Unique cyclolignan’s architecture obtained via acid catalyzed cyclization/intramolecular Friedel-Crafts tandem reaction

We examine the cyclization of chiral bis-benzylidenesuccinate 6 (precursor to podophyllotoxin-related cyclolignans) in different conditions. Under acid catalysis, a new type of cyclolignan was obtained via cyclization/intramolecular Friedel-Crafts tandem reaction. The structure of the product was confirmed by spectroscopic data and X-ray diffraction. Also, a plausible mechanism of this transformation was proposed

8][9] Our research group is interested in the stereoselective synthesis of cyclolignans, achieved by photocyclization of chiral bis-benzylidenesuccinates. 10 We found that L-prolinol is an excellent chiral auxiliary 11 and used for the formal synthesis of PPT 12 and the total synthesis of (+)-epigalcatin. 13We also studied the regioselectivity in the photocyclization of cyclolignan precursors 14 and we used the bis-benzylidenesuccinyl platform as a photolabile protecting group for secondary amines. 15lthough the photocyclization of chiral bis-benzylidenesuccinates is relatively better studied, it is also possible to obtain cyclolignans via acid catalyzed cyclization, as shown in Charlton's total synthesis of (+)-lyoniresinol dimethyl ether, in which (-)-ephedrine was used as a chiral auxiliary and triflic acid (TfOH) as the catalyst. 16erein, we report a case in which the acid catalyzed cyclization of a chiral bis-benzylidenesuccinate was followed by an intramolecular Friedel-Crafts cyclization, leading to a novel type of cyclolignan.The results are compared with these obtained for the photocyclization of 6 in protic and aprotic solvents.

Synthesis of axially chiral bis-benzylidenesuccinate
The synthetic pathway starts with the opening of fulgide 4 (prepared according to previously reported procedure 15 ) with L-prolinol leading to succinamic acid 5.An atroposelective macrolactonization using BOP (benzotriazol-1-yloxytris(dimethylamino)phosphonium hexafluorophosphate ; Castro's reagent) as a coupling agent gave product 6 with 93% yield (Scheme 1).Only one rotamer was obtained, which was confirmed by NMR analysis.In our previous studies 12,14 we have shown that the use of L-prolinol as a chiral auxiliary provides the P-isomer of the macrocyclic compounds of type 6 (bis-benzylidenesuccinate cyclic amide-ester), which additionally had the cis configuration at C-1 and C-2 chiral centers and the same absolute configuration as in 1.This is a clear advantage of L-prolinol over (-)-ephedrine, which induces the M-configuration 16 of the bis-benzylidenesuccinate cyclic amide-ester and, after cyclization, leads to a trans configuration at C-1 and C-2, and the absolute configuration at C-1 opposite to that found in 1. Scheme 1.The synthesis of cyclic amidester 6.

Photocyclization
With product 6 in hand, we started to study the cyclization of this compound in different conditions.We began with a photocyclization using a medium-pressure mercury lamp (λ max ≈ 365 nm).In CHCl 3 and protic solvents (MeOH and AcOH) a strong photochromism was observed during the irradiation, but the reaction mixture gradually bleached and became fluorescent under UV-light.The isolated product was subjected to spectral analysis.Mass spectrometry showed that the molecular weight exceeded by 14 units the mass of 6.Additionally, 1 H NMR spectrum showed singlet at 10.24 ppm characteristic for phenol -OH group.These observations led us to conclusion that an aromatization of the naphthalene ring occurred, along with the incorporation of an oxygen atom at C-4 (Scheme 2).Scheme 2. Photocyclization of 6 in protic solvents.
In aprotic solvent (toluene) we observed a color change from slightly yellow to red during the irradiation, which indicates formation of an intermediate product.The color did not disappear even after long irradiation and no product was isolated.We decided to add an acidic additive to protonate the intermediate.After addition of p-toluenesulfonic acid (p-TsOH, catalytic amount) discoloration has occurred instantaneously and an intense blue fluorescence appeared (Figure 2.).After column chromatography we isolated compounds 8 and 9 as a mixture.Their preparative separation by column chromatography was not possible.We have noticed that aromatization of compounds 8 and 9 to compound 10 (confirmed by NMR and MS analyzes) occurred during column chromatography and in solution (air oxidation) (Scheme 3).Scheme 3. Photocyclization of 6 in aprotic solvent.

Acid catalyzed cyclization
Based on Charlton's results, 16 we performed an analogous experiment, in which a solution of 6 was treated with TfOH.The reaction mixture immediately became coloured upon the addition of TfOH, but after purification we obtained a single product with 83% yield. 1 H NMR analysis revealed only 3 aromatic protons and we thus assumed the creation of an additional ring, most likely as a result of an intermolecular Friedel-Crafts reaction, which occurred after cyclization (Scheme 4).From the HMBC, HSQC and COSY spectra (see Supporting Information) we concluded that the proton at C-1 position is a doublet at 4.13 ppm and J = 8.4 Hz and proton at C-2 position is a doublet at 4.87 ppm and J = 8.4 Hz.In the ROESY spectrum (see Suporting Information) we observed an interaction between these protons which indicates a cis configuration.A reasonable mechanism for this transofmation is presented in Scheme 5.It starts with the protonation of the ester carbonyl group.After migration of the double bond, the carbocation at benzylic position is formed, which is prone to cyclization to form the central dihydronaphtalene motif.The dihydronaphtalene is deprotonated at the C9 position and protonated from the less hindered side at C2, to form the partially aromatic 1,2-dihydronaphtalene ring.This leads to formation of carbocation at the C12 position that allowed a subsequent intramolecular Friedel-Crafts reaction.Then, the 8-membered ring is opened and in the last step oxygen atom is protonated leading to compound 11.Scheme 5.The mechanism of acid catalyzed transformation of 6 into 11.
Johnson et al. obtained a similar polycyclic architecture using Rh catalyzed intramolecular hydroacylation of an appropriate trisubstituted alkene. 17However, compared to our results the previous method requires a heavy metal catalyst, leads to the racemic product and has low yield (43%).

Conclusions
We have studied cyclization of chiral bis-benzylidenesuccinate 6 in different conditions.Acid catalyzed cyclization led us to product with additional ring resulting from intramolecular Friedel-Craft reaction.We have confirmed its structure by X-ray diffraction and proposed a plausible reaction mechanism.The resulting compound belongs to a still unexplored class of cyclolignans with potential biological activity.In comparison to previous synthesis of similar polycyclic compounds, our approach avoids the use of heavy metal catalysts and leads to the product in its single enantiomeric form.We believe that the reaction is driven both by the high protonating power of the superacid TfOH and the high electron density at the trimethoxy-substituted aryl rings.Further experiments with differently substituted bis-benzylidenesuccinates are currently underway.

Experimental Section
General.The reagents and solvents were commercially available and were used without further purification.Thin layer chromatography (TLC) analyses were performed on Merck Kieslgel 60 F-254 plates and visualization was done under UV light.Evaporation of solvents was performed at reduced pressure, using a Büchi rotary evaporator.Melting points were determined on an Electrothermal, Model IA 9200 apparatus and are uncorrected. 1H and 13 C NMR spectra were recorded on a Bruker AVANCE 500 spectrometer.Chemical shifts were reported in ppm from tetramethylsilane with the solvent resonance as the internal standard in CDCl 3 solution.The following abbreviations are used: m-multiplet, s-singlet, d-doublet, t-triplet, q-quartet, br-broad.High-resolution mass (ESI-TOF MS) spectra were recorded on a Micromass LCT spectrometer.(3E,4E)-3,4-Bis-(3,4,5-trimethoxybenzylidene)dihydrofuran-2,5-dione (4) was synthesized as reported previously. 10

f]pyrrolo[2,1-c][1,4]oxazocine-6,13-dione (7).
In a 50 mL quartz cuvette, 50 mg (0.093 mmol) of compound 6 was dissolved in 50 mL of MeOH and the solution was flushed with dry argon for 1 h.The reaction mixture was irradiated using a medium pressure mercury lamp.During irradiation, the solution was vigorously flushed with dry argon, which guaranteed constant mixing and deaeration of the solution.The color of the reaction mixture changed within the first 20 s of irradiation from colorless to yellow with a strong blue fluorescence.After 40 min of irradiation the solution was transferred into a round bottom flask and the solvent was removed under vacuum.The resulting yellow oil was purified on a silica gel column using Et 2 O/EtOAc (gradient, from 50% to 100% of EtOAc) mixture as an eluent.Product 7 (23.5 mg, 0.044 mmol, 47%) was obtained in the form of a thick oil. 1

Figure 2 .
Figure 2. Photographs of reaction mixture during irradiation of 6 in toluene.A -pure toluene in cuvette and solution of 6 in toluene in a vial; B -solutions after mixing in cuvette; C -solution during the irradiation without acidic additive; C -solution during irradiation immediately after addition of acidic additive; Dsolution during irradiation few minutes after addition of acidic additive.

Figure 3 .
Figure 3.The ORTEP view of the molecule.The non-hydrogen atoms are shown as 30% probability ellipsoids.