Introduction

Abstract Introduction Results Discussion References

rac-Mexiletine is a class 1b antiarrhythmic drug, similar in structure to lidocaine, used in the treatment of acute and long-term ventricular arrhythmias. Unlike lidocaine (t_1/2 1.2-1.9 hr), mexiletine is an orally effective agent with a half-life ranging from 6 to 12 hr (1). When administered to healthy volunteers, it is almost completely absorbed (88% bioavailability), with peak plasma concentrations occurring within 2-4 hr (2). In coronary artery disease patients, these numbers are lower due to decreased gastric emptying (3). Mexiletine is a weakly basic drug (pK_a = 8.75) with a narrow therapeutic range. Serum concentrations must remain between 0.8 and 2.0 mg/liter, and levels >2.0 mg/liter can cause neurological side effects (4). Another feature that distinguishes mexiletine from lidocaine is that it possesses a chiral center and is administered as a racemate. Significant differences for the enantiomers have not been reported with respect to the absorption rate constant, peak plasma concentration, nor the time required to achieve peak plasma concentrations (5). However, there are reported differences in serum protein binding (6), receptor binding (7), electrophysiology (8), excretion (9), and metabolism (10) for the enantiomers.

A characteristic of rac-mexiletine therapy is the large interindividual variations in plasma concentrations after all modes of administration (11). One reported source of variation was urine pH. Mitchell et al. (12) demonstrated that physiological changes resulting in a decrease of urine pH from 8.0 to 5.0 produced a consistent increase in renal clearance. Other sources of variation included patient age (13), smoking (14), and use of additional medications, including cimetidine (15) and rifampicin (16). The predominant factor influencing plasma concentrations of rac-mexiletine is the liver's capacity to process the drug. The importance of metabolism is emphasized by reports that <10% of an oral dose is excreted unchanged in urine >3 days (17). The body generates numerous metabolites through various pathways, including oxidation, reduction, deamination, methylation, and conjugation. The principal metabolites (20% of an administered dose) are formed via oxidation of rac-mexiletine to PHM¹, HMM, and their corresponding alcohols, N-hydroxy-PHM and N-hydroxy-HMM (18). However, total urinary recovery of the drug, PHM, HMM, N-hydroxy-PHM, N-hydroxy-HMM, and their conjugated phase II metabolites only accounts for 30% of an administered dose. Turgeon et al. (19), recovered an additional 40% as two glucuronide metabolites formed after N-oxidation and deamination of rac-mexiletine. A summary of rac-mexiletine's metabolic profile in humans is outlined in fig. 1.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Summary of the phase I and II metabolites detected in human plasma and urine after administration of rac-mexiletine. ** Forms glucuronide conjugates.

Although the pathways involved in rac-mexiletine's metabolism were elucidated by 1988, the enzymes responsible were not. Broly et al. (20) investigated rac-mexiletine's metabolism in vitro using human liver microsomes and established that liver preparations were capable of oxidizing rac-mexiletine to HMM and PHM. Further studies revealed that a cytochrome P450 enzyme was responsible for biosynthesis, because metabolite production could be inhibited by heating the reaction mixture, adding carbon monoxide, and excluding NADPH. In addition, specific cytochrome P450 inhibitors (including SKF525-A, metyrapone, -naphthoflavone, and quinidine) were used to determine that the enzyme involved was cytochrome P450db1 (P450IID1 and bufI). This particular enzyme is also responsible for debrisoquine/sparteine-type polymorphism and is now known as CYP2D6 (21).

The role of debrisoquine/sparteine polymorphism in rac-mexiletine metabolism was examined in more detail by Turgeon et al. (22). Using 14 healthy volunteers (10 EMs and 4 PMs), they investigated whether low doses of quinidine, a CYP2D6 inhibitor, could repress this enzyme and turn EMs into PMs after oral administration of rac-mexiletine. When the drug was administered alone, mean plasma concentrations and urinary recovery of HMM and PHM were higher in PMs compared with EMs. However, there was no difference in formation of N-hydroxy metabolites. With coadministration of quinidine, the pharmacokinetics of rac-mexiletine were unchanged in PMs. In EMs, the pharmacokinetics were altered to the point where differences were not observed between EMs and PMs. On average, quinidine coadministration impaired oxidative carbon metabolism by 70%, but had no effect on nitrogen oxidation. These studies provided evidence for the role of CYP2D6 in rac-mexiletine metabolism in vivo.

One other aspect of rac-mexiletine administration is whether the enantiomers have different properties in vivo. The significance of this issue has been described with other chiral drugs (23). Studies with rac-mexiletine have demonstrated that the R-()-enantiomer preferentially binds serum proteins (24), cardiac sodium channels (5), and is taken up by rat cardiac tissue more significantly than S-(+)-mexiletine (7). In addition, it possesses greater antiarrhythmic activity (22) and forms more of the glucuronide conjugate (9).

Human liver microsomes have also been used to investigate stereoselective aspects of rac-mexiletine's metabolism in vitro (10). It was demonstrated that p-hydroxylation to PHM was the favored product from S-(+)-mexiletine, and aliphatic hydroxylation to HMM was preferred with R-()-mexiletine. It was also observed that biosynthesis of these metabolites was inhibited by quinidine and that dextromethorphan O-demethylation was competitively inhibited by S-(+)-mexiletine and R-()-mexiletine. This provided supplementary evidence for the role of CYP2D6 in stereoselective aspects of rac-mexiletine's oxidative metabolism.

These studies have provided detailed information regarding the pathways of rac-mexiletine metabolism in humans and the specific enzymes involved. This information can be used to evaluate the utility of microorganisms as an in vitro modeling system for oxidative biotransformations. Specifically, fungi can be useful tools for the study of drug metabolism in vitro because they possess a cytochrome P450 enzyme system analogous to that found in human liver (25). It has been demonstrated that fungi can generate a metabolic profile of the possible oxidative metabolites before studies involving human administration (26). It has also been shown that metabolite biosynthesis can be scaled up to yield semipreparative amounts of metabolite for structure elucidation (27).

The purpose of our study was to investigate whether the fungus Cunninghamella echinulata (UAMH 4145) was capable of metabolizing rac-mexiletine in a fashion similar to that reported in these studies. In addition, a stereospecific assay was used (28) to resolve simultaneously the enantiomers of mexiletine, HMM, and PHM, and thus determine whether metabolism was stereoselective. Finally, optimal conditions corresponding to maximum metabolite production were determined to facilitate scale-up of metabolite biosynthesis.

Materials and Methods

Chemicals and Reagents. R-()-mexiletine, S-(+)-mexiletine, and rac-mexiletine, HMM, and PHM were kindly donated by Boehringer Ingelheim Ltd. (Ontario, Canada). The internal standard used for high performance liquid chromatographic analysis was (±)-1-(4-hydroxyphenoxy)-3-isopropylamino-propan-2-ol (rac-prenalterol) and was synthesized in our laboratories. NEIC and n-butylamine (99+%) were obtained from Aldrich (Milwaukee, WI). Analytical grade sodium carbonate, sodium hydroxide, sodium phosphate (dibasic), sodium chloride, sodium acetate, HPLC-grade chloroform, and hexane were obtained from Fisher Scientific (Fair Lawn, NJ). Analytical grade D glucose, peptone, hydrochloric acid, maleic acid, acetic acid, methanol, and diethyl ether were obtained from BDH (Toronto, Ontario, Canada). Water was double-distilled and filtered using a Millipore Milli Q filtration system (Mississauga, Ontario, Canada). Bacto yeast extract, Bacto-agar, casamino acids, Czapec-Dox broth, malt extract broth, and Sabouraud dextrose broth were obtained from Difco Laboratories (Detroit, MI). Trypticase soy broth was purchased from Baltimore Biologicals Ltd. (BBL, Cockeysville, MD). C. echinulata (UAMH 4145, ATCC 9244) was obtained from UAMH (Edmonton, Alberta, Canada). The D-Loop bioreactor (300 ml) used in scale-up experiments was manufactured by Technical Services, University of Alberta (Edmonton, Alberta, Canada).

HPLC. HPLC analyses of rac-mexiletine, HMM, and PHM were conducted on a Waters system (Mississauga, Ontario, Canada) consisting of a model 590 pump, 712 WISP autosampler, and 470 scanning fluorescence detector set at 280 nm and 340 nm for excitation and emission as previously described (28). Briefly, enantiomer separation was conducted on a 250 × 4.6 mm (i.d.) stainless-steel Partisil 5 column (Phenomenex, Torrance, CA) using a mobile phase consisting of hexane:chloroform:methanol (65:34:1, v/v). Samples were prepared by adding sodium carbonate (0.100 ml of a 0.2 M solution) and diethyl ether (2.0 ml) to each diluted (1:80) fermentation sample. These mixtures were vortexed for 15 sec and centrifuged at 1,800 g for 4 min. The organic layer was removed and transferred to a clean glass test tube. The ether extraction step was repeated, and the combined extracts were evaporated to dryness under a gentle stream of nitrogen. The residues were reconstituted in chloroform and derivatized with NEIC at room temperature. These solutions were vortexed for 10 sec and again evaporated to dryness under a gentle stream of nitrogen. The residues were reconstituted with chloroform and reacted with n-butylamine to remove excess NEIC. The identity of mexiletine diastereomers was confirmed by comparing run times to similarly derivatized R-()-mexiletine and S-(+)-mexiletine standards. The identity of HMM and PHM diastereomers was tentatively identified by comparing run times of similarly derivatized HMM and PHM biosynthesized by C. echinulata from authentic R-()-mexiletine and S-(+)-mexiletine standards.

General Fermentation Procedure. Fermentations were performed in 125 ml Erlenmeyer culture flasks containing 25 ml medium adjusted to pH 7.0 unless stated otherwise. A model G-25 controlled environment incubator-shaker was used (New Brunswick Scientific Co., Inc.) equipped with 45° angle brackets operating at 250 rpm and at 27°C. The media tested are listed in table 1.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Concentration-time profile of rac-mexiletine, rac-HMM, and rac-PHM enantiomers in yeast extract broth and czapek dox broth. , S-(+)-mexiletine; , R-()-mexiletine; , S-HMM; , R-HMM; , S-PHM; , R-PHM.

	Results

Abstract Introduction Results Discussion References

Altering Fermentation Parameters. In our preliminary studies, it was observed that the medium used influenced the oxidative metabolic profile generated by C. echinulata. It was therefore decided to investigate this effect in greater detail. The results obtained from these experiments are summarized in table 2. In addition, previous biotransformation experiments were arbitrarily performed at pH 7.0. However, it was not known whether this was the optimal pH for metabolite biosynthesis. To determine optimal pH, yeast extract, trypticase soy, peptone, malt extract, and Sabouraud dextrose broth were prepared at pH 5.0, 5.5, 6.0, 6.5, 7.5, and 8.0. The results obtained from these experiments are summarized in table 3. Finally, authentic samples of pure S-(+) and R-()-mexiletine were individually metabolized in yeast extract and trypticase soy broth to determine how the single enantiomers were metabolized compared with the racemate. The results obtained are summarized in table 4.

View this table: [in this window] [in a new window]   TABLE 4 Metabolism of authentic R-() and S-(+)-mexiletine standards Summary of data obtained from the metabolism of authentic S-(+)-mexiletine and R-()-mexiletine standards in yeast extract broth and trypticase soy broth.

	Discussion

Abstract Introduction Results Discussion References

C. echinulata was capable of stereoselectively biotransforming rac-mexiletine into the major human metabolites HMM and PHM. In general, the quantity of metabolite biosynthesized was strongly influenced by the medium used and its pH. Overall, R-() and S-(+)-mexiletine enantiomers were not metabolized equally. R-()-mexiletine was preferentially metabolized, whereas R-PHM was preferentially biosynthesized. The preference for R-()-mexiletine is similar to that seen in human studies. However, this similarity was not observed with HMM biosynthesis in which larger amounts of S-HMM were produced in microbial studies.

It has been reported that the isozyme responsible for the production of HMM and PHM in humans was CYP2D6. To determine whether C. echinulata has a similar enzyme, Foster et al. (29) conducted an experiment to evaluate whether sparteine and quinidine decreased the rate and extent of methoxyphenamine (a human CYP2D6 substrate) biotransformation by the same fungus. They were able to demonstrate inhibition suggesting the fungal enzyme responsible for metabolite biosynthesis may be similar to CYP2D6. This observation correlates with our observation that C. echinulata can produce HMM and PHM.

Our first series of experiments determined that the media used had a dramatic effect on metabolism. For example, yeast extract broth proved to be the preferred medium for maximum HMM and PHM biosynthesis (fig. 2). Under these conditions, 21% more HMM and 25% more PHM were produced, compared with Czapek Dox broth. In addition, under these growth conditions, R-()-mexiletine was preferentially metabolized, whereas R-PHM was preferentially biosynthesized. However, this was not the case with HMM biosynthesis, wherein S-HMM was produced in excess of R-HMM. This pattern was evident with all media evaluated.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Concentration-time profile of rac-mexiletine, rac-HMM, and rac-PHM enantiomers in trypticase soy broth and phosphate buffer. , S-(+)-mexiletine; , R-()-mexiletine; , S-HMM; , R-HMM; , S-PHM; , R-PHM.

The use of trypticase soy broth revealed information about the effect of growth medium on metabolite biosynthesis. When the utilization of trypticase soy broth was compared with phosphate buffer for biotransformation, stereoselective production of HMM and PHM were detected in each environment (fig. 3). However, there was 2.1 times more drug remaining, 1.6 times more PHM, and 1.5 times more HMM produced in the trypticase soy broth medium than in phosphate buffer. It is evident that a medium that supports fungal growth will augment metabolite biosynthesis.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Concentration-time profile of rac-mexiletine, HMM, and PHM in yeast extract broth at pH 5.5 and 6.0.

Another two media that provided high yields of HMM and PHM were casamino acids broth and peptone broth. A distinguishing feature of their use was that 200 hr were required to reach maximum levels of metabolite. When compared with 83 hr for yeast extract broth or 21 hr for trypticase soy broth, it seems that C. echinulata was metabolically active for a longer period when casamino acid broth and peptone broth were used. However, although the amount of metabolites produced were high, it would be convenient to obtain similar results in a shorter period of time.

Experiments in which pH was varied revealed that when yeast extract broth was used, metabolite biosynthesis increased with increasing pH. However, at pH 8.0, metabolites were not formed. In addition, the data indicate that a pH of 7.0 provided the largest concentration of metabolites. This study also demonstrated that changes in pH did not affect preference for R-mexiletine nor the product R-PHM. However, a subtle reversal was observed with HMM production in all incubates. As pH increased during incubation, a preference for R-HMM production was changed into one for S-HMM.

Another trend observed with cultures grown in yeast extract broth was that the measured pH was similar in all incubates when the maximum concentration of metabolite was detected. In addition, increases in pH beyond 7.6 caused a decrease in metabolite concentration. The effect of pH on metabolite biosynthesis can be seen in fig. 4. With the acidic yeast extract broth preparations of pH 5.5 and 6.0, a decrease in medium pH was observed over the first 15 hr, followed by a gradual increase to pH 8.0. During fermentation, the initial decrease in pH is generally attributed to acidic byproducts, thus resulting from carbohydrate metabolism. After the carbohydrate has been depleted, there is a subsequent increase in pH due to the utilization of other medium components for cell growth. Metabolite biosynthesis was observed to coincide with this increase in medium pH.

A cessation of metabolite biosynthesis with basic pH was also observed with trypticase soy broth. In general, as pH increased from 5.0 to 6.5, the concentration of metabolites increased to 12% for HMM and 11% for PHM. This was in agreement with the previous observation that C. echinulata preferred a pH of ~7.0 for optimal metabolite production. In addition, the appearance of metabolites was preceded by an increase in pH.

The pH values for the malt extract and Sabouraud dextrose formulations, when prepared according to the supplier's instructions, was 4.7 and 5.7 respectively. This difference from pH 7.0 had a significant impact on metabolite biosynthesis. When malt extract broth was used at pH 4.7, metabolites were not detected. When Sabouraud dextrose broth was used at pH 5.7, there was a significant decrease in HMM and PHM biosynthesis when compared with pH 7.0. These data support the observation that C. echinulata prefers a neutral pH for optimal production of metabolites, although generally low pH values are optimal for fungal growth.

The changes in pH also had a consistent effect on fungal growth for all media evaluated. As pH was increased from 5.5 to 8.0, fungal morphology changed from small, dense, individual pellets (pH 5.5) to diffuse mycelia (pH 8.0). This change in appearance occurred in small increments over the pH range under study, with the individual pellets slowly congealing until a single mass of widely separated mycelia was visible.

When bioconversion of rac-mexiletine was examined in detail, the combined average recovery of drug and metabolites was calculated to be between 45 and 70%, depending on the media used (i.e. a significant percentage of added drug could not be accounted for). The fungal cytoplasm was a possible location for the unaccounted material. This possibility was evaluated by grinding the mycelia, followed by solvent extraction of the fragments. The maximum amounts of compounds detected by HPLC were 6.7 µg/ml of S-(+)-mexiletine + R-()-mexiletine and 0.9 µg/ml of R + S-PHM. The unaccounted mexiletine substrate was either incorporated into cellular material or eliminated as other metabolite(s) not detected by this assay. Our analytical procedure was designed to detect compounds possessing a primary or secondary amino group. The possibility exists that neutral or acidic metabolites are also being biosynthesized.

The previously described experiments demonstrated that C. echinulata preferentially metabolized R-()-mexiletine and biosynthesized R-PHM and S-HMM. This provided a basis from which a comparison could be made between the metabolism of rac-mexiletine and S-(+) and R-() standards. When using trypticase soy broth, both the enantiomers were completely metabolized within 36 hr. In addition, while a low concentration of HMM was produced when the individual enantiomers were added, large increases in both S-PHM (2.2-fold) and R-PHM (2.5-fold) were observed. When yeast extract broth was used, a different pattern of metabolism was evident. A larger concentration of drug (compared with racemate) remained in the medium at the final sampling time (75 hr). In addition, as observed with trypticase soy broth, small amounts of HMM were produced in association with higher levels of PHM. This suggests that when individual enantiomers are metabolized, aromatic hydroxylation is favored over benzylic. There were, however, some similarities when comparing metabolism of rac-mexiletine with that of individual enantiomers. R-()-mexiletine was the preferred enantiomer metabolized, and R-PHM was the favored enantiomer formed. Thus, as with human studies, R-()-mexiletine was preferentially oxidized by cytochrome P450.

The conditions chosen to scale-up metabolite production were based on previously obtained data that suggested that yeast extract broth at pH 7.0 would provide optimal results. Unfortunately, the concentrations of metabolites in the D-Loop fermentor were less than the amount detected in Erlenmeyer flasks. The calculated yields were 3.0% R + S-HMM (1.7 mg/300 ml) and 8.5% R + S-PHM (6.6 mg/300 ml). An explanation for the lower yields can be attributed to the way C. echinulata grew in the D-Loop fermentor. The inoculum of 25.6 g (wet weight), in combination with the rich growth medium, provided considerable biomass in a matter of hours. Fungal growth occurred so quickly that it plugged the fermentor, thus preventing medium circulation.

C. echinulata was capable of stereoselectively biotransforming rac-mexiletine into the major human metabolites HMM and PHM. In general, the quantity of metabolite biosynthesized was strongly influenced by the medium used and its pH. Yeast extract broth at an initial pH of 7.0 provided the optimal amounts of oxidative metabolites. Under these conditions, 30.1 µg/ml of R + S-HMM and 64.6 µg/ml of R + S-PHM were biosynthesized after 83 hr of incubation. In total, the yield of HMM and PHM from rac-mexiletine was 48%.