New Bromo- and Iodo-Hydroxylactones with Two Methyl Groups Obtained by Biotransformation of Bicyclic Halolactones

: The subject of the research was to determine the ability of the ﬁlamentous fungi to biotrans-form bicyclic halolactones containing two methyl groups in their structure. By chemical synthesis three bicyclic halolactones with two methyl groups, one in the cyclohexane ring and one in the lactone ring, were obtained: 2-chloro-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one, 2-bromo-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one, and 2-iodo-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one. These compounds were formed as mixtures of two diastereoisomers. The obtained halolactones (as mixture of two diastereoisomers) were subjected to screening biotransformation with the use of eight strains of ﬁlamentous fungi: Fusarium culmorum AM10, F. avenaceum AM12, F. semitectum AM20, F. solani AM203, Absidia coerulea AM93, A. cylindrospora AM336, Penicillium chermesinum AM113, P. frequentans AM351. Two of the substrates, 2-bromo-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one and 2-iodo-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one, were hydroxylated without removing the halogen atom from the molecule, giving 2-bromo-7-hydroxy-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one, 2-bromo-5-hydroxy-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one, and 2-iodo-7-hydroxy-4,7-dimethyl-9-oxabicyclo[4.3.0]nonan-8-one as products. The hydroxylation capacity was demonstrated by strains of Absidia cylindrospora AM336, Fusarium avenaceum AM12, and F. solani AM203. The structures of all lactones were determined on the basis spectroscopic data.


Introduction
Biotransformations are a useful tool for obtaining interesting new derivatives of organic compounds. They can be used as an alternative to classical chemical synthesis [1][2][3][4][5]. The undoubted advantage of this method is the relatively low cost and the possibility of conducting transformations of compounds in mild conditions. An equally important aspect is the possibility of obtaining new compounds, which are not available through classical synthesis or require the use of specific catalysts. Such catalysts can be biocatalysts, which include whole cells of filamentous fungi or yeast or enzymes isolated from them. Biotransformations using whole fungal cells are often used to transform bicyclic lactones. These are usually lactones containing in their structure, apart from methyl groups, halogen atoms, or double bond. The most common products of such reactions are hydroxyl derivatives of lactones [6][7][8][9][10][11][12].
Hydroxylation reactions are mainly catalyzed by monooxygenases of cytochrome P450 (CYP). These enzymes are present in all kingdoms of life, with fungi being particularly diverse in terms of CYPs. Using data from genome sequencing projects, more than 300,000 CYPs have been identified to date [13]. Diversification of their forms results from the key function of these enzymes in the biosynthesis of primary and secondary metabolites, but primarily in the detoxification and biodegradation of xenobiotics [14,15].
However, hydroxylation reactions can also be catalyzed by unspecific peroxygenases (UPO), which are also widespread in the fungal kingdom [16]. The analysis of the genomes of basidiomycetes and saccharides revealed over a thousand putative peroxygenase genes [17].
In our research, we used lactones containing halogen atoms in the molecule of substrates. Previous studies carried out so far showed [18][19][20][21][22][23][24] that filamentous fungi, especially those belonging to the genera Fusarium and Absidia, can carry out the hydrolytic dehalogenation of halolactones. As a result, hydroxylactones with a different conformation than halolactones obtained by chemical synthesis were obtained. It is also possible to obtain halo-hydroxylactones, compounds in which the halogen atom remains intact, and additionally a hydroxyl group is introduced [20,25,26]. It is worth emphasizing that, due to the variety of enzymes capable of catalyzing the hydroxylation of xenobiotics, it is not possible to predict the position of hydroxylation without experimental basic research. Additionally, the disadvantage of isolated hydroxylases are their instability and the requirement to use expensive cofactors or auxiliary enzymes [16].
The analysis of 1 H NMR spectra of both isomers A and B of ester 2 and acid 3 showed that in both cases, the methyl group located at C-5 carbon took an equatorial position. This was indicated by the signals from H-5 protons of both A and B isomers, being wide multiplets in the range of 2.45-2.58 ppm (ester 2) and 2.53-2.63 ppm (acid 3). The side chain also occupied an equatorial position, as evidenced by wide multiplets in the range 1.97-2.06 and 1.99-2.09 ppm derived from H-1 protons of both isomers A and B ester 2 and acid 3, respectively ( Figure 2).
Analysis of 1 H NMR spectra of both diastereoisomers A and B chloro-4, bromo-5, and iodolactone 6 allowed to conclude that their spatial structures are similar. The cyclohexane ring was in the chair conformation. The methyl group lying on C-4 carbon took the equatorial position. This was evidenced by signals from H-4 protons belonging to both isomers, being broad multiplets located in the range of 1.   The analysis of 1 H NMR spectra of both isomers A and B of ester 2 and acid 3 showed that in both cases, the methyl group located at C-5 carbon took an equatorial position. This was indicated by the signals from H-5 protons of both A and B isomers, being wide multiplets in the range of 2.45-2.58 ppm (ester 2) and 2.53-2.63 ppm (acid 3). The side chain also occupied an equatorial position, as evidenced by wide multiplets in the range 1.97-2.06 and 1.99-2.09 ppm derived from H-1 protons of both isomers A and B ester 2 and acid 3, respectively ( Figure 2).    The comparison of the chemical shifts of the signals coming from the H-7A and H-7B protons allowed to determine the arrangement of the CH 3 -10 methyl group in the lactone ring. In the case of the A isomer of lactones 4, 5, and 6, the signal coming from the H-7 proton was at 2.34, 2.37, and 2.33 ppm, respectively. For the B isomer of lactones 4, 5, and 6, the signals from the H-7 proton were at 2.77, 2.77, and 2.78 ppm, respectively. This means that the H-7 protons of the A isomers was lying across the plane of the lactone ring, and the H-7 protons of the B isomers-in the plane of the lactone ring. This implied the appropriate position of the CH 3 -10 methyl group in the plane of the lactone ring (isomer A) and across the plane of the lactone ring (isomer B) (Figure 3). the halogen atom in both A and B isomers of lactones 4-6. The signals derived from the protons H-6A and H-6B were broad multiplets. They were in the ranges of 2.32-2.37, 2.39-2.44, and 2.46-2.53 ppm for the A isomer (compounds 4, 5, and 6, respectively). For isomer B, these signals were at 2.67 (dddd, J = 12.4, 10.0, 6.0 and 4.0 Hz), 2.70-2.76, and 2.81-2.89 ppm (compounds 4, 5, and 6, respectively). This means that in both isomers A and B, the C-C bond of the lactone ring assumed the equatorial orientation.
The comparison of the chemical shifts of the signals coming from the H-7A and H-7B protons allowed to determine the arrangement of the CH3-10 methyl group in the lactone ring. In the case of the A isomer of lactones 4, 5, and 6, the signal coming from the H-7 proton was at 2.34, 2.37, and 2.33 ppm, respectively. For the B isomer of lactones 4, 5, and 6, the signals from the H-7 proton were at 2.77, 2.77, and 2.78 ppm, respectively. This means that the H-7 protons of the A isomers was lying across the plane of the lactone ring, and the H-7 protons of the B isomers-in the plane of the lactone ring. This implied the appropriate position of the CH3-10 methyl group in the plane of the lactone ring (isomer A) and across the plane of the lactone ring (isomer B) ( Figure 3).

Preparative Biotransformation of Halolactones 5 and 6. Analysis of the Structures of Obtained Derivatives
After analyzing the results of the screening tests, it was found that that the microorganism transforming substrates with the highest efficiency was strain A. cylindrospora AM336. In order to determine the structure of the resulting products, a preparative biotransformation was carried out using strain A. cylindrospora AM336 as a biocatalyst. Bromolactone 5 and iodolactone 6 were used as substrates in this stage. The results of the preparative biotransformation are presented in Table 2. The biotransformation resulted in three new lactones. Two of them, compounds 7 and 8, were the products of hydroxylation of bromolactone 5, while compound 9 was formed by hydroxylation of iodolactone 6. For substrates (lactones 4-6), two peaks were observed on GC chromatograms for each compound, corresponding to two diastereoisomers A and B. In the case of biotransformation products (lactones 7-9), only single peaks were observed. Analysis of the NMR spectra confirmed that each of the biotransformation products was one compound. Therefore, optical rotation of lactones 7-9 was measured. In all three cases, the values obtained were different from zero. In the next step, the enantiomeric excesses of these compounds were determined using a column with chiral filling (Supplementary Materials, Figures S41-S43). The retention times of the analyzed compounds were consistent with our previous experience with structurally similar compounds [20,26]. The results are presented in Table 3. Analyzing the data presented in Table 3, the enantioselectivity of the hydroxylation process is dependent on the place where the hydroxyl group was introduced. In case of hydroxylation of secondary C-5 carbon being a fragment of cyclohexane ring, the process was highly enantioselective. On the other hand, when the hydroxyl group was introduced to the tertiary C-7 carbon fragment, a part of the lactone ring, the products were characterized by a slight enantiomeric excess.
Analysis of the 1 H NMR spectra of the obtained biotransformation products showed that strain A. cylindrospora AM336 preferred hydroxylation, while the hydrolytic dehalogenation characteristic for filamentous fungi was not observed. In the case of bromolactone 5, hydroxylation of carbon C-7 in the lactone ring (compound 7) or C-5 carbon in the cyclohexane ring (compound 8) took place. In the case of iodolactone 6, the hydroxyl group was introduced into the lactone ring only (compound 9). The structures of the obtained compounds are presented in Figure 4. observed on GC chromatograms for each compound, corresponding to two diastereoisomers A and B. In the case of biotransformation products (lactones 7-9), only single peaks were observed. Analysis of the NMR spectra confirmed that each of the biotransformation products was one compound. Therefore, optical rotation of lactones 7-9 was measured. In all three cases, the values obtained were different from zero. In the next step, the enantiomeric excesses of these compounds were determined using a column with chiral filling (Supplementary Materials, Figures S41-S43). The retention times of the analyzed compounds were consistent with our previous experience with structurally similar compounds [20,26]. The results are presented in Table 3. Analyzing the data presented in Table 3, the enantioselectivity of the hydroxylation process is dependent on the place where the hydroxyl group was introduced. In case of hydroxylation of secondary C-5 carbon being a fragment of cyclohexane ring, the process was highly enantioselective. On the other hand, when the hydroxyl group was introduced to the tertiary C-7 carbon fragment, a part of the lactone ring, the products were characterized by a slight enantiomeric excess.
Analysis of the 1 H NMR spectra of the obtained biotransformation products showed that strain A. cylindrospora AM336 preferred hydroxylation, while the hydrolytic dehalogenation characteristic for filamentous fungi was not observed. In the case of bromolactone 5, hydroxylation of carbon C-7 in the lactone ring (compound 7) or C-5 carbon in the cyclohexane ring (compound 8) took place. In the case of iodolactone 6, the hydroxyl group was introduced into the lactone ring only (compound 9). The structures of the obtained compounds are presented in Figure 4.  Analysis of the 1 H NMR spectra of the biotransformation products showed that these compounds were formed from the B isomer of bromolactone 5 or iodolactone 6. In the case of bromolactone, it was observed that tertiary carbon C-7 (lactone 7) or secondary carbon C-5 (lactone 8) were hydroxylated, while in the case of iodolactone, only the carbon C-7 (lactone 9) was hydroxylated. In all three lactones, the chair conformation of the cyclohexane ring was preserved. The H-6 proton retained its axial orientation, while H-1 and H-2 protons retained their equatorial orientation. The disappearance of the signal from the H-7 proton observed in the NMR spectra of compounds 7 and 9 indicates Analysis of the 1 H NMR spectra of the biotransformation products showed that these compounds were formed from the B isomer of bromolactone 5 or iodolactone 6. In the case of bromolactone, it was observed that tertiary carbon C-7 (lactone 7) or secondary carbon C-5 (lactone 8) were hydroxylated, while in the case of iodolactone, only the carbon Catalysts 2021, 11, 73 6 of 12 C-7 (lactone 9) was hydroxylated. In all three lactones, the chair conformation of the cyclohexane ring was preserved. The H-6 proton retained its axial orientation, while H-1 and H-2 protons retained their equatorial orientation. The disappearance of the signal from the H-7 proton observed in the NMR spectra of compounds 7 and 9 indicates hydroxylation of C-7 carbon. It is known from previous considerations that in the case of the B isomer, the CH 3 -10 methyl group lied across the plane of the lactone ring. This means that the hydroxyl group in products 7 and 9 was introduced into an equatorial position. In the case of lactone 8, the hydroxyl group was also introduced into an equatorial position, but this time on the C-5 carbon. This was evidenced by the position of the signal coming from the H-5 proton lying at 2.97 ppm and its shape (a triplet with a coupling constant of 10.0 Hz) and the differences in the position of the H-1 proton and the CH 3 -10 methyl group, which shifted towards the lower field as well as H-6 proton that shifted towards the higher field.
The formation of halo-hydroxylactones during biotransformation is not common. During our previous studies, the formation of halo-hydroxylactones was observed in two cases only. This was the case when halolactones with one methyl group located at C-6 carbon were used as substrates and the microorganism was the filamentous fungus Penicillium vermiculatum AM30 [20]. The second time, the microorganisms inducing the formation of such compounds were edible fungi of the genus Pleurotus ostreatus. At that time, the only effectively transformed substrate was chlorolactone with four methyl groups, two at C-4 carbon and one each at C-6 and C-7 carbon [26].
Based on our experience with the biotransformation of halolactones with whole cells of filamentous fungi, the formation of products such as hydroxylactone depends on the type of halogen. Typically, filamentous fungi slightest transforms lactones with chlorine atom in the molecule. The lack of products derived from chlorolactone 4 was, therefore, not surprising. Bromo-and iodolactones are much more frequently transformed by filamentous fungi [19][20][21][22]. Such a dependence can be seen here as well. Three strains were able to hydroxylate bromolactone 5: A. cylindrospora AM336, F. avenaceum AM12, and F. solani AM203, and for hydroxylation of iodolactone 6-only strain A. cylindrospora AM336. It should be emphasized that out of two diastereoisomers of halolactones 4-6, only one from them was hydroxylated, the B isomer. This means that the presence of a methyl group in the lactone ring has a great influence on the course of biotransformation. The presence of this methyl group prevents the process of hydrolytic dehalogenation, which was observed many times during the biotransformation of halolactones with Fusarium fungi. In the situation discussed here, the A isomer remained unreacted, only the B isomer, in which the methyl group lies across the plane of the lactone ring, was hydroxylated. This orientation of the methyl group allows the hydroxyl group to approach in the plane of the lactone ring. According to our experience, the hydroxyl group introduced into the cyclohexane ring of the lactone molecule usually took the equatorial position. This trend has been maintained here as well.

General Methods
The progress of subsequent steps of synthesis and the course of biotransformation was analyzed on GC on CP-3380 instrument (Varian, Agilent Technologies, Santa Clara, CA, USA) using DB-1 column (cross-linked methyl silicone gum 30 m × 0.25 mm × 0.25 µm). The following temperature program was used: injector 250 • C and detector (FID) 300 • C, initial column temperature 140 • C (hold for 1 min), ramp 140-180 • C at 15 • C/min, ramp 180-300 • C at 40 • C/min and hold 1 min at 300 • C. For purification of products obtained from the synthesis and biotransformation reactions, column chromatography was used, using as a carrier silica gel (Kieselgel 60, grid 230-400) and as an eluent a mixture of hexane and acetone in a ratio of 3:1. The enantiomeric excess was analyzed on GC on 6890N instrument (Agilent Technologies, Santa Clara, CA, USA) using a chiral column CP-Chirasil-B-325 (25 m × 0.25 mm × 0.25 µm, Supelco, Bellefonte, PA, USA). The temperature program for lactones 7 and 9 was as follows: injector 200 • C, detector (FID) 200 • C, initial column temperature 140 • C, ramp 140-200 • C at the rate of 1.0 • C/min, and hold at 200 • C for 1 min. For lactone 8, a slightly different temperature program was used: 200 • C injector, 200 • C detector (FID), 140 • C initial column temperature, ramp 140-160 • C at the rate of 0.5 • C/min, ramp 160-200 • C at the rate of 20 • C/min, and hold at 200 • C for 1 min. To confirm the molar masses of the compounds obtained, a high-resolution mass spectrometry analysis was performed using the Waters LCT Premier XE instrument (ESI ionization, Waters, Milford, MA, USA). The NMR spectra were recorded on the JEOL DeltaTM 400 MHz spectrometer (JEOL USA, Inc., Peabody, MA, USA) or on the Bruker Avance TM 600 MHz spectrometer, using CDCl 3 as a solvent. To determine the optical rotation of biotransformation products, the Jasco P-2000 polarimeter (Jasco, Easton, PA, USA) was used. Measurements were performed for chloroform solutions, in concentrations expressed in g/100 mL.

Organic Synthesis
Ethyl ester of (4,7-dimethylcyclohex-2-en-1-yl)acetic acid 2A + 2B The known allylic alcohol 1 [27] was used as a starting substrate. This compound was subjected to Claisen rearrangement with orthopropionate modification [28]. Briefly, 3.0 g, 0.027 mol of alcohol 1 was heated with 18 mL, 0.1 mol of triethyl orthopropionate and catalytic amount of propionic acid. After completion of the reaction (30 h), the product was purified by column chromatography giving 3.1 g (59%) of ester 2 as a pair of diastereoisomers A and B with the following spectral properties: 1  (4,7-dimethylcyclohex-2-en-1-yl)acetic acid 3A + 3B Basic hydrolysis of the obtained mixture of two diastereoisomers A and B of ester 2 was carried out according to the procedure described previously [27]. Briefly, 3.1 g of ester 2 was heated with 5% potassium hydroxide solution in methanol for 4 h. After evaporation of methanol, the residue was dissolved in water and acidified with 1 molar hydrochloric acid, and the product was extracted with diethyl ether. Then, 1.9 g (yield 72%) of acid 3 also as two diastereoisomers A and B was obtained. The spectral data of acid are as follows: 1   Chlorolactonization of acid 3 was carried out according to the procedure of Grabarczyk and Białońska [29]. Briefly, 0.6 g, 0.0036 mol of THF solution of acid 3 was stirred with 1.1 g of N-chlorosuccinimide for 24 h. After completion of the reaction, product was extracted with diethyl ether giving 0.31 g (43%) of chlorolactone 2 as a pair of diastereoisomers A and B with the following physical and spectral properties: 1  Bromolactonization of acid 1 was carried out according to the known method Grabarczyk and Białońska [29]. Briefly, 0.6 g (0.0036 mol) of tetrahydrofuran solution of acid 1 was stirred with 0.8 g N-bromosuccinimide for 24 h. After completion of the reaction product was extracted with diethyl ether giving 0.49 g (56%) of bromolactone 5 also as a pair of diastereoisomers A and B was obtained. The physical and spectral data of this products are as follows: 1  Iodolactonization of acid 3 was carried out according to the known procedure [27]. Briefly, 0.6 g (0.0036 mol) of acid 3 dissolved in diethyl ether was stirred with 1 g of sodium bicarbonate for an hour, and then 1 g of iodine in 2 g of potassium iodide was added. After 24 h, product was extracted with diethyl ether giving 0.39 g (37%) of iodolactone 6 as a pair of diastereoisomers A and B with the spectral properties: 1 Figures S21-S25).

Microorganisms
The biotransformations were carried out with the use of eight strains of filamentous fungi belonging to the own collection of the Department of Chemistry of Wrocław University of Life Sciences. These were Fusarium culmorum AM10, F. avenaceum AM12, F. semitectum AM20, F. solani AM203, Absidia coerulea AM93, A. cylindrospora AM336, Penicillium chermesinum AM113, P. frequentans AM351. These strains were cultured on Sabouraud's agar (0.5 g aminobac, 0.5 g peptone, 4 g glucose, and 1.5 g agar dissolved in 100 mL of water) at 28 • C, and after growing, stored at 4 • C.

Screening Biotransformation
Erlenmeyer flasks with a capacity of 300 mL containing 100 mL of Sabouraud medium, consisting of 3 g of glucose and 1 g of peptone dissolved in 100 mL of water, were used for biotransformation. Three days after inoculation of the medium with the given microorganism, 10 mg of substrate (halolactone 4-6) dissolved in 1 mL of acetone was added to each of the flasks with the mycelium growth. Shaking cultures were incubated with the substrate for the next 7 days. After 3, 5, and 7 days, samples were taken (about 30 mL each). The medium and mycelium were extracted with dichloromethane (15 mL), dried with magnesium sulfate, and analyzed on GC (DB-1 column). Each biotransformation was performed in two repetitions.

Preparative Biotransformation
For this, 100 mg of bromo-5 or iodolactone 6 dissolved in 10 mL of acetone was added to 10 Erlenmeyer flasks (300 mL) containing 3-day cultures of A. cylindrospora AM336 strain, prepared as described above. After 7 days of incubation of the shaking culture, the entire contents of 10 flasks were extracted with dichloromethane (3 × 40 mL). The combined organic fractions were dried with anhydrous magnesium sulphate. After evaporation of the solvent in vacuo, the products were purified on the chromatographic column. Spectral data of the obtained compounds are presented below.