First total synthesis of chromanone A, preparation of related compounds and evaluation of their antifungal activity against Candida albicans, a biofilm forming agent

A straightforward and convenient approach for the first total syntheses of chromanone A and a related 7-OMe substituted natural product is reported. These unique C-3 substituted 2-hydroxymethyl chromones were recently isolated as fungal metabolites. Chromanone A was synthesized in 25.3% overall yield from the readily available pyrocatechol, whereas the second natural product was prepared in 39.7% global yield. A small library of chromones, including both natural products and some of their synthetic heterocyclic precursors, was evaluated against Candida albicans ATCC 10231, a biofilm forming agent. It was found that 8-methoxy-3-methyl-4-oxo-4H-chromene-2-carbaldehyde, a partially oxidized form of chromanone A, exhibited a minimum inhibitory concentration of 7.8 μg mL−1 and significantly inhibited the yeast's virulence factors, including the adherence to buccal epithelial cells and the secretion of phospholipases, as well as the formation of germ tubes and the generation of the hyphal pseudomycelium. In addition, despite the heterocycle exhibiting non-significant inhibition of the formation of the Candida biofilm, it completely inhibited the growth of C. albicans in preformed biofilms at 62.5 μg mL−1.


Introduction
Biolms are structured microbial communities, embedded in a self-produced polymeric matrix of protein and polysaccharides. They are frequently found on interfaces or attached to surfaces, and are the scenario of complex interactions. These enduring structures are elusive to the defense system of the host, reluctant to undergo phagocytosis, and offer the producing microorganism a markedly increased resistance to antibiotics and chemical disinfectants. 1a Therefore, biolms are an important form of microbial resistance and transmission of chronic infections. 1b The biolms are ubiquitous, and their presence on surfaces such as biomedical implants, may put at risk complex surgical interventions and ultimately compromise the life of the patient.
In this context, there is a need for antimicrobial agents capable of attacking microorganisms within the biolms. Hence, the search for new compounds endowed with this property is currently relevant.
The chromone skeleton is a "privileged structure", 2a meaning a promising motif for drug development. Not surprisingly, substituted chromones are known to display a variety of useful biological properties, including antifungal activity. 2b,c In 2009, Gamal-Eldeen and coworkers reported the isolation of chromanone A (A, Fig. 1), 3 from an algicolous marine Penicillium species, cultivated on a solid biomalt medium. In turn, this fungus was isolated as an endophyte of the Egyptian green alga Ulva sp. The natural product inhibits the activity of CYP1A at a level of 4 mg mL À1 , being a potential cancer chemopreventive agent. 3 The structure of the natural product was assigned based on an analysis of 1D and 2D (HSQC, HMBC) NMR spectra. Being a member of the very small family of the naturally occurring 2hydroxymethylchromones which bear a C-3 functionalization, chromanone A is a structurally unique chromone. Its special structural characteristics are shared only by a handful of natural products, such as its isomer B, recently isolated from Rhinocladiella sp., a fungus obtained from the marine sponge Ircinia oros. 4a These features are also found in heterocycles C 4b and D, 4c as well as in boeravinone Q (E) 5a and its congener mirabijalone C (F). 5b Our work is focused on the synthesis of structurally unique heterocyclic natural products, 6 as well as in their evaluation. 7a,b Further, among chromone derivatives, we have developed the total synthesis of the structure assigned to aspergillitine, 7c a 2,3dimethyl chromone derivative isolated from a marine Aspergillus species.
In pursuit of these interests, here we report an efficient approach to the rst total syntheses of chromanone A (A) and of its isomer, the related natural product B. The results of the evaluation of both natural products and their heterocyclic synthetic intermediates, as antifungal agents against the bio-lm forming yeast Candida albicans ATCC 12031, are also discussed.

Chemistry
The initial step toward chromanone A as the rst target (1), was a retrosynthetic analysis of the product (Scheme 1). Initially, focus was made on the oxygen bearing functionality of the 2hydroxymethyl feature; a functional group interconversion was proposed, conjecturing that the proper oxidation stage of the C-2 substituent could be set either through reduction (aldehyde, ester, 2a) or by means of a selective oxidation. 7a The latter possibility revealed the 2,3-dimethylchromone 2b as a suitable precursor.
Then, two different alternatives were considered. In one of them (Route a), the heterocyclic ring of 2a,b was disassembled as shown, to unveil a propiophenone derivative (3), which was further submitted to a C-O disconnection on the aliphatic side of the carbonyl moiety (bond a), to uncover ortho-vanillin (5) as a suitable starting material.
In the second case (Route b), an additional methyl ether C-O disconnection was considered on the chromone 2, which suggested the catechol derivative 4 as the most appropriate synthetic intermediate. In turn, a C-O disconnection on the aromatic side of the carbonyl motif of propiophenone 4 (bond b) set aside the three-carbon side chain and determined that the logical starting material for this approach was the commercially available pyrocatechol (6).
For simplicity, the Route a was explored rst. Thus, orthovanillin (5) was subjected to a 1,2-addition to the carbonyl with excess ethyl Grignard reagent (Scheme 2), freshly prepared from iodoethane and activated magnesium. The reaction was executed at room temperature (RT), giving alcohol 7 in 97% yield. 8 Subsequently, 7 was submitted to a selective oxidation of its secondary alcohol moiety. Different conditions [PDC, CH 2 Cl 2 ; H 2 N-NH 2 $xH 2 O, DMSO/I 2 , MeCN/H 2 O (5 : 1, v/v); DMSO/Ac 2 O (2.3 equiv.); IBX (1.5 equiv.), EtOAc] 9 were tested; however, the reaction proved most successful with IBX in EtOAc, furnishing ketone 3 in up to 81% yield in just one hour at room temperature. 10 Unfortunately, this transformation proved to lack robustness, being highly dependent on the experimental conditions, including the source of the oxidizing reagent. This characteristic frequently resulted in an exacerbated concomitant oxidation of the ortho-phenol, ultimately causing tarry materials and low yields. Scheme 1 Retrosynthetic analyses of chromanone A (1).
Next, a one-pot Kostanecki-Robinson cyclization protocol was practiced on the propiophenone 3 in order build the chromone system. The reaction of aromatic ketones with a glycolate ether was considered as an alternative, to access a 2alkoxymethylene chromone; however, it generally proved to proceed in low yield 11a and was soon discarded due to inability of the product to withstand the harsh conditions required to break the ether bond and the lack of selectivity of the reaction. In addition, model experiments with the scarcely precedented use of glyoxal 11b as a two-carbon synthon, which would have delivered a 2-hydroxymethyl chromone directly, met with failure; no reaction was observed, despite different basic conditions (KOH, K t BuO and NH 4 OH in EtOH) were explored. Therefore, the ketone 3 was exposed to ethyl chlorooxoacetate, 12 which can be regarded as an oxidized form of glyoxal or glycolate ethers. Luckily, the reaction took place in CH 2 Cl 2 , in the presence of triethylamine at 100 C under microwaves irradiation (MW), giving chromone 8 in a reproducible but rather moderate (42%) yield.
Finally, the required adjustment of the oxidation state of the C-2 substituent was carried out by means of a selective reduction of 8; however, since the use of NaBH 4 as reducing agent 13 did not perform as expected, this challenging transformation was carried out with Ca(BH 4 ) 2 . 14 The reagent was prepared in situ by adding a stoichiometric amount of CaCl 2 to NaBH 4 in EtOH. 14c Under these conditions, chromanone A (1, compound A of Fig. 1) was obtained in 41% yield. Its spectroscopic data in CDCl 3 conrmed its structure, whereas the NMR spectra taken in MeOH-d 4 were in full agreement with those reported for the natural product. 3 Signal enhancement (NOE) of the CH 2 OH moiety (0.8%) was observed upon irradiation of the hydrogen atoms of the 3-Me group, as well as between the hydrogen atoms of the 8-OMe group and H-7 (1.7%). Notably, the spectra taken in both solvents were alike, and differences exceeding 10 ppm were found between them, particularly for the carbon atoms of the isocyclic ring attached to oxygen.
Despite this rst approach to 1 afforded the natural product in just four steps and 13.5% overall yield, the lack of robustness of the selective oxidation of 7, coupled to the moderate yields of the cyclization toward 8 and its selective ester moiety reduction stages, suggested the need to devise an improved alternative. Therefore, we resorted to the second synthetic plan toward 1, commencing with the selective Friedel-Cras ortho-acylation of pyrocatechol (6) under BF 3 $OEt 2 promotion. 15 However, since in our hands the reported procedure proved hard to be reproduced, a series of optimization experiments were run in order to nd the proper reaction conditions.
In the process, it was soon found that ZnCl 2 and AlCl 3 are not suitable promoters, and that the use of 1,2-dichlorobenzene as solvent negatively affects the reaction performance. It was also discovered that under conventional thermal conditions (Table  1), the transformation hardly proceeded aer 3 days at 110 C (entry 1), whereas only small amounts of product were recovered when the reaction was performed at 180 C for 5 h (entry 2).
On the other hand, employing microwaves radiation, no product was isolated aer 10 min at 180 C (entry 3), whereas slightly milder conditions (170 C, 5 min) resulted in 33% yield of product (entry 4), when 10 equiv. EtCO 2 H were used. Increasing the amount of acid caused a further increase in the yield to 61% (entry 5), while unexpectedly, additional amounts of acid produced a drastic yield reduction (entry 6). In addition, performing the reaction at 160 C for 10 min (entry 7) or in the presence of propionic anhydride (entry 8) did not improve the results.
Therefore, the reaction was best performed as in entry 5, in a solventless condition and under microwaves irradiation, affording consistently over 60% yield of the expected product 4. 16 The use of 6 instead of guaiacol for this approach was based on literature precedents, which suggested that the latter should not be a suitable starting material because it would afford the unwanted propiophenone isomer. 17a,b Next, 4 was cyclocondensed and further cyclized with Ac 2 O under Kostanecki-Robinson conditions, uneventfully affording the expected 2,3-dimethylchromone 9 (ref. 7c) in 76% yield (Scheme 3). A subsequent Williamson O-methylation of 9 with MeI/K 2 CO 3 in reuxing acetone gave 94% yield of 2. 17c This was followed by a selective oxidation with the versatile I 2 /DMSO reagent system, which conveniently furnished aldehyde 10 in 83% yield. The transformation, which was carried out aerobically, required the addition of TsOH. 18a Finally, the reduction of the formyl moiety with NaBH 4 in EtOH uneventfully provided the expected product 1 in 70% yield. Interestingly, unlike the reduction of compound 8, in this case the use of Ca(BH 4 ) 2 proved to be unnecessary, since the formyl group attached to C-2 is much more reactive than the C-4 ketone moiety.
This optimized approach gave 1 in 25.3% overall yield, aer ve synthetic steps. The NMR spectroscopic data of the synthetic compound (in MeOH-d 4 ) were in excellent agreement with those of the literature, 3 and the heterocycle obtained through Route a, conrming the structure of the natural product.
The intimate details of the mechanism of the key I 2 /DMSOmediated oxidation toward 10 remain unclear; however, based on some previous literature precedents, a polar rather than a free-radical reaction mechanism can be proposed. Further, considering that the 2-methylchromone moiety may be regarded as a vinylogous a-methylketone, 18 it can be conjectured that a Kornblum-like oxidation is at the heart of the mechanistic sequence (Scheme 4). In this scenario, at rst the carbonyl group of the substrate (2) would be activated by the added TsOH, giving rise to intermediate I. In turn, this intermediate would be subjected to deprotonation to provide the dienol II, being followed by an iodination with I 2 , to generate the reactive iodide III.
Next, in the presence of DMSO which plays the dual role of solvent and oxidant, the iodide III would undergo a SN 2 -type reaction with the nucleophilic oxygen atom of the reagent, losing iodide and forming the alkoxysulfonium salt intermediate IV. The latter would undergo a proton abstraction, resulting in the aldehyde product 10. Interestingly, it has been reported that reaction of a primary iodide like III with KO 2 in DMSO afforded a hydroxymethyl chromone, albeit in rather low yield (27%). 19 During the reaction two molecules of HI and one of SMe 2 are produced. Although DMSO can oxidize iodide ions to iodine, generating SMe 2 (DMSO + 2I À / SMe 2 + I 2 + H 2 O), it is assumed that the presence of oxygen (air) under the strenuous reaction conditions (130 C) would serve to reoxidize all the SMe 2 formed in the transformation and/or to regenerate the iodine in the presence of DMSO.
The general guidelines provided by the retrosynthetic analysis of Scheme 1 were used to synthesize compound 16 ( Fig. 1, compound B) and to access additional heterocyclic intermediates for bioactivity testing.
Scheme 4 The Kornblum oxidation-based mechanistic proposal for the conversion of 2,3-dimethylchromone 2 into 10 with the I 2 /DMSO reagent system.
The phenol was methylated under conventional conditions, with MeI in reuxing acetone, using K 2 CO 3 as base to afford 14 (93%). This is a natural product, which has been recently isolated from Rhinocladiella sp. 4b and from the co-culture of a marine-derived Actinomycete (Streptomyces rochei MB037) and the fungus Rhinocladiella similis 35. Compound 14 proved to display weak antibacterial activity against Staphylococcus aureus and Pseudomonas aeruginosa. 21 Next, 14 was selectively oxidized with the DMSO-I 2 reagent system, providing aldehyde 15 in 78% yield. 22 Finally, the aldehyde was selectively reduced with the aid of the NaBH 4 , affording the alcohol 16 (Fig. 1, compound B) in 71% yield. This synthetic route provided the natural product 16 with an overall yield of 39.7% in just four steps.

Evaluation of bioactivity
The methods employed are fully described in the ESI. † Initially, antifungal susceptibility tests were carried out, with the determination of the minimum inhibitory concentration (MIC) and the minimum fungicidal concentration (MFC). The antifungal activity of the compounds was assessed in the concentration range 0.49-250 mg mL À1 with the standardized CLSI microbroth dilution method for yeasts 23 against C. albicans ATCC 10231. Initially, the aldehyde 10 and the natural product (1) were screened; however, in order to obtain a better insight into the structural factors implied in the bioactivity, their precursors 2 and 9 as well as the related 7-substituted 2,3-dimethylchromones 13 and 14 were also tested, along with the aldehyde 15 and the natural product 16.
The results are collected in Fig. 2, which depicts the percentage of growth inhibition of a standardized inoculum of C. albicans (1 Â 10 3 CFU per mL) plotted against the logarithm of the corresponding concentration of each compound.
The assay revealed that the natural product (1) is a very weak inhibitor, being considered inactive. It barely caused 50% inhibition at a concentration of 250 mg mL À1 (MIC > 250 mg mL À1 ); quite a similar prole was exhibited by its phenolic precursor 9 (63% inhibition at 250 mg mL À1 ). On the contrary, the aldehyde 10 proved to be a very good inhibitor with MIC ¼ 7.8 mg mL À1 .
Regarding the 7-substituted chromones, compounds 13, 14 and 16 demonstrated to be inactive. Compound 14 exhibited 72.8% inhibition at 250 mg mL À1 (MIC > 250 mg mL À1 ), whereas aldehyde 15 was moderately active, with a MIC value of 62.5 mg mL À1 . The MFC values for the isomeric active aldehydes 10 and 15 were also determined according to the established protocol, 23 being 125 mg mL À1 in both cases.
The results suggested that the presence of a formyl moiety is relevant for the observed activity, and that the latter can be modulated by the position of the oxygenated substituent in the homocyclic ring.
Next, the effect of the heterocycles on the virulence factors of C. albicans were examined. The tests included inhibition of the adherence to buccal epithelial cells, inhibition of the formation of the germ-tube, morphogenesis of C. albicans on solid media and the inhibition of lytic enzymes.
The rst step by which a microorganism can initiate an infection is through adherence to an epithelial surface; this ability enables it to exist in biolms and is in clear association with its virulence. 24 In order to evaluate whether the active compounds are able to affect this process, at sub-lethal concentrations, compounds 10 and 15 were submitted to the assay of inhibition of the adherence to buccal epithelial cells (BEC).
The results (Fig. 3A) showed that the number of yeasts adhered to 100 BEC decreased from 2972 AE 233 in the untreated control cells to 177 AE 60, in the presence of compound 10 at MFC/2; interestingly, the compound was still active at MFC/32 (number of yeasts adhered to 100 BEC ¼ 1724 AE 399). Further, as seen in Fig. 3B, treatment with sub-lethal concentrations of compound 15 also caused a remarkable decrease in yeast adherence to BEC at levels ranging from MFC/2 (482 AE 187) to MFC/8 (1334 AE 233).  These results clearly indicated that the amount of adhered fungal cells to BEC was signicantly lower (Wilcoxon test, p < 0.0001) in yeasts treated with both aldehydes than in the untreated cells, suggesting that the presence of these heterocycles causes some degree of resistance to the colonization of BEC by C. albicans.
In the germ-tube inhibition assay, the effect of different geometrically distributed sub-lethal concentrations (MFC/64-MFC/2) of compounds 10 and 15 on the formation of germ tubes (GT) in C. albicans was assessed.
Compound 15 proved to be a less powerful inhibitor (Fig. 4B), which inhibited hyphae formation up to a concentration of MFC/8 (GT% ¼ 59 AE 1.4%). At MFC/2, the observed GT% was 16.0 AE 1.4%. The results were all statistically signicant according to the Holm-Sidak test (p < 0.05).
The morphogenesis of C. albicans on solid media was studied by examining the effects of compounds 10 and 15 on the formation of pseudomycelium in C. albicans in the nutrientpoor Spider medium, that induces pseudohyphal morphogenesis. It was observed that the aldehyde 10 effectively reduced formation of the hyphal pseudomycelium up to a concentration of MFC/4. This effect was apparent by the smoother aspect of the colonies and the notable reduction of hyphae at the edges. On the other hand, in the presence of compound 15, the C. albicans colonies showed their typical lamentation at the edges, suggesting that this heterocycle is not an effective inhibitor of pseudomycelium formation.
Finally, in the study of lytic enzyme inhibition, it was detected a statistically signicant inhibition of phospholipases secretion in the presence of compound 10 at MFC/2 (Pz ¼ 0.92 AE 0.01) and MFC/4 (Pz ¼ 0.82 AE 0.02) with respect to the control (Pz ¼ 0.74 AE 0.02). Contrarily, however, no signicant changes in the Pz index were observed between the untreated (control) cells and those exposed to compound 15. This signaled that the secretion of phospholipases was not inhibited.
On the other hand, it was also observed that at sub-lethal concentrations, compounds 10 and 15 did not inhibit secretion of C. albicans esterases, since their values of the Pz index (Pz ¼ 0.82 AE 0.02) were the same as the control.
In view of these promising observations, the interaction of the chromone derivatives on formation of the C. albicans biolm and on preformed biolms was examined. In the rst case, an inoculum prepared according to Pierce et al. was used, employing yeast extract-peptone-dextrose (YPD) medium. 25 Fig . 5A displays the data related to the inhibition of the formation of the C. albicans biolm in the presence of different concentrations of compounds 10 and 15. A compound was considered active if it signicantly inhibited biolm formation at concentrations below the MIC (sub-inhibitory). However, it was detected that none of the heterocycles managed to completely avoid formation of the biolm.
In the presence of 10 at a level of 250 mg mL À1 , the inhibition was 67.6% (54.5% at 125 mg mL À1 ), while 15 exhibited meager values of 45.6% and 23.7% inhibition at the same concentrations, respectively. Not unexpectedly, at the corresponding MIC values, their degree of inhibition lacked statistical signicance; therefore, based on these results, none of the heterocycles inhibited biolm formation.
On the other hand, when the antifungal activity against preformed C. albicans biolms was evaluated, it was observed that compounds 10 and 15 displayed almost complete inhibition at concentrations of 62.5 and 250 mg mL À1 , respectively. Further, compound 10 caused over 60% inhibition when the concentration was halved. The effect was concentration dependent, as reected in the progressive increase in cell viability with reducing concentrations of both compounds.
The results indicate that, while the MIC of compound 10 against planktonic cells of C. albicans is 7.8 mg mL À1 , the aldehyde can completely inhibit the growth of the biolm yeast colonies at a concentration of 62.5 mg mL À1 , suggesting that it may be a useful and promising advantage in the development of more effective anti-biolm agents.

Conclusions
Two facile and expedient approaches to the rst total synthesis of chromanone A, a unique chromone isolated from an algicolous fungus Penicillium sp., endophyte of the Egyptian green alga Ulva sp., were developed. In the most efficient sequence, the heterocycle was accessed in only ve steps and 25.3% overall yield from pyrocatechol, a commercial and easily available starting material.
The key stages of the synthesis included a selective Friedel-Cras acylation of the starting catechol under BF 3 $Et 2 O promotion and the aerobic oxidation of the C-2 methyl group of a chromone intermediate by the I 2 /DMSO reagent system. To the best of our knowledge, this is the rst report of the use of this approach and this reagent system for the synthesis of 2formyl-and 2-hydroxymethyl chromones. A 7-methoxy isomer of the natural product, which is also a fungal natural product, was prepared following the same strategy. Neither of the synthetic sequences required the use of protecting groups, favouring the efficiency of the syntheses.
The tests of antifungal activity against C. albicans revealed that 8-methoxy-3-methyl-4-oxo-4H-chromene-2-carbaldehyde, the 2-formyl analogue of chromanone A, exhibited a MIC ¼ 7.8 mg mL À1 (MFC ¼ 125 mg mL À1 ). In addition, at sub-lethal concentration levels, this compound signicantly inhibited various C. albicans virulence factors, including the secretion of phospholipases and the yeast adherence to buccal epithelial cells, as well as the formation of the hyphal pseudomycelium and germ tubes. On the other hand, the formyl derivative completely inhibited the growth of C. albicans in preformed biolms at 62.5 mg mL À1 .
Taken together, the results indicate that the 2-formyl analogue of chromanone A not only kills C. albicans, but also inhibits its virulence factors, suggesting that it is able to target both, the cell growth and the pathogenic process. These features turn the aldehyde into a potentially useful lead to develop more effective agents to ght this yeast in biolm scenarios. These ndings also support the hypothesis that marine fungi may be important sources of inspiration for the development of new agents to combat biolm-forming microorganisms, and perhaps help to overcome difficulties or failure of current therapy for fungal infections.

General information
The chemical transformations were carried out under dry argon atmospheres, employing freshly distilled anhydrous solvents and oven-dried glassware. Anhydrous CH 2 Cl 2 was obtained from an MBraun solvent purier and dispenser system. Absolute EtOH was prepared by reuxing the commercial solvent over clean Mg/I 2 and distilling from the resulting magnesium alkoxide. Anhydrous Et 3 N was prepared by distillation of the commercial product from CaH 2 . The anhydrous solvents were stored in dry Young ampoules. The rest of the solvents and reagents were used as received.
The reactions were monitored by TLC, employing aluminium supported silica gel 60 GF 254 plates run in different hexanes:EtOAc mixtures. The chromatographic spots were revealed by exposure to UV light (254 and 365 nm), followed by spraying with the ethanolic p-anisaldehyde/sulfuric acid reagent and gentle heating to improve selectivity.
The ash column chromatographies were developed under positive pressure with slurry-packed silica gel 60 H for thin layer chromatography (particle size < 55 mm), employing gradient of solvent polarity techniques with hexanes:EtOAc.

Equipment
The melting points were determined on an Ernst Leitz Wetzlar model 350 hot-stage microscope and are informed uncorrected. The FT-IR spectra were acquired on a Shimadzu Prestige 21 spectrophotometer, as thin lms held between two NaCl disks or as solid dispersions in KBr (solid samples). The NMR spectra were recorded in CDCl 3 with a Bruker Avance 300 NMR spectrometer (300.13 MHz for 1 H and 75.48 MHz for 13 C). The chemical shis are reported in the d scale, in ppm downeld from TMS (d ¼ 0.0 ppm). The residual solvent peaks of CDCl 3 (d H ¼ 7.26 ppm, d C ¼ 77.16 ppm), were used as internal references. The scalar coupling constant (J) and width and half-height (w 1/2 ) values are given in Hertz. NOE and 2D NMR experiments (HSQC and HMBC) were also acquired in order to ensure unambiguous signal assignment. The HRMS were obtained with a Bruker MicroTOF-Q II instrument from UMyMFOR (Buenos Aires, Argentina). The microwave-assisted reactions were carried out in a CEM Discover microwave reactor. The light microscopy observations and determinations were performed with an Eclipse E100 (Nikon Corp., Tokyo, Japan) instrument. The microplates were read in a VERSA Max microplate reader (Molecular Devices, Sunnyvale, CA, USA).

Ethyl 8-methoxy-3-methyl-4-oxo-4H-chromene-2carboxylate (8)
Under an argon atmosphere, anhydrous triethylamine (198 mg, 1.95 mmol) and ethyl chlorooxoacetate (67 mg, 0.49 mmol) were successively added to a stirred solution of 3 (44 mg, 0.24 mmol) in anhydrous CH 2 Cl 2 (1 mL), placed in a microwave vial. The mixture was heated at 100 C in a microwave oven for 1 h; aer its completion was conrmed by TLC, it was cooled to room temperature and diluted with brine (10 mL). The reaction products were extracted with EtOAc (3 Â 20 mL). The combined organic extracts were washed with brine (5 mL

Author contributions
All the authors were involved in the manuscript, performing conceptualization (IC, TSK

Conflicts of interest
There are no conicts to declare.