Journal of Industrial Microbiology & Biotechnology

, Volume 34, Issue 3, pp 219–224

Identification of metabolites produced from N-phenylpiperazine by Mycobacterium spp

Authors

  • M. D. Adjei
    • Division of MicrobiologyNational Center for Toxicological Research, US Food and Drug Administration
    • Norfolk Department of Public Health
  • J. Deck
    • Division of MicrobiologyNational Center for Toxicological Research, US Food and Drug Administration
  • T. M. Heinze
    • Division of Biochemical ToxicologyNational Center for Toxicological Research, US Food and Drug Administration
  • J. P. Freeman
    • Division of Biochemical ToxicologyNational Center for Toxicological Research, US Food and Drug Administration
  • A. J. Williams
    • Division of MicrobiologyNational Center for Toxicological Research, US Food and Drug Administration
    • Division of MicrobiologyNational Center for Toxicological Research, US Food and Drug Administration
Original Paper

DOI: 10.1007/s10295-006-0189-x

Cite this article as:
Adjei, M.D., Deck, J., Heinze, T.M. et al. J Ind Microbiol Biotechnol (2007) 34: 219. doi:10.1007/s10295-006-0189-x

Abstract

Mycobacterium sp. 7E1B1W and seven other mycobacterial strains known to degrade hydrocarbons were investigated to determine their ability to metabolize the piperazine ring, a substructure found in many drugs. Cultures were grown at 30°C in tryptic soy broth and dosed with 3.1 mM N-phenylpiperazine hydrochloride; samples were removed at intervals and extracted with ethyl acetate. Two metabolites were purified from each of the extracts by high-performance liquid chromatography; they were identified by mass spectrometry and 1H nuclear magnetic resonance spectroscopy as N-(2-anilinoethyl)acetamide and N-acetyl-N′-phenylpiperazine. The results show that mycobacteria have the ability to acetylate piperazine rings and cleave carbon-nitrogen bonds.

Keywords

BiotransformationFluoroquinolonesMycobacteriumN-phenylpiperazinePiperazine

Introduction

Many synthetic drugs, including antibacterial and antifungal agents, anthelmintics, anxiolytics, and antidepressants, contain a piperazine ring that plays a significant role in potency [14, 16, 18, 29]. For example, the antibacterial activity of some fluoroquinolone drugs is reduced or lost upon the substitution or partial removal of the piperazinyl moiety [33]. Also, because piperazinyl compounds are used in textile dyeing and analytical chemistry [19], they have been found in industrial wastewater [26].

Piperazine is biodegraded by several bacteria in the genera Arthrobacter and Mycobacterium [13, 15]. Mycobacterium smegmatis, Nocardia sp., Pseudomonas aeruginosa, P. fluorescens, and Streptomyces griseus demethylate the piperazine ring of the fluoroquinolone drug danofloxacin to produce N-desmethyldanofloxacin; and M. smegmatis and P.fluorescens degrade the piperazine ring of danofloxacin to produce 1-cyclopropyl-6-fluoro-7-amino-4-oxo-1,4-dihydroquinoline-3-carboxylic acid [8]. The ability of mycobacteria to biotransform piperazine rings [1, 2] may be especially important because piperazinyl fluoroquinolones are used in combination therapy to treat drug-resistant tuberculosis [7].

Several fungi metabolize piperazine rings in fluoroquinolones. For instance, Penicillium spp. strains cleave the piperazine ring of danofloxacin [8] and Gloeophyllumstriatum mineralizes carbon from the piperazine rings of enrofloxacin and ciprofloxacin to CO2 [31, 32]. Umbelopsisramanniana (Mucorramannianus) produces N-acetylated metabolites and piperazine ring-cleavage products from enrofloxacin and sarafloxacin [20, 21] and Pestalotiopsisguepini similarly metabolizes the piperazine rings of ciprofloxacin and norfloxacin [22].

N-Phenylpiperazine, a model compound structurally related to piperazinyl fluoroquinolones [14], is also known to have adrenergic blocking properties [9]. Mycobacterium sp. 7E1B1W (ATCC 29676), a non-pathogenic, fast-growing soil bacterium that degrades propane and n-tetradecane but not phenanthrene or pyrene [35], and seven other mycobacteria were investigated for the potential to metabolize N-phenylpiperazine.

Materials and methods

Cultural conditions

Cultures of Mycobacterium spp. were grown for 3 days in 125-ml flasks containing 30 ml tryptic soy broth (TSB) (Remel Inc., Lenexa, KS). They were then dosed with 3.1 mM (final concentration) N-phenylpiperazine hydrochloride (99%, Aldrich Chemical Co., Milwaukee, WI) that had been dissolved in water and filter-sterilized. Controls consisted of cultures that were not dosed and non-inoculated flasks of TSB dosed with N-phenylpiperazine. Cultures and controls were incubated aerobically at 30°C with shaking at 200 rpm. After 4 days, they were extracted with ethyl acetate for analysis by HPLC.

For kinetic studies, cultures and controls were incubated for 3 days in 500 ml flasks containing 100 ml TSB and then were dosed with 3.1 mM N-phenylpiperazine. At various intervals from 0 to 8 days, samples were withdrawn aseptically and extracted with ethyl acetate for analysis.

To determine the effect of N-phenylpiperazine concentration on metabolite production, cultures were grown for 3 days and then were dosed with N-phenylpiperazine (0, 0.4, 0.8, 1.5, or 3.1 mM). They were incubated as described above, harvested 4 days after dosing, and extracted with ethyl acetate for analysis.

Extraction and HPLC analysis

Cultures were centrifuged at 15,000 g for 10 min at 4°C and the supernatants were extracted three times with equal volumes of ethyl acetate. The extracts were dried in vacuo and dissolved in methanol. Metabolites were detected and purified by high-performance liquid chromatography (HPLC) using an Agilent Technologies (Palo Alto, CA) 1100 Series chromatograph; the diode array detector was monitored at 280 nm. A Phenomenex (Torrance, CA) Prodigy 5-μm ODS-3 column (250 × 4.6 mm) was used with a solvent system consisting of A (methanol: water: acetic acid, 10:90:0.2) and B (methanol: water: acetic acid, 90:10:0.2) at a flow rate of 2.0 ml/min. Solvent B was increased from 10 to 95% in a 20-min linear gradient and held at 95% for 10 min.

Mass spectrometry

Direct exposure probe/electron ionization mass spectrometry (DEP/EI-MS) was performed on a ThermoFinnigan (San Jose, CA) TSQ700 mass spectrometer in the electron-ionization (EI) single quadrupole mode. The ion source temperature was 150°C and the electron energy was 70 eV (uncorrected). The first quadrupole analyzer was scanned from m/z 50 to 550 in 0.7 s. The rhenium wire of the DEP was heated from 0 to 800 mA with a linear ramp of 5 mA/s.

Liquid chromatography/electrospray ionization tandem mass spectrometry (LC/ESI-MS/MS) analyses were performed on a ThermoFinnigan Quantum Ultra mass spectrometer equipped with an Agilent Technologies 1100 Series HPLC. The mass spectrometer was operated in the positive-ion ESI mode with an in-source collision-induced dissociation (CID) offset of −15 or −20 V. Other ESI conditions were: spray voltage 3.0 kV, capillary temperature 350°C, sheath gas pressure 40 psi, ion sweep gas 10, and auxiliary gas 15. MS/MS conditions were: argon collision pressure 1.5 mTorr, collision energy 15 eV, parent set masses (m/z) 163, 179, and 205, and Q3 scanned m/z 30–250 per 0.5 s. HPLC was performed with a Phenomenex Prodigy 5-μm ODS-3 column (2.0 × 250 mm). The mobile phase was a linear gradient from 5% acetonitrile to 95% acetonitrile in 40 min, with constant 0.1% formic acid, at a flow rate of 0.2 ml/min.

NMR spectroscopy

Samples were dissolved in deuterated methanol and 1H nuclear magnetic resonance (NMR) spectroscopy was performed at 500 MHz using a Bruker Instruments (Billerica, MA) AM500 NMR spectrometer [21].

Statistical analysis

A one-way analysis of variance (anova) was performed to determine the effect of substrate concentration on metabolite production, using the statistical software JMP v 5.1 (SAS Institute). Differences among concentration means were compared using Tukey’s test.

Results

N-Phenylpiperazine was analyzed by LC/ESI-MS/MS for comparison with the metabolites. Fragmentation of the protonated molecule at m/z 163 gave the product-ion mass spectrum shown (Fig. 1 a). Ions included the residual protonated molecule at m/z 163, a minor N,N-divinylbenzenaminium ion at m/z 146, and a major N-vinylbenzenaminium ion at m/z 120.
https://static-content.springer.com/image/art%3A10.1007%2Fs10295-006-0189-x/MediaObjects/10295_2006_189_Fig1_HTML.gif
Fig. 1

Product-ion mass spectra obtained by LC/ESI-MS/MS: (a) N-phenylpiperazine (MH+ = 163), (b) metabolite I, N-(2-anilinoethyl)acetamide (MH+ = 179), and (c) metabolite II, N-acetyl- N′-phenylpiperazine (MH+ = 205). The metabolites were produced from N-phenylpiperazine by Mycobacterium sp. 7E1B1W

During the HPLC analysis of extracts from dosed Mycobacterium sp. 7E1B1W cultures, N-phenylpiperazine eluted at 9.8 min and two small peaks (I and II) not found in control cultures eluted at 16.8 and 22.5 min, respectively. In extracts from cultures harvested 4 days after dosing, peaks I and II formed about 1.7 and 0.9%, respectively, of the total integrated peak area at 280 nm.

The DEP/EI mass spectrum of metabolite I had ions at m/z 178 (22) [M .], 119 (39) [C6H5-NH-CH = CH2, 118 (12), 107 (7), 106 (100) [C6H5-NH-CH2, 79 (5), and 77 [C6H5. Metabolite I was analyzed by LC/ESI-MS/MS. Fragmentation of the protonated molecule at m/z 179 gave the product-ion mass spectrum shown (Fig. 1b). Ions included an extremely minor protonated molecule at m/z 179, a minor N-vinylbenzenaminium ion at m/z 120, a major N-acetylethylenaminium ion at m/z 86, and a minor ethylenaminium ion at m/z 44. The NMR resonances for metabolite I (Table 1) were nearly identical to those of N-phenylpiperazine except that the H3 resonances had been shifted slightly downfield. The H2 and H3 resonances integrated as two each, indicating that four proton resonances of the original piperazine moiety were missing. A sharp singlet at 1.92 ppm integrated as three. From the NMR spectrum, the singlet was determined to be an acetyl group attached to the terminal nitrogen.
Table 1

1H NMR data for N-phenylpiperazine and metabolites I and II produced by Mycobacterium sp. 7E1B1W

Compounda

Chemical shifts (ppm)b

Coupling constants (Hz)b

H2′, H6′

H3′, H5′

H4′

H2, H6

H3, H5

Acetyl

N-Phenylpiperazine

6.99

7.26

6.89

3.22

3.29

 

J2,3 = 5.2, J2′,3′ = 8.6, J2′,4′ = 1.1, J3′,4′ = 7.3

N-(2-Anilinoethyl)-acetamide (I)

6.63

7.07

6.60

3.21

3.36

1.92

J2,3 = 6.5, J2′,3’ = 8.6, J2′,4′ = 1.1, J3′,4′ = 7.3

N-Acetyl-N′-phenyl-piperazine (II)

6.98

7.24

6.85

3.12, 3.18

3.68, 3.72

2.13

J2,3 = 5.2, J2′,3′ = 8.6, J2′,4’ = 1.1, J3′,4’ = 7.3

aCompounds were dissolved in deuterated methanol

bFor NMR assignments, piperazine-ring positions are numbered from 1 to 6 and phenyl-ring positions from 1′ to 6′. (The piperazine ring in metabolite I is incomplete and lacks positions 5 and 6.)

A standard for N-(2-anilinoethyl)acetamide was synthesized by dissolving 50 mg N-phenylethylenediamine (Fisher Scientific, Pittsburgh, PA) in 4.5 ml water and adding 0.3 ml acetic anhydride (Eastman Chemical, Kingsport, TN). Of the two products formed, one was identified by LC/ESI-MS/MS as N-(2-anilinoethyl)acetamide (MH+ = m/z 179) by the prominent product ion at m/z 86. Because the HPLC retention times, DEP/EI and product-ion mass spectra, and NMR spectra were identical for metabolite I and the synthetic N-(2-anilinoethyl)acetamide standard, metabolite I was identified as N-(2-anilinoethyl)acetamide (Fig. 1b).

The DEP/EI mass spectrum of metabolite II had ions at m/z 204 (48) [M .], 189 (4), 161 (14), 132 (100) [C6H5-N(CH2)CH = CH2, 120 (21), 119 (26) [C6H5-NH-CH = CH2, 106 (12) [C6H5-NH-CH2, 105 (21), 104 (21), 77 (15) [C6H5, and 56 (18). The mass spectrum matched that of 1-(4-phenyl-1-piperazinyl)ethanone (CAS# 2155713-1, National Institute of Standards and Technology reference library) with a reverse fit of 94%. Metabolite II was analyzed by LC/ESI-MS/MS. Fragmentation of the protonated molecule at m/z 205 gave the product-ion mass spectrum shown (Fig. 1c). Ions included the residual protonated molecule at m/z 205, a major N-phenylpiperazinium ion at m/z 163, an N-vinylbenzenaminium ion at m/z 120, and an N-acetyl-N-vinylethylenaminium ion at m/z 112. The NMR data for metabolite II (Table 1) showed a downfield chemical shift for the H3 resonance when compared to that of N-phenylpiperazine. A resonance representing the three protons of an acetyl group had been added. The H2, H6 and H3, H5 resonances of the piperazine ring were split into two triplet resonances each due to the asymmetry of the acetyl group.

A standard for N-acetyl-N′-phenylpiperazine was synthesized by dissolving 50 mg N-phenylpiperazine hydrochloride in 4.5 ml water and adding 0.3 ml acetic anhydride. Because the HPLC retention times, DEP/EI and product-ion mass spectra, and NMR spectra were identical for metabolite II and the synthetic standard and matched the data from the reference library, metabolite II was identified as N-acetyl-N′-phenylpiperazine (Fig. 1c).

In kinetics experiments with Mycobacterium sp. 7E1B1W, metabolites I and II were detected simultaneously 6 h after the addition of N-phenylpiperazine to duplicate cultures (Fig. 2) and were at approximately equal concentrations at 12 h. After 18 h, about twice as much of metabolite I as of metabolite II was found. Between 1 and 4 days, the rates of formation of the metabolites were approximately linear; both compounds continued to increase in concentration until the experiment was terminated at 8 days.
https://static-content.springer.com/image/art%3A10.1007%2Fs10295-006-0189-x/MediaObjects/10295_2006_189_Fig2_HTML.gif
Fig. 2

Time course of N-phenylpiperazine metabolism by Mycobacterium sp. 7E1B1W incubated in TSB for 8 days after dosing. Results are means (n = 2), with error bars, of the peak areas at 280 nm for metabolites I (filled squares) and II (open squares)

The rates of formation of metabolites I and II appeared to be correlated with substrate concentration (Fig. 3). Increasing the concentration of N-phenylpiperazine from 0.4 to 3.1 mM in triplicate cultures led to a significant increase at 4 days in metabolite I (P < 0.0001 using Tukey’s test) but not in metabolite II. Starting with 1.5–3.1 mM N-phenylpiperazine, the amount of metabolite I detected was about twice that of metabolite II.
https://static-content.springer.com/image/art%3A10.1007%2Fs10295-006-0189-x/MediaObjects/10295_2006_189_Fig3_HTML.gif
Fig. 3

Formation of metabolites I (filled bars) and II (open bars) by Mycobacterium sp. 7E1B1W incubated for 4 days in TSB with different concentrations of N-phenylpiperazine. Results are means (n = 3), with error bars, of the peak areas at 280 nm

To find whether other mycobacteria would produce metabolites from N-phenylpiperazine, seven other strains were tested, including M. frederiksbergense FAn9 (DSM 44346), M. gilvum ATCC 43909, M. gilvum BB1 (DSM 9487), M. gilvum PYR-GCK (ATCC 700033) [17], M. smegmatis mc2155 (ATCC 700084), Mycobacterium sp. PYR100 [17], and Mycobacterium sp. RJGII-135 [28]. The same two metabolites (I and II) were detected in extracts from dosed cultures of all strains; the identities were confirmed by HPLC retention times and product-ion spectra (data not shown).

Discussion

Eight different Mycobacterium spp. strains, when tested for the ability to metabolize N-phenylpiperazine, produced N-acetylated metabolites. The piperazine rings of fluoroquinolone drugs may be N-acetylated by fungi [2022] and mycobacteria [1, 2]; similar reactions also occur in experimental animals [11] and in humans [23]. The acetylated drug metabolites found in animals and human urine could also result from microbial metabolism. N-Acetyl-N′-phenylpiperazine has been synthesized previously for use as an activator in the radical polymerization of methyl methacrylate [30].

Although arylamine N-acetyltransferases from bacteria metabolize arylamines, arylhydrazines, and arylhydroxylamines [6, 24, 25], these enzymes are not known to acetylate piperazine rings. Mycobacterium sp. 7E1B1W also N-acetylates the piperazine rings of norfloxacin and ciprofloxacin [1, 2] but it did not acetylate p-aminobenzoic acid (data not shown).

Carbon–nitrogen bond cleavage, which occurred in the formation of N-(2-anilinoethyl)acetamide, is known in the metabolism of other cyclic secondary amines by mycobacteria [10, 12, 27]. It is typical of the metabolism of piperazine-containing fluoroquinolones by fungi [2022], animals [11], and humans [23].

The metabolism of N-phenylpiperazine by Mycobacterium sp. 7E1B1W may proceed in two possible sequences, depending on whether N-acetylation precedes or follows C–N bond cleavage. Because of the simultaneous appearance of the two metabolites, no evidence is available at this time to show the sequence.

In summary, eight strains of Mycobacterium spp. modified the piperazine ring of N-phenylpiperazine, forming N-(2-anilinoethyl)acetamide and N-acetyl-N′-phenylpiperazine as products.

Acknowledgments

We thank Dr. C. E. Cerniglia for providing the Mycobacterium strains and also thank him and Dr. F. Rafii for helpful comments on the manuscript. This work was supported in part by an appointment to the postgraduate research program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the US Food and Drug Administration. The views presented in this article do not necessarily reflect those of the Food and Drug Administration.

Copyright information

© Society for Industrial Microbiology 2006