Metabolism of L (−)-Carnitine by Enterobacteriaceae Under Aerobic

FEMS Microbiology Letters 174 (1999) 295^301

Metabolism of L(3)-carnitine by Enterobacteriaceae under aerobic conditions Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 Thomas ElMner a, Andrea PreuMer a, Ulrich Wagner b, Hans-Peter Kleber a;*

a Institut fuër Biochemie, Universitaët Leipzig, Talstr. 33, D-04103 Leipzig, Germany b Institut fuër Zoologie, Fakultaëtfuër Biowissenschaften, Pharmazie und Psychologie, Universitaët Leipzig, Talstr. 33, D-04103 Leipzig, Germany Received 11 January 1999; received in revised form 16 March 1999; accepted 16 March 1999

Abstract

Different Enterobacteriaceae, such as Escherichia coli, Proteus vulgaris and Proteus mirabilis, are able to convert L(3)- carnitine, via crotonobetaine, into Q-butyrobetaine in the presence of carbon and nitrogen sources under aerobic conditions. Intermediates of L(3)-carnitine metabolism (crotonobetaine, Q-butyrobetaine) could be detected by thin-layer chromatography. In parallel, L(3)-carnitine dehydratase, carnitine racemasing system and crotonobetaine reductase activities were determined enzymatically. Monoclonal antibodies against purified CaiB and CaiA from E. coli O44K74 were used to screen cell-free extracts of different Enterobacteriaceae (E. coli ATCC 25922, P. vulgaris, P. mirabilis, Citrobacter freundii, Enterobacter cloacae and Klebsiella pneumoniae) grown under aerobic conditions in the presence of L(3)-carnitine. z 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords: Enterobacteriaceae; Carnitine; Trimethylammonium compound; Carnitine dehydratase; Crotonobetaine reductase

1. Introduction Bacteria are able to metabolise L(3)-carnitine in di¡erent ways [9] using this quaternary ammonium L(3)-carnitine (R(3)-3-hydroxy-4-trimethylami- compound as sole source of carbon and nitrogen nobutyrate) is a ubiquitously occurring compound (e.g. Pseudomonas sp.; [10]) or only as sole source in nature. In eukaryotes, L(3)-carnitine is essential of carbon (e.g. Acinetobacter sp.; [11,12]) under for the transport of long-chain fatty acids through aerobic conditions. Enterobacteriaceae are able to the inner mitochondrial membrane [1,2]. In bacteria, convert L(3)-carnitine via crotonobetaine into Q-bu- the physiological function of L(3)-carnitine is un- tyrobetaine in the presence of carbon and nitrogen known. In addition to glycine betaine, one of the sources under anaerobic conditions, but they do not most widely distributed osmoprotectants, L(3)- assimilate the carbon skeleton and nitrogen [13,14]. carnitine was shown to serve as osmoprotectant Two enzymes, L(3)-carnitine dehydratase (EC in Escherichia coli [3] and other microorganisms 4.2.1.89) and crotonobetaine reductase, were found [4^8]. in E. coli to catalyse this reaction sequence [15,16]. A carnitine racemase activity able to convert D(+)-car- * Corresponding author. Tel.: +49 (341) 97 36992; nitine into L(3)-carnitine was subsequently also Fax: +49 (341) 97 36998; E-mail: [email protected] postulated [17]. Studies using whole cells of E. coli

FEMSLE 8746 29-4-99 296 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301 have shown that these enzymes are inducible in the 2.2. Cultivation of microorganisms and cell disruption presence of L(3)-carnitine or crotonobetaine under anaerobic conditions [18]. Although a role as exter- The strains were inoculated from agar slopes into nal electron acceptor of anaerobic respiration like complex media and cultivated as preculture under nitrate or fumarate [19] is postulated for crotonobe- aerobic conditions at 30³C up to the middle of the taine [14], the precise function of this two-step path- exponential growth phase. The complex media con- way is still unknown. The stimulation of anaerobic tained 17 g pancreatic peptone, 3 g yeast extract and growth of Enterobacteriaceae by crotonobetaine sup- 5 g NaCl per litre of deionised water. Cultivation ports this hypothesis [20]. was carried out in 1000 ml Erlenmeyer £asks con- L(3)-carnitine dehydratase has been puri¢ed and taining 250 ml complex medium supplemented either characterised [15]. A still unknown cofactor essential with 0.5% carnitine, 0.5% crotonobetaine or 0.5% Q- for enzyme activity was separated during puri¢cation butyrobetaine on a rotary shaker (175 rpm) at 30³C. Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 procedure. The addition of this low molecular mass Growth of cells was followed by measuring the ap- e¡ector ( 6 1000 Da) caused reactivation of the parent absorbance of the culture at 600 nm. Cells apoenzyme. It is not possible to replace the e¡ector were harvested at the end of the exponential growth with known coenzymes or cofactors involved in de- phase by centrifugation at 5000Ug for 15 min and hydration (hydration) reactions. The caiB gene-en- washed twice with 67 mM phosphate bu¡er (pH 7.5). coding L(3)-carnitine dehydratase was isolated by Cells were disrupted by grinding with Alcoa and oligonucleotide screening from a genomic library of protein was extracted with 10 mM phosphate bu¡er E. coli [21]. CaiB belongs to an operon, which con- (pH 7.5). Cell-free extracts were obtained by cen- sists of six open reading frames (caiTABCDE) [22]. trifugation at 15 000Ug for 45 min. Current studies have shown that caiA encodes cro- Cultivation under anaerobic conditions was car- tonobetaine reductase, which converts, together with ried out at 37³C in 1000-ml £asks ¢lled to the neck CaiB and the unknown cofactor, crotonobetaine into with medium as described above and stoppered air- Q-butyrobetaine [23]. Crotonobetaine reductase has tight. been puri¢ed and characterised [23]. The cai operon from E. coli O44K74 was induced by the global reg- 2.3. Enzyme assays ulatory proteins CRP and FNR and repressed by the histone-like protein H-NS [22,24]. L(3)-carnitine dehydratase assay was carried out The aim of our studies was to investigate the oc- according to Jung et al. [15]. The conversion of currence of carnitine metabolising enzymes by Enter- D(+)-carnitine into L(3)-carnitine was determined obacteriaceae under aerobic conditions. as described by Jung and Kleber [17]. Crotonobe- taine reductase activity was determined according to Dickie and Weiner [25] with crotonobetaine as 2. Materials and methods substrate [16]. Protein concentrations were determined according 2.1. Microorganisms to Bradford [26] using bovine serum albumin as standard. The speci¢c activity was de¢ned as Wmol E. coli O44K74, E. coli ATCC 25922, Proteus mi- substrate conversion per min mg protein. rabilis, Proteus vulgaris, Citrobacter freundii, Kleb- siella pneumoniae, Klebsiella oxytoca, Hafnia alvei, 2.4. Preparation of cofactor Morganella morganii, Providencia rettgeri, Serratia marcescens, Enterobacter agglomerans and Entero- E. coli O44K74 was grown at 37³C in complex bacter cloacae were used in the experiments [13]. medium supplemented with 0.13% L-carnitine, 0.2% All strains were obtained from `Institut fuër Medizi- fumarate and 1% glycerol under anaerobic condi- nische Mikrobiologie und Infektionsepidemiologie', tions. Cultivation was carried out in 1000 ml Erlen- Universitaët Leipzig. meyer £asks ¢lled to neck with medium and stop-

FEMSLE 8746 29-4-99 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301 297 pered air-tight. Cells were harvested at the end of the 2.8. SDS-PAGE and Western blotting exponential growth phase by centrifugation (5000Ug; 15 min), suspended in aqua dest. and dis- Antigen samples were separated by 12% SDS- rupted by two passages through a French pressure PAGE [28] and transferred onto 0.2 Wm nitrocellu- cell (SLM Instruments, Urbana, USA) operating at lose membranes (Serva, Heidelberg, Germany) [29]. 20 000 psi. Unbroken cells and debris were removed Following transfer, membranes were stained for pro- by centrifugation at 15 000Ug for 45 min at 4³C. tein with Ponceau S (0.2 mg ml31 in 2% acetic acid). Proteins were separated from cofactor by ultra¢ltra- After destaining, blots were blocked overnight at 4³C tion using an Amicon YM 01 membrane at 4³C. in PBS/0.2% Tween-20. Hybridoma supernatants (di- Afterwards cofactor was concentrated by lyophilisa- luted 1:3 in PBS/0.1% Tween-20) were incubated tion (Christ, Germany). Cofactor was suspended in with the membrane for 1 h at room temperature. aqua dest. and stored at 320³C. The negative mab 9D5 was employed for speci¢city Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 control. Immunoreactive bands were detected by per- 2.5. Determination of trimethylammonium compounds oxidase-conjugated goat anti-mouse IgG (Dianova) diluted 1:500 in PBS/0.1% Tween-20 and visualised Carnitine and the other quaternary ammonium ¢nally via 3,3P-diaminobenzidine tetrahydrochloride compounds were examined by thin-layer chromatog- as chromogen. raphy. Adsorbents, solute systems and Rf values have already been described [11]. 2.9. Chemicals

2.6. Puri¢cation of CaiA and CaiB L(3)-carnitine, D(+)-carnitine and crotonobetaine were generous gifts from Sigma Tau, Rome, Italy. Q- CaiA has been puri¢ed to electrophoretic homo- Butyrobetaine was a gift from Lonza AG, Basel, geneity from E. coli BL21(DE3) containing the Switzerland. Carnitine acetyltransferase was pur- plasmid pT7-7CaiA [23]. CaiB has been puri¢ed to chased from Boehringer, Mannheim, Germany. All electrophoretic homogeneity from a cell-free extract other chemicals were of analytical grade. of E. coli O44K74 according to Jung et al. [15].

2.7. Preparation of monoclonal antibodies 3. Results and discussion

Monoclonal antibodies against CaiA and CaiB 3.1. Metabolism of carnitine under aerobic conditions were obtained according to Preusser et al. [23]. Fe- male mice were immunised with puri¢ed CaiB and The ability of di¡erent Enterobacteriaceae to me- CaiA, which was covalently linked with biotin. Gen- tabolise L(3)-carnitine under aerobic conditions is eration of hybridomas and immunoglobulin isotyp- summarised in Table 1. Under aerobic conditions ing were performed as described previously [27]. Hy- P. vulgaris and P. mirabilis are able to convert bridoma supernatants were assayed for speci¢c L(3)-carnitine via crotonobetaine into Q-butyrobe- monoclonal antibodies (mabs) using enzyme-linked taine as under anaerobic conditions [13]. In contrast immunosorbent assays (ELISA) [23]. Each hybrido- to E. coli ATCC 25922, E. coli O44K74 showed no ma was cloned twice by limiting dilution prior to metabolism of L(3)-carnitine. E. coli O44K74 pos- characterisation of secreted mabs. Anti-CaiA-mab sesses an inducible, active and carrier-mediated up- and anti-CaiB-mab were selected for immunological take system for trimethylammonium compounds, enzyme analyses after cloning and stabilisation of which is repressed just as the expression of L(3)- related hybridomas. The speci¢city of the isolated carnitine dehydratase by oxygen [30]. But our studies mabs to the antigens was also tested by immunoblot- have shown that carnitine metabolism is also possi- ting experiments [23]. ble in genera Escherichia and Proteus under aerobiosis. This implies that the carrier-mediated carnitine uptake must be active under aerobiosis. C. freundii

FEMSLE 8746 29-4-99 298 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301

Table 1 Metabolism of L(3)-carnitine by di¡erent Enterobacteriaceae under aerobic conditions Strain Degradation of L(3)-carnitine Formation of metabolites Crotonobetaine Q-Butyrobetaine E. coli O44K74 333 E. coli ATCC 25922 + + + P. mirabilis +++ P. vulgaris +++ C. freundii + 33 P. rettgeri + 33 Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 and P. rettgeri convert L(3)-carnitine into Q-butyro- robically in the presence of crotonobetaine or betaine under anaerobic conditions in presence of L(3)-carnitine [18]. For the ¢rst time we could detect carbon and nitrogen sources [13]. In contrast to these carnitine metabolising enzymes in cell-free ex- anaerobic conditions, formation of crotonobetaine tracts of di¡erent Enterobacteriaceae grown under and Q-butyrobetaine could not be observed under aerobic conditions in presence of L(3)-carnitine (Ta- aerobiosis. Both strains degrade L(3)-carnitine com- ble 2). pletely. Up to now the degradation pathway could L(3)-carnitine dehydratase was detectable in cell- not be identi¢ed, because neither L(3)-carnitine de- free extracts of P. mirabilis, P. vulgaris and E. coli. hydrogenase nor L(3)-carnitine dehydratase were In contrast to E. coli ATCC 25922, E. coli O44K74 detectable. Other Enterobacteriaceae (E. agglomer- shows only very small L(3)-carnitine dehydratase ans, E. cloacae, K. pneumoniae, K. oxytoca, H. alvei, activity without cofactor supplementation. Isolated M. morganii and S. marcescens) were also investi- cofactor from anaerobic grown E. coli O44K74 in- gated, but they showed no metabolism of L(3)-car- creased not only the L(3)-carnitine dehydratase ac- nitine similarly as described under anaerobic condi- tivity in cell-free extracts from E. coli O44K74 but tions [13]. also in E. coli ATCC 25922 and P. mirabilis. This indicates the existence of a very similar or same co- 3.2. Occurrence of carnitine metabolising enzymes factor by other Enterobacteriaceae under aerobiosis. Only the crude extract of P. vulgaris showed cofactor In E. coli O44K74 it was shown that L(3)-carni- saturation. Absence of metabolism of L(3)-carnitine tine dehydratase and crotonobetaine reductase are in E. coli O44K74 (Table 1) is obviously caused by inducible enzymes detectable in cells grown anae- the lacking cofactor. In C. freundii the expression of

Table 2 Speci¢c activities of L(3)-carnitine dehydratase, crotonobetaine reductase and the D(+)-carnitine racemasing system in cell-free extracts from di¡erent Enterobacteriaceae grown aerobically in presence of L(3)-carnitine Strain Speci¢c activity (mU mg31) L(3)-carnitine dehydratase L(3)-carnitine dehydratase Crotonobetaine D(+)-carnitine racemasing +cofactor reductase+cofactor system+cofactor E. coli O44K74 0.6 (25.8)a 8.2 (231.0) ^ (27.6) ^ (15.9) E. coli ATCC 25922 61.1 (138.3) 74.3 (647.8) 4.6 (70.2) ^ (13.4) P. mirabilis 34.4 (19.8) 51.9 (2708.7) 1.2 (20.7) 0.1 (10.2) P. vulgaris 75.0 (6.1) 74.0 (901.1) 5.1 (1.2) 3.4 (10.5) C. freundii ^ (207.5) ^ (1215.7) ^ (20.5) ^ (51.7) P. rettgeri ^ (71.0) ^ (2171.6) ^ (75.1) ^ (6.5) The same amount of cofactor was added to all performed incubation mixtures. aSpeci¢c activities of corresponding enzymes in cell-free extracts from di¡erent Enterobacteriaceae grown under anaerobic conditions in presence of L(3)-carnitine.

FEMSLE 8746 29-4-99 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301 299 Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021

Fig. 1. Evidence of CaiB in cell-free extracts of di¡erent Enterobacteriaceae by monoclonal antibodies. Cell-free extracts were separated by SDS-PAGE, blotted onto nitrocellulose membranes and immunostained with antibodies against CaiB from E. coli O44K74 (cf. Section 2.7). Puri¢ed CaiB shows two additional bands at 25 and 20 kDa due to fragments of CaiB (cf. lane 8). BOEHRINGER protein standard as shown were: triosephosphate isomerase, 26 600; aldolase, 39 200; glutamate dehydrogenase, 55 600; fructose-6-phosphate kinase, 85 200; L-galactosidase, 116 400.

L(3)-carnitine dehydratase and crotonobetaine re- low concentrations of the cofactor and/or the en- ductase seems to be repressed by oxygen. Cell-free zymes or the a¤nity of cofactor to L(3)-carnitine extracts of E. agglomerans, E. cloacae, K. oxytoca, K. dehydratase is higher than to crotonobetaine reduc- pneumoniae, H. alvei, M. morganii and S. marcescens tase and carnitine racemasing system. D(+)-carnitine were also investigated, but no L(3)-carnitine dehy- racemasing system and crotonobetaine reductase dratase activity could be detected even with cofactor were tested additionally in cell-free extracts of E. supplementation. In contrast to the L(3)-carnitine cloacae and K. pneumoniae, but no signi¢cant activ- dehydratase, addition of the cofactor was always es- ity could be found. sential for evidence of D(+)-carnitine racemasing Under anaerobic conditions E. coli, P. mirabilis, P. system and the crotonobetaine reductase (Table 2). vulgaris and C. freundii convert L(3)-carnitine into The reason for the discrepancies are probably very Q-butyrobetaine almost quantitatively [13]. These re-

Fig. 2. Evidence of CaiA in cell-free extracts of di¡erent Enterobacteriaceae by monoclonal antibodies. Cell-free extracts were separated by SDS-PAGE, blotted onto nitrocellulose membranes and immunostained with antibody against CaiA from E. coli O44K74 (cf. Section 2.7). Protein standard cf. Fig. 1.

FEMSLE 8746 29-4-99 300 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301 sults were achieved by thin-layer chromatography against CaiB was observed [23]. Screening of cell- and are compatible with detected enzyme activities free extracts by means of a monoclonal antibody (cf. Table 2). In comparison, the enzyme activities against CaiA is shown in Fig. 2. In E. coli O44K74 under anaerobiosis are up to 50-fold higher than caiA is expressed only under anaerobic conditions. under aerobic conditions. Obviously, the transcrip- However, the expression of caiA is realised under tion of the cai operon is reduced under aerobiosis. aerobic conditions by E. coli ATCC 25922, P. mira- This supports the hypothesis of crotonobetaine as bilis and P. vulgaris. external electron acceptor of anaerobic respiration In summary, the expression of caiB under aerobic [14]. conditions could be veri¢ed by E. coli O44K74, E. Besides L(3)-carnitine (cf. Table 2) crotonobe- coli ATCC 25922 and P. vulgaris using antibodies taine also acts as inducer under aerobic conditions. and L(3)-carnitine dehydratase activity by measur- In the presence of crotonobetaine the following ing in cell-free extracts. Although L(3)-carnitine de- Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 L(3)-carnitine dehydratase activities (with addition hydratase activity was detected enzymatically in P. of cofactor) were obtained: E. coli O44K74 (23.9 mirabilis, no cross reactivity with anti-CaiB-mab was mU mg31), E. coli ATCC 25922 (16.5 mU mg31), observed. CaiA is expressed under aerobiosis only in P. mirabilis (60.1 mU mg31) and P. vulgaris (64.2 E. coli ATCC 25922, P. mirabilis and P. vulgaris. mU mg31). Q-Butyrobetaine does not show any e¡ect In further investigations, we intend to carry out on expression of carnitine metabolising enzymes regulation studies in order to understand the aero- under aerobiosis similar to anaerobic conditions [18]. bic/anaerobic shift.

3.3. Detection of carnitine metabolising enzymes with monoclonal antibodies Acknowledgments

It is supposed that structural analogies exist be- This work was supported by the Deutsche For- tween CaiB and CaiA of di¡erent Enterobacteria- schungsgemeinschaft grant No. K/A 911/1-2. The ceae. That's why di¡erent cell-free extracts were authors thank Stan Theophilou for help with the screened with monoclonal antibodies against CaiB manuscript. obtained from E. coli O44K74. CaiB from E. coli O44K74 has a relative molecular mass of 85 kDa and is a dimer consisting of two identical subunits References (45 kDa) [15]. Monoclonal antibody raised against CaiB from E. coli O44K74 showed only reactivity [1] Bremer, J. (1963) Carnitine in intermediary metabolism - the with the puri¢ed subunit of CaiB while no reaction biosynthesis of palmitoylcarnitine by cell subfractions. J. Biol. against CaiA was observed. Anti-CaiB-mab recog- Chem. 238, 2774^2779. nises proteins from E. coli ATCC 25922 and P. vul- [2] Fritz, I.B. and Yue, K.T.N. (1963) Long chain acylcarnitine acyltransferase and the role of acylcarnitine derivatives in the garis (Fig. 1). Cell-free extract from C. freundii also catalytic increase of fatty acid oxidation induced by carnitine. showed a low cross reactivity with anti-CaiB-mab. J. Lipid Res. 4, 279^288. Obviously the amount of L(3)-carnitine dehydratase [3] Jung, H., Jung, K. and Kleber, H.-P. (1990) L-Carnitine me- was not su¤cient for an enzymatic determination. tabolism and osmotic stress response in Escherichia coli. These results suggest that the subunits of CaiB in J. Basic Microbiol. 30, 409^413. [4] Jebbar, M., Champion, C., Blanco, C. and Bonnassie, S. E. coli ATCC 25922, P. vulgaris and C. freundii (1998) Carnitine acts as a compatible solute in Brevebacterium have nearly the same molecular mass as in E. coli linens. Res. Microbiol. 149, 211^219. O44K74. [5] Kappes, R.M. and Bremer, E. (1998) Response of Bacillus CaiA from E. coli O44K74 has a native molar subtilis to high osmolarity: uptake of carnitine, crotonobe- mass of 164.4 kDa and is composed of four identical taine and Q-butyrobetaine via ABC-transport system Opu C. Microbiology 144, 83^90. subunits with relative molar masses of 41.5 kDa [23]. [6] Kets, E.P.W., Galinski, E.A. and De Bont, J.A.M. (1994) By Western blotting anti-CaiA-mab was found to Carnitine: a novel compatible solute in Lactobacillus planta- recognise the subunit of CaiA while no reaction rum. Arch. Microbiol. 162, 243^248.

FEMSLE 8746 29-4-99 T. ElMner et al. / FEMS Microbiology Letters 174 (1999) 295^301 301

[7] Lucchesi, G.I., Lisa, T.A., Casale, C.H. and Domenech, C.E. [19] Haddock, B.A. and Jones, C.W. (1977) Bacterial respiration. (1995) Carnitine resembles choline in the induction of chol- Bacteriol. Rev. 41, 47^99. inesterase, acid phosphatase, and phospholipase C and in its [20] Seim, H., Loëster, H., Claus, R., Kleber, H.-P. and Strack, E. action as an osmoprotectant in Pseudomonas aeruginosa. Curr. (1982) Stimulation of the anaerobic growth of Salmonella ty- Microbiol. 30, 55^60. phimurium by reduction of L-carnitine, carnitine derivatives [8] Smith, L.T. (1996) Role of osmolytes in adaptation of osmoti- and structure-related trimethylammonium compounds. Arch. cally stressed and chill-stressed Listeria monocytogenes grown Microbiol. 132, 91^95. in liquid media and on processed meat surfaces. Appl. Envi- [21] Eichler, K., Schunck, W.-H., Kleber, H.-P. and Mandrand- ron. Microbiol. 62, 3088^3093. Berthelot, M.-A. (1994) Cloning, nucleotide sequence and ex- [9] Kleber, H.-P. (1997) Bacterial carnitine metabolism. FEMS pression of the Escherichia coli gene encoding carnitine dehy- Microbiol. Lett. 147, 1^9. dratase. J. Bacteriol. 176, 2970^2975. [10] Aurich, H., Kleber, H.-P. and Schoëpp, W. (1967) An inducible [22] Eichler, K., Bourgis, F., Buchet, A., Kleber, H.-P. and Mand- carnitine dehydrogenase from Pseudomonas aeruginosa. Bio- rand-Berthelot, M.-A. (1994) Molecular characterization of

chim. Biophys. Acta 139, 505^507. the cai operon necessary for carnitine metabolism in Esche- Downloaded from https://academic.oup.com/femsle/article/174/2/295/502865 by guest on 29 September 2021 [11] Kleber, H.-P., Seim, H., Aurich, H. and Strack, E. (1977) richia coli. Mol. Microbiol. 13, 775^786. Verwertung von Trimethylam-moniumverbindungen durch [23] Preusser, A., Wagner, U., Elssner, T. and Kleber, H.-P. (1999) Acinetobacter calcoaceticus. Arch. Microbiol. 112, 201^206. Crotonobetaine reductase from Escherichia coli consists of [12] Seim, H., Loëster, H., Claus, R., Kleber, H.-P. and Strack, E. two proteins. Biochim. Biophys. Acta, in press. (1982) Splitting of the C-N-bond in carnitine by an enzyme [24] Eichler, K., Buchet, A., Lemke, R., Kleber, H.-P. and Mand- (trimethylamine forming) from membranes of Acinetobacter rand-Berthelot, M.-A. (1996) Identi¢cation and characteriza- calcoaceticus. FEMS Microbiol. Lett. 15, 165^167. tion of the caiF gene encoding a potential transcriptional [13] Seim, H., Ezold, R., Kleber, H.-P. and Strack, E. (1980) Sto¡- activator of carnitine metabolism in Escherichia coli. J. Bac- wechsel des L-Carnitins bei Enterobakterien. Z. Allg. Mikro- teriol. 178, 1248^1257. biol. 20, 591^594. [25] Dickie, P. and Weiner, J.H. (1979) Puri¢cation and character- [14] Seim, H., Loëster, H., Claus, R., Kleber, H.-P. and Strack, E. ization of membrane-bound fumarate reductase from anae- (1982) Stimulation of the anaerobic growth of Salmonella ty- robically grown Escherichia coli. Can. J. Biochem. 57, 813^ phimorium by reduction of L-carnitine, carnitine derivatives 821. and structure-related trimethylammonium compounds. Arch. [26] Bradford, M.M. (1976) A rapid and sensitive method for the Microbiol. 132, 91^95. quantitation of microgram quantities of protein utilizing the [15] Jung, H., Jung, K. and Kleber, H.-P. (1989) Puri¢cation and principle of protein-dye binding. Anal. Biochem. 72, 248^254. properties of carnitine dehydratase from Escherichia coli -a [27] Wagner, U., Lachmann, I., Haëdge, D. and Droëssler, K. (1997) new enzyme of carnitine metabolization. Biochim. Biophys. Development and characterization of monoclonal antibodies Acta 1003, 270^276. speci¢c for two di¡erent exoproteases of aeromonas salmoni- [16] Roth, S., Jung, K., Jung, H., Hommel, R.K. and Kleber, cida. J. Fish Dis. 20, 223^235. H.-P. (1994) Crotonobetaine reductase from Escherichia coli [28] Laemmli, U.K. (1970) Cleavage of structural proteins during - a new inducible enzyme of anaerobic metabolization of the assembly of the head of bacteriophage T4. Nature 227, L-(3)-carnitine. Antonie van Leeuwenhoek 65, 63^69. 680^685. [17] Jung, H. and Kleber, H.-P. (1991) Metabolism of D-(+)-car- [29] Towbin, H., Staehlin, T. and Gordon, J. (1979) Electropho- nitine by Escherichia coli. Appl. Microbiol. Biotechnol. 35, retic transfer of proteins from polyacrylamide gels to nitro- 393^395. cellulose sheets: procedure and some applications. Proc. Natl. [18] Jung, K., Jung, H. and Kleber, H.-P. (1987) Regulation of L- Acad. Sci. USA 76, 4350^4354. carnitine metabolism in Escherichia coli. J. Basic Microbiol. [30] Jung, H., Jung, K. and Kleber, H.-P. (1990) L-carnitine up- 27, 131^137. take by Escherichia coli. J. Basic Microbiol. 30, 507^514.

FEMSLE 8746 29-4-99