Transformation of Verapamil by Cunninghamella blakesleeana

Abstract

A filamentous fungus, Cunninghamella blakesleeana AS 3.153, was used as a microbial model of mammalian metabolism to transform verapamil, a calcium channel antagonist. The metabolites of verapamil were separated and assayed by the liquid chromatography-ion trap mass spectrometry method. After 96 h of incubation, nearly 93% of the original drug was metabolized to 23 metabolites. Five major metabolites were isolated by semipreparative high-performance liquid chromatography and were identified by proton nuclear magnetic resonance and electrospray mass spectrometry. Other metabolites were characterized according to their chromatographic behavior and mass spectral data. The major metabolic pathways of verapamil transformation by the fungus were N dealkylation, O demethylation, and sulfate conjugation. The phase I metabolites of verapamil (introduction of a functional group) by C. blakesleeana paralleled those in mammals; therefore, C. blakesleeana could be a useful tool for generating the mammalian phase I metabolites of verapamil.

Verapamil, α-[3-[[2-(3,4-dimethoxyphenyl)ethyl]methylamino]propyl]-3,4-dimethoxy-α-(1-methylethyl)-benzeneacetonitrile, is a calcium channel antagonist widely used in the treatment of supraventricular arrhythmias, coronary heart disease, and hypertension (18). The metabolism of verapamil is known in humans and animals (6, 9, 16), where it is extensively metabolized. After oral administration, less than 5% of the dose is excreted as the unchanged drug in the urine; 12 metabolites (including six trace metabolites) have been identified in urine and feces (6, 16). N demethylation and N dealkylation are the main metabolic pathways of verapamil degradation. Verapamil and its N-dealkylated metabolites are further metabolized by O demethylation. Most N-dealkylated metabolites of verapamil have no effect on vasodilation, but norverapamil possesses about 20% of the vasodilating activity of verapamil (8). O-demethylated metabolites of verapamil have the same potency as the parent drug, but their contribution to the overall pharmacological effect is negligible, since these metabolites usually are present as inactive conjugates (8, 14).

Some microorganisms can transform drugs and other xenobiotic compounds in a manner similar to that in mammals, and the utilization of microbial systems as models for mimicking and predicting the metabolism of drugs in humans and animals has received considerable attention (3, 5, 10, 22). Many mammalian phase I metabolic reactions (introduction of a functional group) and phase II metabolic reactions (conjugation with endogenous compounds), including hydroxylation, O and N dealkylation, dehydrogenation, and glucuronide and sulfate conjugation, also occur in microbial models (1). Microbial models have a number of advantages over studies with animals and humans, due to reduced animal use, ease of setup and manipulation, higher yield and diversity of metabolite production, and lower cost of production. In some cases microbial models can provide sufficient amounts of putative metabolites for complete structural elucidation. Metabolite production is especially important for metabolites that are not easily synthesized by organic chemical methods. A major goal of studies on these microbial models is to develop a list of microorganisms that can mimic the metabolic profiles of drugs in mammals. Filamentous fungi from the genus Cunninghamella, particularly Cunninghamella elegans, often metabolize drugs in a manner similar to that in mammals (2, 3, 7, 10, 11, 15, 22).

Although the metabolism of verapamil in mammals has been studied, little is known about the potential of nonligninolytic fungi to metabolize verapmil. The objective of this paper is to identify a fungus that can extensively metabolize verapamil and to chemically identify the metabolites. If the metabolic profile of verapamil by this microorganism is similar to that in mammals to some extent, then the fungus could be used as a model in studies of mammalian drug metabolism.

MATERIALS AND METHODS

Chemicals.

Verapamil hydrochloride was supplied by Zhongyang Pharmaceutical Factory (Tianjin, China). Methanol and acetonitrile were of high-performance liquid chromatography (HPLC) grade. Peptone and yeast extract were obtained from Aoboxing Co. (Beijing, China). All other chemicals were of analytical grade.

Microorganisms.

C. blakesleeana AS 3.153, C. echinulata AS 3.2004, C. elegans AS 3.156, and AS 3.2028 were provided by the Institute of Microbiology, the Academy of Sciences of China (Beijing, China). Stock cultures were maintained on potato dextrose agar (Aoboxing Co.) slants at 4°C and were transferred every 6 months.

Fermentation conditions.

Microbial fermentation was carried out in a liquid broth containing (per liter) glucose (20 g), peptone (5.0 g), yeast extract (5.0 g), K₂HPO₄ (5.0 g), and NaCl (5.0 g). The broth was adjusted to pH 6.5 with 6.0 M HCl, autoclaved in Erlenmeyer flasks at 115°C for 30 min, and cooled before inoculation. Transformation of verapamil was performed by incubating cultures on a rotary shaker at 220 rpm and 28°C according to a two-stage fermentation procedure (1).

For each of four strains of Cunninghamella, the first-stage fermentation was initiated by inoculating a 250-ml Erlenmeyer flask containing 50 ml of broth with a loop of spores obtained from a freshly growing agar slant. After incubation for 24 h, a 1.0-ml portion (about 200 cells) from the first-stage culture was used to inoculate a second-stage 100-ml flask containing 20 ml of broth. The second-stage culture was incubated for 24 h, before verapamil in sterile water was added to a final concentration of 500 μg/ml. After 96 h of additional incubation, the culture was centrifuged at 1,500 × g for 20 min and the supernatant was decanted and stored at −20°C until analysis.

Preparative-scale fermentation of verapamil by C. blakesleeana followed the same procedure as the screening experiments, except that the volume of broth was increased. Two first-stage flasks were generated as described above, and five second-stage flasks of 1-liter capacity containing 200 ml of broth were each inoculated with 10 ml of the first-stage culture. Verapamil was added to a concentration of 500 μg/ml, and then the culture was incubated for an additional 96 h.

Two kinds of controls were run simultaneously with the fermentation and were analyzed in the same way. Culture controls consisted of a fermentation blank in which strains of Cunninghamella were grown under identical conditions without verapamil. The substrate control consisted of verapamil and sterile broth incubated without fungal inoculum to determine whether verapamil could chemically decompose or spontaneously transform under fermentation conditions.

Extraction of metabolites and liquid chromatography-mass spectrometry (LC/MSⁿ) assay.

A 100-μl aliquot of each sample or control was diluted with 900 μl of distilled water and was filtered through a membrane (0.45-μm pore size). The filtrate was applied to a preconditioned Bond Elute C₁₈ cartridge (Teda Fuji Co., Tianjin, China). The column was washed with water, and the metabolites were eluted with methanol (2.0 ml). The methanol solution was evaporated to dryness under a gentle stream of nitrogen at 40°C, and the residue was dissolved in a 1.0-ml mobile phase (acetonitrile:water:formic acid, 30:70:1 [vol/vol/vol]). A small aliquot of the solution (20 μl) was injected into the chromatographic system. The transformation yield was determined according to the total signal abundance of various metabolites relative to the sum of metabolites and verapmil.

LC/MSⁿ analysis was performed on a Finnigan LCQ ion trap mass spectrometer (San Jose, Calif.) equipped with an atmospheric pressure ionization interface. The instrument was operated in positive electrospray ionization (ESI) mode. The capillary voltage was fixed at 16 V, and its temperature was maintained at 200°C. The spray voltage was set at 4.25 kV. The HPLC fluid was nebulized by using N₂ as both the sheath gas at a flow rate of 0.75 liter/min and the auxiliary gas at a flow rate of 0.15 liter/min. Multistage mass spectra (MS/MS or MS³) were produced by collision-induced dissociation of the selected precursor ions with He present in the ion trap, and the relative collision energy was set at 30 to 40%. Data were collected and analyzed with the Navigator software (version 1.2; Finnigan). Liquid chromatography was carried out with a Shimadzu LC-10AD solvent delivery system (Kyoto, Japan). The samples were separated on a Diamonsil C₁₈ column (particle size, 5 μm; inside diameter, 4.6 by 200 mm; Dikma Co., Beijing, China). The mobile phase consisted of acetonitrile:water:formic acid (30:70:1 [vol/vol/vol]) at a flow rate of 0.3 ml/min.

Isolation and identification of major metabolites.

To isolate the major metabolites of verapamil in sufficient quantities for structural elucidation, the transformation of verapamil by C. blakesleeana was carried out on a preparative scale. After fermentation, the flask contents were centrifuged at 1,500 × g for 20 min. The combined supernatant was adjusted to pH 10 with 2 M NaOH and was extracted with 3 equal volumes of diethyl ether. The organic layer was dried over sodium sulfate and was evaporated to dryness in vacuum. The residue dissolved in methanol was subjected to separation.

The semipreparative HPLC was performed with a Hewlett-Packard 1100 system (Palo Alto, Calif.) consisting of a quaternary pump, a vacuum degasser, a variable wavelength detector, and an autosampler. Separation was achieved by using an Inertsil C₁₈ column (particle size, 10 μm; inside diameter, 10.7 by 250 mm; Gasukuro Kogyo, Tokyo, Japan). The mobile phase consisted of a linear 48 to 65% methanol gradient in 50 mM NH₄H₂PO₄ for 50 min at a flow rate of 3 ml/min. The UV detection wavelength was set at 276 nm. To isolate the metabolites, the methanol solution of the extract was injected repeatedly and major peaks with the same retention time were pooled. The methanol in the combined HPLC eluate was evaporated in vacuum; then the metabolites were extracted from the remaining solution by diethyl ether as described above and were stored at −20°C until structural elucidation.

The purified metabolites and verapamil were dissolved in CDCl₃ for nuclear magnetic resonance (NMR) analysis. The ¹H NMR measurements were carried out at 300 MHz on a Bruker ARX 300 NMR spectrometer (Faellanden, Switzerland). Chemical shifts were reported as parts per million relative to tetramethylsilane as the internal standard.

RESULTS

Screening of cultures.

Four strains of Cunninghamella could all transform verapamil to some extent. The frequently used C. elegans AS 3.156 and AS 3.2028 transformed 10 and 12% of the verapamil, respectively, whereas C. blakesleeana AS 3.153 gave the best yield at 93%, and C. echinulata AS 3.2004 was intermediate at 26%. Since C. blakesleeana transformed the highest proportion of the verapamil, it was selected for further investigation.

Identification of verapamil metabolites.

LC/MSⁿ chromatograms of the culture control showed that secondary metabolites produced by C. blakesleeana did not affect the analysis of verapamil and its metabolites. The substrate control contained only verapamil. Twenty-three metabolites were found in the C. blakesleeana cultures, in addition to verapamil (Fig. 1), that were not found in the control cultures.

FIG. 1.

Total ion current (TIC) and selected ion monitoring (SIM) chromatograms of verapamil and its metabolites in the sample of C. blakesleeana incubated with 500 μg of verapamil/ml for 96 h. (NL is a shortened form of normalized y-axis normalization mode, in which the instrument automatically sets the vertical scale equal to the height of the largest peak in an MS.)

Following preparative-scale transformation of verapamil by C. blakesleeana, five major metabolites (M1, M2, M4, M5, and M6) were isolated by semipreparative HPLC and their structures were determined by ¹H NMR (Table 1) and ESI mass spectrometry (Table 2). The other metabolites (Table 3) cannot be separated completely with this method, and no further purification was conducted, since their structures could be identified based on their MS and retention time (Table 2).

TABLE 1.

¹H NMR parameters of verapamil and five metabolites purified from the preparative-scale fermentation of verapamil by C. blakesleeana

Proton	Parameter of:
	Verapamil	M1	M2	M4	M5	M6
H-2	6.68	6.67	6.65	6.72		6.76
H-3	6.79	6.78	6.82	6.79		6.86
H-6	6.70	6.69	6.67	6.71		6.80
H-7	2.68	2.67	2.67	2.80		2.86
H-8	2.52	2.52	2.51	2.75		2.79
H-9				2.11	1.65	2.50
H-10	2.38	2.37	2.37	2.61	2.56
H-11a	1.19	1.19	1.19	1.19	1.19
H-11b	1.59	1.55	1.57	1.56	1.57
H-12a	1.85	1.82	1.84	1.86	1.89
H-12b	2.11	2.09	2.10	2.15	2.17
H-15	6.91	6.88	6.91	6.90	6.92
H-16	6.83	6.81	6.82	6.83	6.83
H-19	6.86	6.86	6.86	6.85	6.85
CH3-21	3.85	3.85		3.85		3.88
CH3-23	3.87	3.86	3.87	3.86		3.90
CH3-24	2.20	2.20	2.19		2.36	2.47
H-25	2.08	2.06	2.05	2.07	2.08
CH3-26	0.79	0.79	0.79	0.79	0.79
CH3-27	1.18	1.17	1.18	1.18	1.19
CH3-31	3.88		3.88	3.88	3.88
CH3-33	3.89	3.88	3.89	3.89	3.89

TABLE 2.

LC/MSⁿ data of verapamil and its metabolites produced by incubation with C. blakesleeana

Metabolite	Retention time (min)	MS [M+H]+	m/z (relative abundance, %) of:
			MS2 fragment ion	MS3 fragment ion
Verapamil	46.6	455	303 (95), 260 (8), 165 (100), 150 (12)
M1	27.2	441	289 (70), 246 (5), 165 (100), 150 (14)
M2	29.9	441	303 (19), 291 (100), 260 (6), 151 (9)
M3	34.4	441	303 (100), 291 (46), 260 (11), 151 (20)
M4	42.2	441	398 (18), 289 (9), 260 (4), 165 (100), 150 (14)
M5	13.1	291	260 (78), 248 (100)
M6	6.1	196	165 (100)
M7	18.6	427	289 (18), 277 (100), 246 (6), 151 (14)
M8	20.9	427	289 (100), 277 (44), 246 (6), 151 (28)
M9	25.3	427	384 (14), 275 (12) 246 (4), 165 (100), 150 (18)
M10	28.6	127	384 (10), 289 (6), 277 (100), 260 (14), 151 (36)
M11	9.6	277	246 (100), 234 (91)
M12	12.2	277	260 (100), 234 (78)
M13	29.7	521	441 (100)	303 (100), 291 (46), 260 (11), 151 (19)
M14	40.1	521	441 (100)	289 (70), 246 (5), 165 (100), 150 (15)
M15	48.2	521	441 (100)	303 (19), 291 (100), 260 (6), 151 (9)
M16	19.0	507	427 (100)	384 (14), 275 (11) 246 (4), 165 (100), 150 (18)
M17	26.8	507	427 (100)	289 (18), 277 (100), 246 (6), 151 (14)
M18	29.8	507	427 (100)	289 (18), 277 (100), 246 (6), 151 (15)
M19	42.7	507	427 (100)	275 (51), 232 (3), 165 (100), 150 (14)
M20	49.2	507	427 (100)	384 (10), 289 (6), 277 (100), 260 (15), 151 (36)
M21	52.6	507	427 (100)	275 (50), 232 (3), 165 (100), 150 (14)
M22	53.5	507	427 (100)	303 (42), 291 (100), 260 (9), 137 (4)
M23	61.9	507	427 (100)	303 (42), 291 (100), 260 (9), 137 (4)

TABLE 3.

The chemical structures of verapamil and its metabolites produced by incubation with C. blakesleeana

^aThe derivative of N dealkylation of the phenylethyl moiety from verapamil.

^bThe derivative of N dealkylation of the phenylacetonitrile moiety from verapamil.

^c—, the metabolite is being described here for the first time.

Parent drug.

The compound eluting at 46.6 min possessed the same pseudomolecular ion, full-scan MS/MS, and chromatographic behavior as authentic verapamil.

Metabolites M1, M2, M3, and M4.

The retention times of M1, M2, M3, and M4 were 27.2, 29.9, 34.4, and 42.2 min, respectively. They all had pseudomolecular ions [M+H]⁺ at m/z 441 in the full-scan MS but had different MS/MS, indicating that they were isomers. Their pseudomolecular ions were 14 Da lower than that of the parent drug, suggesting the loss of a methyl group from verapamil.

The MS/MS of M1 had fragment ions at m/z 289, 246, 165, and 150, which indicated that a methyl group had been lost from the phenylacetonitrile moiety. The ¹H NMR data for M1 differed from those of verapamil in the signal for a methoxy group in the phenylacetonitrile moiety. Two-dimensional nuclear Overhauser enhancement (NOE) spectroscopy was used to determine whether the O demethylation occurred meta or para to the substituted alkyl. There were NOE cross-peaks between the signal of the remaining methoxy protons in the phenylacetonitrile moiety and the signal of H-19 in the NOE spectroscopy spectrum of M1, suggesting that these protons are close to each other in space. Thus, the O demethylation in M1 occurred para to the substituted alkyl and M1 was 30-O-demethyl-verapamil.

The MS/MS of M2 and M3 had the same fragment ions at m/z 303, 291, 260, and 151, which suggested the loss of a methyl group in the phenylethyl moiety. The relative abundance of the fragment ions at m/z 303 and 291 differed in M2 and M3, indicating that O demethylation in the phenylethyl moiety occurred at different positions in M2 and M3. Compared with verapamil, the signal of one group of methoxy protons in the phenylethyl moiety disappeared in the NMR spectrum of M2. The remaining protons were close in space to H-6. Thus, the O demethylation in M2 was para to the substituted alkyl in the phenylethyl moiety, and M2 was 20-O-demethyl-verapamil, whereas M3 was identified as 22-O-demethyl-verapamil.

The MS/MS of M4 had fragment ions at m/z 398, 289, 260, 165, and 150, suggesting that the four O-methyl moieties were unaltered and that oxidative N demethylation occurred. The lack of N-methyl protons in the ¹H NMR data for M4 supported N demethylation. On the basis of these data, M4 was confirmed as norverapamil.

Metabolite M5.

M5 with a retention time of 13.1 min had a pseudomolecular ion [M+H]⁺ at m/z 291 in the full-scan MS, consistent with N dealkylation of the phenylethyl moiety (164 Da) of verapamil. The MS/MS of M5 had two fragment ions at m/z 260 and 248. The ¹H NMR spectrum of M5 also lacked the signals belonging to the phenylethyl moiety but had a new proton signal at 1.65 ppm, assigned to a hydrogen at the 9-N position. Based on these data, we concluded that M5 was N-methyl-4-(3,4-dimethoxy-phenyl)-4-cyano-5-methylhexylamine.

Metabolite M6.

The retention time of M6 was 6.1 min. It had a pseudomolecular ion [M+H]⁺ at m/z 196 in the full-scan MS, which was 259 Da lower than that of verapamil, suggesting the cleavage of the phenylacetonitrile moiety from the parent drug. In the ¹H NMR spectrum of M6, the signals for the phenylacetonitrile moiety were absent and a new signal at 2.50 ppm was present, which was indicative of a proton connected to a nitrogen atom. Therefore, M6 was confirmed as N-methyl-2-(3,4-dimethoxy-phenyl)ethylamine.

Metabolites M7, M8, M9, and M10.

The retention times of M7, M8, M9, and M10 were 18.6, 20.9, 25.3, and 28.6 min, respectively. They all had the same pseudomolecular ions [M+H]⁺ at m/z 427 in the full-scan MS, but each had a different MS/MS, indicating that they were isomers. The pseudomolecular ions were 28 Da lower than that of the parent drug, suggesting the loss of two methyl groups from verapamil.

The MS/MS of M7 and M8 had identical fragment ions at m/z 289, 277, 246, and 151. The diagnostic ions at m/z 289 and 246 suggested the loss of a methyl group from the phenylacetonitrile moiety. The lost methyl group from the phenylacetonitrile moiety in M7 and M8 was at the 30-O position. The fragment ion at m/z 151 indicated that the other demethylation occurred in the phenylethyl moiety. The fragment ions at m/z 289 and 277 were 14 Da lower than those in the MS/MS of M2 and M3, suggesting that M7 and M8 were demethylated derivatives of M3 and M4, respectively. The relative abundance of fragment ions at m/z 303 and 291 in the MS/MS of M3 and M4 was determined by the positions of the O demethylation; i.e., 20-O-demethyl-verapamil had the base fragment ion at m/z 291, and 22-O-demethyl-verapamil had the base fragment ion at m/z 303. Therefore, the prominent fragment ion at m/z 277 indicated that O demethylation in the phenylethyl moiety in M7 was para to the substituted alkyl, whereas the base fragment ion at m/z 289 in the MS/MS of M8 suggested that O demethylation in the phenylethyl moiety occurred meta to the substituted alkyl. On the basis of these data, M7 was characterized as 20,30-O-didemethyl-verapamil and M8 as 22,30-O-didemethyl-verapamil.

The MS/MS of M9 had fragment ions at m/z 384, 275, 246, 165, and 150. The diagnostic ion at m/z 246 suggested that one methyl group had been lost from the phenylacetonitrile moiety. The presence of fragment ions at m/z 275 and 384 suggested that the other demethylation occurred in the N-methyl moiety. Based on these data, M9 was characterized as 30-O-demethyl-N-demethyl-verapamil.

The MS/MS of M10 had fragment ions at m/z 384, 289, 277, 260, and 151. The diagnostic ion at m/z 151 suggested that the loss of one methyl group occurred in the phenylethyl moiety, whereas the relative abundance of fragment ions at m/z 289 and 277 indicated that the demethylation occurred at the 20-O-methyl moiety. The presence of the fragment ions at m/z 289 and 384 suggested that the other methyl group was lost from the N-methyl moiety. Thus, M10 was characterized as 20-O-demethyl-N-demethyl-verapamil.

Metabolites M11 and M12.

The retention times of M11 and M12 were 9.6 and 12.2 min, respectively. They both had pseudomolecular ions [M+H]⁺ at m/z 277 in the full-scan MS but had different MS/MS fragment ions, indicating that they were isomers. Their pseudomolecular ions were 14 Da lower than that of M5, suggesting that they were further demethylated derivatives of M5. M11 had fragment ions at m/z 246 and 234 in the MS/MS, whereas M12 had fragment ions at m/z 260 and 234. The diagnostic ion at m/z 246 suggested that a methyl group was lost from the phenylacetonitrile moiety, so M11 was characterized as N-methyl-4-(3-methoxy-4-hydroxyphenyl)-4-cyano-5-methylhexylamine. The fragment ion at m/z 260 indicated that the phenylacetonitrile moiety was unaltered and that the demethylation was carried out in the N-methyl moiety. Based on these data, M12 was characterized as 4-(3,4-dimethoxyphenyl)-4-cyano-5-methylhexylamine.

Metabolites M13, M14, and M15.

The retention times of M13, M14, and M15 were 29.7, 40.1, and 48.2 min, respectively. They displayed the same pseudomolecular ions [M+H]⁺ at m/z 521 in the full-scan MS, and all yielded only one fragment ion at m/z 441 in the MS/MS, but each gave different MS³, indicating that they were isomers. Their pseudomolecular ions [M+H]⁺ at m/z 521, a mass shift of 66 Da compared with verapamil, suggested that they were sulfate conjugates of demethylated derivatives of verapamil. The base fragment ions at m/z 441, corresponding to the protonated aglycone, were formed by the loss of 80 Da from the precursor ions, which also was indicative of sulfate conjugates (17, 19). On the basis of these spectral data, the three metabolites appeared to be sulfate conjugates of demethylated derivatives of verapamil. The MS³ of M13 was identical with the MS/MS of M3, suggesting that M13 was 22-O-demethyl-verapamil-22-sulfate. M14 yielded the same MS³ as the MS/MS of M1, indicating that M14 was 30-O-demethyl-verapamil-30-sulfate. The MS³ of M15 gave the same fragment ions as those in the MS/MS of M2, so it was characterized as 20-O-demethyl-verapamil-20-sulfate.

Metabolites M16 to M23.

M16 to M23 all had pseudomolecular ions [M+H]⁺ at m/z 507 in the full-scan MS, and all had a single fragment ion at m/z 427 in the MS/MS, but each displayed a different MS³, indicating that they were isomers. The pseudomolecular ions [M+H]⁺ at m/z 507 were a mass shift of 80 Da compared with didemethylated derivatives of verapamil. The fragment ions at m/z 427 possessing the same mass-to-charge ratio with the protonated aglycone were produced by the loss of 80 Da (a characteristic loss of sulfate conjugates) from the precursor ions, suggesting that they appeared to be all sulfate conjugates of didemethylated derivatives of verapamil.

M16 had the same MS³ as the MS/MS of M9, so it was characterized as 30-O-demethyl-N-demethyl-verapamil-30-sulfate. The MS³ of M17 and M18 were identical and gave the same fragment ions as those in the MS/MS of M7, suggesting that M17 and M18 were both sulfate conjugates of 20,30-O-didemethyl-verapamil. Because the retention times of the two metabolites were different, they were the isomers with different positions of sulfate conjugation, 20,30-O-didemethyl-verapamil-20-sulfate and 20,30-O-didemethyl-verapamil-30-sulfate, respectively. These compounds were not further differentiated.

M19 and M21 had the same MS³ with fragment ions at m/z 275, 232, 165, and 150. The presence of fragment ions at m/z 165 and 150 suggested that the phenylethyl moiety was unaltered. The diagnostic ions at m/z 275 and 232 were 28 Da lower than those fragment ions in the MS/MS of verapamil, which suggested that two methyl groups were lost from the phenylacetonitrile moiety. The difference between M19 and M21 is in the position of sulfate conjugation. They were 30,32-O-didemethyl-verapamil-30-sulfate and 30,32-O-didemethyl-verapamil-32-sulfate, respectively, but were not further differentiated. The MS³ of M20 was identical with the MS/MS of M10, suggesting that M20 was 20-O-demethyl-N-demethyl-verapamil-20-sulfate.

The MS³ of M22 and M23 were identical and had fragment ions at m/z 303, 291, 260, and 137. The presence of the fragment ions at m/z 303 and 260 suggested that the phenylacetonitrile moiety and the N-methyl moiety were unaltered. The diagnostic ion at m/z 137 was 28 Da lower than that in the MS/MS of verapamil, which suggested that the two methyl groups were lost from the phenylethyl moiety. Based on these data, M22 and M23 differ in the position of sulfate conjugation and were characterized as 20,22-O-didemethyl-verapamil-20-sulfate and 20,22-O-didemethyl-verapamil-22-sulfate, respectively, but were not further differentiated.

DISCUSSION

The metabolism of verapamil is established in humans and animals (6, 9, 16), but this is the first report of its metabolism by microorganisms. Twenty-three metabolites, 12 phase I metabolites and 11 phase II metabolites, were produced by C. blakesleeana from verapamil. The major metabolic pathways (Fig. 2) of verapamil by C. blakesleeana were N dealkylation, O demethylation, and sulfate conjugation.

FIG. 2.

Proposed metabolic profile of verapamil by C. blakesleeana.

The concept of using microorganisms, and in particular the fungi belonging to the genus Cunninghamella, as models for mammalian metabolism has been well documented (1, 7, 11, 15). The fungi are highly efficient in their production of metabolites from a wide variety of xenobiotics, and direct parallels have been observed between fungal metabolism and mammalian metabolism (11, 15). In our study, of the 12 phase I metabolites of verapamil by C. blakesleeana, 10 were known mammalian metabolites. The results are consistent with the hypothesis that C. blakesleeana can mimic mammalian phase I metabolism of drugs. Previous studies of verapamil metabolism found that conjugated metabolites occurred following β-glucuronidase hydrolysis (6, 9, 16). Because β-glucuronidase possesses both glucuronidase and sulfatase activities, the conjugated metabolites were not differentiated as glucuronides or sulfate conjugates. The 11 conjugated metabolites produced by C. blakesleeana that we characterized were all sulfate conjugates. Though previous studies with microbial models indicated that these organisms could produce glucuronide conjugates of drugs (2, 4), we did not find any such metabolites. This result is consistent with our previous study of naproxen metabolism, in which only sulfate conjugates were produced by C. blakesleeana (23).

The regiochemistry of N dealkylation of verapamil by C. blakesleeana is consistent with previous reports of verapamil metabolism in rat and human liver microsomes (12). Among the three pathways of C-N cleavage of verapamil, N dealkylation of the short phenylethyl chain to yield the secondary amine (M5) occurs most readily, followed by N demethylation to yield norverapamil (M4), and N dealkylation of the long phenyl-acetonitrile chain to produce M6. Four regioisomeric phenolic metabolites could potentially be produced by O demethylation of verapamil, and three of these are reported in human and animal studies. C. blakesleeana produced a novel metabolite, 22-O-demethyl-verapamil (M4), but not a previously described one, 32-O-demethyl-verapamil. As in rat and human liver microsomes (13), O demethylation of verapamil by C. blakesleeana occurred more in the phenylethyl moiety than in the phenylacetonitrile moiety. In both side chains, O demethylation appeared to occur more readily para to than meta to the substituted alkyl. The similarity of the regiochemistry of dealkylation of verapamil in both C. blakesleeana and mammals suggests that a similar set of P450 isozymes could be responsible for the transformations. Walles et al. (21) reported that N-demethyl-verapamil and O-demethyl-verapamil were produced via carbinolamine, N-formyl, hemiacetale, and formate intermediates in incubation experiments with verapamil and microsomes isolated from the human heart tissue. They carried out the microsomal incubation for 4 h, whereas our microbial culture was incubated with verapamil for 96 h. These intermediates could be transformed to the corresponding products during incubation. The metabolites of verapamil in human heart microsmes were found in the fungal system as M1, M2, M4, and M5, but the unstable intermediates were not identified in the present study.

In conclusion, C. blakesleeana can transform verapamil into 23 metabolites by N dealkylation, O demethylation, and sulfate conjugation. The phase I metabolites paralleled those in mammals. Therefore, C. blakesleeana could be used as a model for generating the mammalian phase I metabolites of verapamil.