Degradation of Caffeine and Related Methylxanthines by Serratia

Transcription

1 Microb Ecol (1996) 31: MICROBIAL ECOLOGY 1996 Springer-Verlag New York Inc. Degradation of Caffeine and Related Methylxanthines by Serratia marcescens Isolated from Soil Under Coffee Cultivation E Mazzafera, I O. Olsson, 2 G. Sandberg 2 ~Departamento de Fisiologia Vegetal, IB, CP6109, Unicamp, CEP , Campinas, SR Brazil 2Department of Forest Genetics and Plant Physiology, The Swedish University of Agricultural Sciences, S , Urne~, Sweden Received: 10 April 1995; Revised: 28 June 1995 Abstract. A strain of Serratia marcescens showing the ability to degrade caffeine and other methylxanthines was isolated from soil under coffee cultivation. Growth was observed only with xanthines methylated at the 7 position (caffeine, 1,3,7-trimethylxanthine; paraxanthine, 1,7-dimethylxanthine; theobromine, 3,7-dimethylxanthine and 7-methylxanthine). Paraxanthine and theobromine were released in liquid medium when caffeine was used as the sole source of carbon and nitrogen. When paraxanthine or theobromine were used, 3-methylxanthine, 7-methylxanthine, and xanthine were detected in the liquid medium. Serratia marcescens did not grow with theophylline (1,3-dimethylxanthine), 1-methylxanthine, and 3-methylxanthine, and poor growth was observed with xanthine. Methyluric acid formation from methylxanthines was tested in cell-free extracts by measuring dehydrogenase reduction of tetrazolium salt in native-polyacrylamide gel electrophoresis gel. Activity was observed for all methylxanthines, even those with which no bacterial growth was observed. Our results suggest that in this strain of S. marcescens caffeine is degraded to theobromine (3,7-dimethylxanthine) and/or paraxanthine (1,7-dimethylxanthine), and subsequently to 7-methylxanthine and xanthine. Methyluric acid formation could not be confirmed. Introduction The organic matter of soils is defined as a mixture of plant and animal residues in various stages of decomposition, compounds produced during the decomposition process, and microorganisms and small animals and their decomposing remains [6]. For simplicity, the organic matter has been defined as humic and nonhumic substances. The first group is constituted of amorphous material with a high degree of complexity and with no consistent chemical or physical characteristics. The Correspondence to: Paulo Mazzafera.

2 200 R Mazzafera et al. second group, composed of carbohydrates, proteins, amino acids, fats, waxes, alkanes, and low molecular weight organic acids, is readily used by microorganisms in the soil [11]. Soil structure, fertility, water content, and aeration can significantly influence the diversity as well as the microbial activity in the soil. However, soil characteristics and microorganism populations can be strongly affected by the plants grown on the soil surface. Soil continuously cultivated with the same crop might have in the nonhumic fraction a specific compound, found in abundance in the tissues of a certain species, which could favor growth of microorganisms able to use it as a source of energy. Caffeine has been mentioned as a potent allelopathic compound, inhibiting selectively weed growth under coffee trees (for a review see Waller et al. [14]) as well as being toxic to coffee roots [3]. It is clear from laboratory studies that caffeine can cause inhibition of seed germination in different species [7, 8], including coffee seeds [3]. However, degradation of caffeine by microorganisms has been neglected in these studies. In a survey of bacteria with the ability to degrade caffeine as the only source of nitrogen and carbon, we have isolated a strain of Serratia marcescens from soil continuously cultivated with coffee for -40 years. Serratia marcescens is an enterobacteria, being considered as an opportunistic pathogen of human patients in hospitals [12]. Breakdown of uric acid and xanthine by S. marcescens was previously reported [5, 9]. In this paper we report degradation of caffeine and related methylxanthines by S. marcescens, and suggest a degradation pathway. Materials and Methods Isolation of Serratia marcescens Soil was collected under coffee trees (-40 years old) growing at the Experimental Station of the Instituto Agron6mico, Campinas, Brazil. Leaves and debris were removed from the soil surface, and the topmost 15-cm layer was collected. In the laboratory 1 g of soil was agitated overnight (50 rpm) with 50 ml of distilled water in a 250-ml flask, at room temperature (25 C). The suspension was filtered with ordinary filter paper, and 50- to 100-1xl aliquots of the filtrate were plated in M9 agar plates [10] containing only caffeine (100 mg liter-) as the source of carbon and nitrogen. Incubation proceeded at 28 C until bacterial growth was observed. Single colonies were transferred to new M9- caffeine plates, and colonies growing on these plates were transferred once more to a new set of plates. Then, single colonies from these plates were used as inoculum for M9-caffeine liquid medium to confirm the ability of the isolates to grow on caffeine. Inoculated flasks (250 ml) containing 100 ml of medium were kept in a rotatory shaker (120 rpm) at room temperature (25 C). Bacterial Growth With Caffeine and Other Methylxanthines as the Only Sources of Carbon and Nitrogen The ability of the Serratia strain to grow in liquid medium containing other methylxanthines instead of caffeine was tested. Single colonies from caffeine plates were used as inoculum. Theobromine (3,7-dimethylxanthine), theophylline (1,3-dimethylxanthine), paraxanthine (1,7-dimethylxanthine), 1- methylxanthine, 3-methylxanthine, and 7-methylxanthine were tested at 200 mg liter ~ concentration. All methylxanthines were Sigma grade. Bacterial growth was also investigated with caffeine at 100, 300, 600, and 1,200 mg liter ~, and theobromine at 100, 300, and 600 mg liter -l. Xanthine was tested

3 Degradation of Caffeine by Serratia 201 at 100 mg liter '. Theobromine and xanthine were used in the lower concentrations because of their low solubility. Three replicates were made for each concentration. Serratia growth was followed by measuring the optical density at 600 nm in 1-ml aliquots taken from the flasks. These samples were centrifuged at 14,000 rpm in a bench-top centrifuge, and the supernatant was stored at -21 C until analysis for methylxanthine concentrations. Methylxanthines Estimation by High-Performance Liquid Chromatography High-performance liquid chromatography (HPLC) was performed in a Waters 510 HPLC system, and the compounds were separated on a C~s ODS-hypersil column with a 25-min gradient of 0-45% methanol in 0.5% acetic acid, ph 5 (with 1 N NaOH), at a flow rate of 1.0 ml rain -~. Compounds eluting from the column were detected at 280 nm with an ultraviolet (UV) monitor, and the peak areas were compared with those obtained with standards of known concentration. Identification of Compounds Released in the Liquid Medium Compounds released in the liquid medium with similar retention times to those of methylxanthines in the HPLC chromatograms were collected after elution from the UV monitor. They were freezedried, resuspended in distilled water, and subjected to isocratic chromatography using either 20% methanol (xanthine, monomethylxanthines, and theobromine) or 35% methanol (paraxanthine) in 0.5% acetic acid, ph 5 (with NaOH), at a flow rate of 1.2 ml min ~. The compounds collected were freezedried, redissolved in distilled water, and identified by co-chromatography with standards, by UVspectrophotometry, and by analysis by gas chromatography-mass spectrometry (GC-MS). The compounds were scanned between 210 and 350 nm in a DU-7 spectrophotometer (Beckman, Sweden) with distilled water as a blank. For GC-MS analysis, the compounds were butylated, silylated, and methylated. Dried samples were butylated with dimethylformamide di-n-butylacetal (Pierce, Sweden) for 1 h at 60 C. Dried samples were redissolved in pyridine and silylated with N-methyl-N-trimethylsilyltrifluoroacetamide containing 1% trimethylchlorosilane (MSTFA/TCMS, Pierce, Sweden) for 1.5 h at 80 C in a proportion of 1:1 (v/v). The samples were dried in a speed-vac (Heto, Denmark) and redissolved in pyridine for analysis. Methylation was performed at room temperature by addition of trimethylanilinium hydroxide (MethElute, Pierce, Sweden) to dried samples for 5 min. The compounds were separated in a Hewlett Packard 5890A gas chromatograph using a fused silica capillary column with a flow rate of 2 ml min -~ of He. The temperature in the column was held at 60 C for 5 rain then raised to 280 C at 10 C/min. Mass spectrometry was performed by using a Hewlett Packard 5970 series mass selective detector at a source temperature of 250 C and 70 ev. A Hewlett Packard 59970C work station was used for data acquisition. Cell Free Extracts Preparation and Assay for Dehydrogenase Activity Serratia marcescens was grown in liquid medium (1.5 liters) containing 500 nag ml ~ of caffeine and harvested by centrifugation at 7,000 rpm, 4 C, for 15 rain. The pellet was resuspended in 25 mm sodium phosphate buffer, ph 7.5, containing 10 mm 2-mercaptoethanol, and subjected to membrane disruption by ultrasonication (Soniprep 150, MSE Scientific Instruments). Four pulses of 15-s sonication with intermittent 5-rain cooling (4 C) were applied. Debris was eliminated by centrifugation at 14,000 rpm, 4 C, and the supernatant was filtered through Sephadex G25 PD- 10 columns (Pharmacia, Sweden). Proteins were eluted with 25 mm sodium phosphate buffer, ph 7.5, and the concentrations were determined with the Bradford [2J reagent. After electrophoresis dehydrogenase was assayed by active staining of native polyacrylamide gel electrophoresis gels (native-page). Proteins were separated in a Pharmacia Phast System Unit using minigels with 10-15% gradient of polyacrylamide. Protein solution was loaded in a continuous comb, and the electrophoresis was conducted according to the

4 202 R Mazzafera et al "~ ug/=z 300~g/=z e00ug/=l A 2, 4o ~ 2o Oi oo ~ zooo ~ - ~ _ ~ ~oo! C~ o o ~ B Fig. 1. Serratia marcescens growth. A, Growth in M9- caffeine liquid medium with the substrate in different concentrations. B, Growth in caffeine concentration in the medium during the incubation period. (The symbol key for B is the same as for A.) manufacturer's instructions. After electrophoresis the gel was cut into eight strips, and dehydrogenase enzyme activity was detected by reduction of nitro-blue tetrazolium salt [ 13]. Different methylxanthines were used as substrates for each gel strip. Results Active growth of bacteria and fungi was observed in caffeine plates receiving the soil filtrate. When transferred to a new set of caffeine plates, fungi could not be observed and most of the bacterial colonies did not grow, indicating that residues from the soil filtrate were supplying nutrients for growth in the first plating. Identification of the s. marcescens strain was easily made in the second set of plates because of its characteristic red-pink color. A selective caprylate-tallous agar medium was used to confirm the identity of the Serratia strain [12]. Caprylate is a selective carbon source for practically all strains of Serratia. The ability of the isolated strain to degrade caffeine was checked by further plating in M9-caffeine plates and inoculation in M9-caffeine liquid medium. The Serratia strain grew best in liquid medium containing 600 mg liter 1 of caffeine (Fig. 1). At the beginning of the incubation period, we observed a slight growth inhibition at 1,200 mg liter -L caffeine, which was probably due to a toxic effect of caffeine.

5 Degradation of Caffeine by Serratia 203 loo A,~ ~ 300ug/~ 2o ~ B c~ O i Fig. 2. A, Serratia marcescens growth in M9-theobromine liquid medium with the substrate in different concentrations. B, Theobromine concentration in the medium during the incubation period. (The symbol key for B is the same as for A.) Theobromine was degraded also by Serratia, and the best growth was observed at 300 mg liter -1 (Fig. 2). Although at day 3 for 100 mg liter -1 caffeine concentration, and at day 7 for 300 mg liter -1, caffeine was not detected in the medium (Fig. 1B), still, some bacterial growth was observed on subsequent days (Fig. 1A). A similar situation was detected for theobromine at the same concentrations (Fig. 2A,B). Analyses showed that catabolites of caffeine and theobromine were released in the medium (Fig. 3), and their reabsorption could explain the continued growth after the disappearance of the original substrates in the medium. When caffeine was used as substrate, paraxanthine and theobromine were released in the medium (Fig. 3A). 7-Methylxanthine and 3-methylxanthine were detected in M9-theobromine medium (Fig. 3B). Traces of xanthine were detected in the late incubation period when theobromine was used as substrate. Small amounts of xanthine and 7-methylxanthine were identified also in the liquid medium when paraxanthine was the substrate. Substrate concentration at 600 mg liter -~ was more adequate for facile detection of the released compounds. Support for the hypothesis of reuse of these compounds was found when other methylxanthines were used as substrates (Fig. 4). 7-Methylxanthine and paraxanthine, which accumulated in the medium to a lower extent when caffeine and theobromine were the substrates (Fig. 3), were a much better substrate than 3- methylxanthine (Fig. 4). The compounds released in the medium were identified by co-chromatography

6 204 P. Mazzafera et al. ~E 600 "~ 5oo! 400 -I 300,-I 201) 10 A ~ Caffeine ParaxantNne ~ Theobromine I ~ (~ o o + Theobrom,ne 3-methylxanthine 7-methylxanthine Fig. 3. Release of other methylxanthines when Serratia marcescens was incubated in M9 liquid medium with caffeine (A) and theobronaine (B) at 600 nag nal -~ as the only sources of carbon and nitrogen. in HPLC, by UV spectrophotometry and GC-MS (Table 1). Only theobromine and paraxanthine could be silylated for GC-MS analysis (Table 1). Xanthine, monomethylxanthines, and dimethylxanthines were all converted to caffeine when methylation was attempted. Caffeine and other methylxanthines except xanthine were successfully butylated. The observed spectra matched authentic standards. Although theophylline and 3-methylxanthine concentration remained constant when they were used as substrates (Fig. 4A), some growth of Serratia was observed with these methylxanthines (Fig. 4B). We believe that very small amounts of these compounds were used to support growth; however, this could not be detected, which was due to slight variations in their concentrations caused by medium evaporation. No growth was observed with 1-methylxanthine, while paraxanthine proved to be a substrate as good as caffeine for Serratia growth. Only after a long incubation period was some growth observed with xanthine as substrate. Native-PAGE gels revealed bacterial dehydrogenase activity toward all tested xanthines (results not shown). For all substrates, a single band, with the same Rf, was observed at the top of the gel. Nicotinamide Adenine Dinucleotide was essential for activity, indicating a NAD+-dependent xanthine dehydrogenase. Discussion Serratia grew in liquid culture containing caffeine, theobromine, paraxanthine, or 7-methylxanthine, with a common methyl group at the 7-position. Consequently,

7 Degradation of Caffeine by Serratia ' i ~ C A F A i 8o ~ 1~ jr.~ ~ 40 ~ '71~ o~ I i 100 5o U B Fig. 4. A, Serratia marcescens growth in M9 liquid medium with different methylxanthines as substrates at 200 mg ml concentration. B, Their concentrations in the medium during the incubation period. XAN, xanthine; 1MX, 1- methylxanthine; 3MX, 3- methylxanthine; 7MX, 7- methylxanthine; THF, theophyltine; PAR, paraxanthine; CAF, caffeine. (The symbol key for B is the same as for A.) caffeine degradation in this bacteria could follow demethylation either to theobromine or paraxanthine and subsequent demethylation to 7-methylxanthine and xanthine. Xanthine was then oxidized to uric acid. Support for the suggested pathway comes from the identification of paraxanthine and theobromine in liquid culture with caffeine as substrate and of 7-methylxanthine and xanthine when theobromine or paraxanthine was used as a substrate. A nonspecific demethylating enzyme system would account for both degradation of paraxanthine and theobromine. A proposed degradation pathway suggested for Pseudomonas putida is similar to that suggested for Serratia [1, 4]. Investigating caffeine degradation by P. putida, Woolfolk [15] suggested that besides N-demethylation, caffeine and related methylxanthines were converted to the respective methyluric acids. Various dehydrogenase activities were measured with cell free extracts by coupling the oxidation of the substrate to the reduction of ferricyanide. All substrates tested (caffeine, theobromine, theophylline, and 1, 3, and 7-methylxanthines) were oxidized by the protein preparations. Xanthine dehydrogenase showed no specificity for the substrates used in assays performed on native-page. As activity was also observed with theophylline, 1- methylxanthine, and 3-methylxanthine, for which no consistent bacterial degradation was observed in liquid culture, at least for these methylxanthines it seems unlikely that they were converted to their respective methyluric acids. A strong dehydrogenase activity for xanthine in Serratia cells could explain the lack of specificity for substrates, as suggested by Woolfolk [15]. Additional experiments

8 206 R Mazzafera et al. Table 1. Mass spectra of methylxanthines released in the liquid culture medium Compound Butylation Theobromine ~,,b Paraxanthine 7-Methylxanthine 3-Methylxanthine Methylation Xanthine c Silylation Theobromine Paraxanthine Major peaks (m/z) 55 (22) 67 (32) 82 (14) 109 (42) 137 (20) 180 (100) 194 (28) 207 (7) 219 (28) 236 (56) 68 (22) 81 (13) 95 (7) 109 (13) 123 (38) 136 (95) 150 (18) 165 (12) 180 (100) 194 (40) 219 (15) 236 (96) 67 (16) 82 (15) 95 (7) 108 (7) 123 (21) 136 (74) 150 (48) 166 (50) 180 (72) 193 (10) 205 (27) 222 (44) 235 (22) 249 (5) 261 (37) 278 (100) 68 (17) 81 (19) 95 (20) 109 (12) 123 (25) 136 (14) 152 (17) 166 (83) 180 (70) 193 (21) 207 (22) 222 (83) 235 (29) 249 (12) 261 (49) 278 (100) 55 (39) 67 (46) 82 (27) 109 (46) 137 (7) 165 (6) 194 (100) 55 (39) 67 (46) 82 (26) 109 (28) 137 (14) 180 (100) 53 (35) 68 (90) 82 (9) 95 (18) 123 (34) 150 (8) 180 (100) anumbers in parenthesis correspond to relative abundance. bthe last number of the sequence is the molecular ion (M+). CAnalyzed as caffeine. are necessary to ascertain the participation of methyluric acids in caffeine degradation by S. marcescens. Many studies have indicated caffeine as an allelopathic compound [7, 8, 14]. However, no attention has been paid for transformation of caffeine by soil microorganisms. Because readily usable nonhumic substrates are scarce in the soil, keeping microorganisms under constant starvation [6], it would not be rare to find bacteria degrading caffeine in soil under coffee cultivation. Therefore, the presence of soil bacteria that could metabolize caffeine may alter the allelopathic potential of this compound in the soil near coffee trees. Acknowledgments. P. Mazzafera thanks CNPq, Brazil, for a fellowship during a leave of absence at The Swedish University of Agricultural Sciences, Ume& Sweden. References 1. Blecher R, Lingens F (1977) The metabolism of caffeine by a Pseudomonas putida strain. Hoppe- Seyler's Z Physiol Chem Bd 358: Bradford MN (1976) A rapid and sensitive method for the Quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: Friedman J, Waller GR (1985) Allelopathy and autotoxicity. TIBS 47-50

9 Degradation of Caffeine by Serratia Gluck M, Lingens F (1987) Studies on the microbial production of theobromine and heteroxanthine from caffeine. Appl Microbiol Biotechnol 25: Mimura A, Akimoto T, Kodaira R (1969) Transformation of an organic compound using microorganisms; on the formation of 6,8-dihydroxypurine. J Ferment Technol 47: Nelson DW, Sommres LE (1982) Total carbon, organic carbon, and organic matter. In: Page AL (ed). Methods of soil analysis. Agronomy monograph No. 9, Part 2. Chemical and microbiological properties. American Society of Agronomy and Soil Science Society of America, Madison, Wisconsin, pp Rizvi SJH, Mukerji D, Mathur SN (1980) A new report on a possible source of natural herbicide. Indian J Exp Biol 18: Rizvi SJH, Mukerji D, Mathur SN (1981) Selective phyto-toxicity of 1,3,7-trimethylxanthine between Phaseolus mungo and some weeks. Agric Biol Chem 45: Rouf MA, Lomprey Jr RF (1968) Degradation of uric acid by certain aerobic bacteria. J Bacteriol 96: Sambrook J, Fritscb EF, Maniatis T (1989) Molecular cloning: a laboratory manual, vol. 1, appendix A, 2nd ed. Cold Spring Harbor Laboratory Press, New York 11. Schnitzer M (1982) Organic matter characterization. In: Miller RH, Keeney DR (eds), Methods of soil analysis, vol. 9, part 2. Chemical and microbiological properties. American Society of Agronomy and Soil Science Society of America, Madison, Wisconsin, pp Starr MR Grimont PAD, Grimont E Starr PB (1976) Caprylate-thallous agar medium for selectively isolating Serratia and its utility in the clinical laboratory. J Clin Microbiol 4: Vallejos CE (1983) Enzyme activity staining. In: Tanksley SD, Orton TJ (ed), Isozymes in plant genetics and breeding, part A. Elsevier Science Publishers Amsterdam, pp Waller GR, Kumari D, Friedman J, Friedman N, Chou C-H (1986) Caffeine autotoxicity in Coffea arabica L. In: Putnam AR, Tang C-S (ed), The science of allelopathy. John Wiley & Sons. New York, pp Woolfolk CA (1975) Metabolism of N-methylpurines by a Pseudomonas putida strain isolated by enrichment on the caffeine as sole source of carbon and nitrogen. J Bacteriol t23: