyze the relative cell survival in each well. IC₅₀ values were calculated using Graphpad Prism software.