Degradation and product analysis of caffeine and related dimethylxanthines by filamentous fungi

Transcription

1 1 Degradation and prodct analysis of caffeine and related dimethylxanthines by filamentos fngi M. Hakil," S. Denis: G. Viniegra-González: and *RSTM, nstitt Français de Recherche Scientifiqe por le Développement en Cooperation, RSTM-Mexiqe, Los Morales, México 'Departamento de Biotecnología, División de Ciencias *A Biológicas y de la Sald, Universidad Atónoma Metropolitana, ztapalapa, MéYíco Twenty strains offilamentos fngi were tested for their ability to grow on a liqid medim containing caffeine (1,3,7-trimethylxanthine) as a sole nitrogen sorce. Seven strains which were able to grow on hffeine belong to the Aspergills and Penicillim genses. They all presented the same degradation pathways 'bt varied in the eficiency by which degradation took place. Caffeine was first demethylated in position 7; this lead to the formation of theophylline (1,3-dimetliylxaiitliine). This compond was next demethylated in position to give 3-metliylxanthine. All these strains were also able to grow on theophylline, theobromine (3,7-diniethylxanthine), and paraxanthine (1,7-dimethylxanthiiie) as a sole nitrogen sorce. All degradation intermediates detected were metliylxanthines lsevier Science nc. L.1.,, Keywords: Caffeine; theophylline; theobromine; paraxanthine; degradation; filamentos fngi._.., :.:?*cl i ( ntrodction Coffee plp is one of the major agro-indstrial wastes prodced dring the plping operations of the coffee cherry to obtain coffee bean in many prodction areas of the tropics.' For the season, 2.8 million tons of coffee plp were prodced in the world.' Almost all the plp prodced each year is discarded as a waste prodct and is considered as the most abndant polltant material of lakes and rivers located near coffee processing sites. Coffee plp is ;ch in carbohydrates and prote in^^.^ bt the presence of antintritional factors sch as caffeine (1,3,7-trimethylxanthine, Table ), polyphenols, and tannins restricted its se as animal food. Attempts to redce caffeine levels in coffee plp by solid-state fermentation sing a Penicillini strain have been sccessf~l~~~~ bt no stdies abot the degradation prodcts have been performed. There are few pblications abot caffeine degradation in filamentos fngi; nevertheless, it has been shown in a strain of Aspergills niger' and in a Penicillim roqefortii' that the first prodct to appear from degradation of caffeine Address reprint reqests to Dr. C. Agr, nstitt Français de Recherche Scientifiqe por le D6veloppement en Cooperation, RSTM-Mexiqe, Ciceron 609, Los Morales, 1530 México D.F., México Received 13 Jn 1997; revised 19 September 1997; accepted 19 September 1997 in liqid cltres is theophylline (1,3-dimethylxanthine). This compond can have greater toxic and pharmacological effects than ~affeine.~ This cold also be a problem in trials to obtain a decaffeinated coffee throgh se of filamentos fngi. The prpose of this stdy was to investigate whether the caffeine degradation pathway was the same for filamentos fngi in general. We assessed the ability of several filamentos fngi to degrade caffeine as well as the corresponding dimethylxanthines. We tested 20 different strains for their ability to se caffeine as the sole sorce of nitrogen in liqid media. The kinetics of caffeine and related dimethylxanthine degradation were established for all caffeine-degrading strains. n each experiment, degradation prodcts were characterized. Fo 1 ds QSC W Tlentai re 0 RST. Gof& :A 9 AS 993 X:^ Materials and me o s. Microorganisins All fngal strains sed are from the RSTM-UAM collection.'' Cltre media All microorganisms were maintained on coffee infsion medim (CS) prepared as follows. scrose, 2.0 g; KH,P04, nzyme and Microbial Technology 22: ,> lsevier Science nc. All rights reserved, 655 Avene of the Americas, New York, NY loplo /98/$19.00 P S (97)

2 ~ Papers R3 Table 1 Caffeine and derived componds Prines 1,3,7-Trimethylxanthine (caffeine) 1.3-Dimethylxanthine (theophylline) 1,7-Dimethylxanthine (paraxanthine) 3,7-Dimethylxanthine (theobromine) -Methylxanthine 3-Methylxanthine 7-Methylxanthine Xanthine R R3 R7 CH, CH, CH, CH, CH, H CH, H CH3 H CH3 CH3 CH, H H H CH, H H H CH, H H H 1.3 g; Na,HP, - 2H,.. 12 g; MgS, 7H,, 0.3 : and CaC, * 3H,, 0.3 g were dissolved in an infsion of commercial grond coffee (Grand Mère familial, café Grand Mère S.A.. wattinities, France) in distilled water. The ph was adjsted to 5.6 with 1 M KH and the volme was broght to 1. The medim was sterilized at 121 C for 20 min after spplementation with 20 g of agar. Liqid, Cltres were performed in a caffeine-scrose medim (CS) prepared as follows. The chosen niethylxanthine, 1 g: scrose, 28.4 g; KH,P,, 1.3 g; Na,HP, * 2H,,. 11 g; MgS, 7H20, 0.3 g and trace elements were dissolved in water. The ph was adjsted to 4 by addition of 1 M H,S, and the final volme was broght to 1. The medim was sterilized at 121 C for 20 min. CLr1tLir.e.r Spores of 6-day-old cltres on CS were harvested with 30 ml of distilled water containing 0.2% Tween 80 (w/v) nder agitation provided by magnetic stir bar ( 1 0 rpm). rlenmeyer flasks cpntaining 50 ml of liqid media were inoclated with thè spore sspension at 6 sp ml-. Flasks were incpated at 27 C with an &tation of 160 rpm. n the case of caffeine degradation experiments, after 48 h of cltre, 0.2 g of anhydros caffeine previosly heated to 100 C was added to each flask. This moment was chosen as the initial time for HPLC analysis. Samples analyzed by HPLC were obtained as follows. A sspension cltre (300 pl) was homogeneosly removed every 12 h from each incbated flask. Biomass was then discarded from the cltre medim by centrifgation followed by filtration (0.45 pm). Filtered samples were stored at -20 C ntil analysis. H PL C A 17 ci lwis nj me thy lxnn th in es Chromatographic conditions. HPLC analyses were performed on ii Beckman GLD system with a l 6 pmp, a 507 atosampler with a 100 pl injection loop, and a 168 detector. Componds were separated on ii C8 5 pni Ultrasphere colmn ( 1.6 mm X 25 cm) at room temperatre with a flow rate of 2 mi min- by a gradient eltion system. The elents were 1.75 mm KH,P,, acetonitrile, and tetrahydrofran (98: 1 : 1 ; v/v/v) for pmp A. Acetonitrile and tetrahydrofran were sed (99:l ;v/v) for pmp B. Solvent A was sed first for 7 min before starting a linear gradient reaching 20% of solvent B in 5 min. Methylxanthines were identified by appropriate comparison of their retention times to those of the pre prodcts and qantified by coniparison to standard crves established for each methylxanthine. Sample preparation. Samples were dilted in Solvent A in order to obtain a concentration in the range of the standard crves of each methylxanthine. Reslts t of 20 strains tested for their ability to grow in a medim containing caffeine as the sole nitrogen sorce (CS medim), only seven strains were able to grow (TiiDle 2). Their ability to degrade caffeine was checked by HPLC analysis of the cltre spernatant. n each case, there was caffeine degradation only if fngal growth was observed. The seven ~ strains were not able to grow when caffeine was sed as a ] y sole sorce of carbon and nitrogen. n later experiments. only the seven strains able to grow on the CS-medim were sed. The strains were qantitatively tested for their ability to degrade caffeine. theophylline. theobromine. and paraxanthine as a sole nitrogen sorce. Cq ffe in e [leg mdci t ior 7 We sed a high concentration of caffeine in caffeine degradation experiments (4 g - after 48 h of cltre at g - ) to be able to see as many intermediates as possible. Asper-gillirs ramcirii (V l2a25) and all Penicilliirrii ~orizriiiine strains showed a good ability to degrade caffeine whereas A. iziger (C28B25) and Aspergilliis fiirnignts (C25 A35) were less effective (Tcrhlr 2, Figre ). The A. /Tiger degradation profile (C28B25) is the same as that of A.,firi?zigats (C25A35) (Ficsirrr 2B). The P. comzne degradation profile (V14A35) (Fi<yirre 2A) is representative of all other strains except Aspergills oryzae. n this case, biomass prodction was higher than with other strains, scrose disappeared from ; the cltre medim after 48 h of cltre, and no frther degradation was noted at this incbation time (Figre 1). Table 2 Caffeine degradation by the selected strains Code % Degradation V mean nmber G. species at 48 h (mg/- h- ) V12A25 A. tamarii V33A25 P. commne C25A35 A. fmigats V29A25 P. commne V14A35 P. commne C28825 A. niger C7A25 P. commne nzyme Microb. Technol., 1998, vol. 22, April

3 Caffeine degradation by fngi: M. Hakil et al..m i re in. nd d A :nt :re es on in rd m ìi- :ir ;is ne -n - a4 W a- /- : l ne g e. lie A. re in 11s le 11s ln er ti s l Figre 1 Caffeine degradation by the seven selected strains. C28B25 (); V12A25 (+); C25A35 (A); V14A35 (X); V29A25 (*); C7A35 (t); and V33A Figre 2A shows clearly that theophylline appeared first followed by 3-methylxanthine. These two componds represented p to 60% of initial caffeine present dring caffeine degradation by P. ~~ZZLZ~ (V 14A35). Theobromine and paraxanthine were detected in the cltre media at concentrations of abot 40 mg -'. Trace amonts of 1 -methylxanthine were also detected. n all HPLC assays, no trirnethylric acid or dimethylric acids was detected. DinzetlzZilxniztlziiie degmdntion The lower concentration sed in dimethylxanthine degradation experiments (1 g 1-' for all cltres) was de to the lower solbility of dimethylxanthines. The amont of each degradation prodct detected in these experiments was sally less than 20 mg -'. With the initial qantity of each dimethylxanthine being low, the intermediary prodcts are degraded as they appear. All strains except A. rcminrii seemed to be more efficient in degrading theobromine than theophylline. n all cases, paraxanthine was the least easily degradable dimethylxanthine (Figre 3C). For all strains, degradation of theobromine and theophylline led to the appearance of 3-methylxanthine. Degradation of paraxanthine led to the formation of - and 7-methylxanthine. These prodcts were only present in trace amonts (data not shown) Discssion Twenty different strains of filamentos fngi were tested for their ability to se caffeine as well as different dimethylxanthines (Tble!) as a sole nitrogen sorce. nly Aspergills and Perzicillim genses were able to grow on caffeine as a sole nitrogen sorce. All Perzicilliiirii strains tested were P. corizrizrie and were able to grow. For Aspergills, only for strains ot of nine were able to grow. t appears that Periicillirii and A.spergil1.s are the more freqent caffeine-degrading genses. t therefore seems logical that the majority of the stdies done on caffeine degradation by filanientos fngi are related to Aspergills and Perzicilli~rm genses. Krtzman and Schwimmer' ' isolated a strain of Stemph?'lirii able to degrade caffeine which sggests that this characteristic is not restricted to Aspergillirs and Perzicilliriz. n addition, not all Aspergilli are able to degrade caffeine. Similar stdies have been ndertaken by Bchanan et cil." This capacity is therefore a cha'racter$ic of a strain bt not of a gens. The Fsnri~n strains tested did not seem to be very efficisnt in caffeine degradation. None of the fngal strains tes.ted were able to grow when caffeine was sed as the sole sorce of both carbon and nitrogen. t has been reported that on the contrary, specific bacteria can grow nder the former conditions.13"j This particlarly sggests enzymatic differences between sch microorganisms. The seven selected strains do not have the same caffeinedegrading efficiency; nevertheless, they all yield the same degradation prodcts which are theophylline and 3-methylxanthine. This cold be de to two sccessive demethylations. First, a 7-demethylation, giving theophylline from caffeine, followed by a 1 -demethylation leads.to 3-methylxanthine from theophylline. This hypothesis is spported by data presented in Figre 2 where theophylline appears first after caffeine degradation. 3-Methylxanthine appears only when theophylline is present in the cltre medim. Spport for this hypothesis also comes from Schwimmer er al.' who identified theophylline as the first prodct in the caffeine degradation pathway by a Periicillirk strain. Figre 3 shows that the sin of theophylline and 3-methylxanthine represents more than 50% of the initial caffeine concentration after 110 h of cltre. This sggests that other degradation pathways are not qantitatively important. The small f, & U $ 1000 (A) !s S i (B) ----= Figre 2 Caffeine degradation and appearance of prodcts. V14A35 (A); C25A35 (B); caffeine (e); theophylline (D); and 3-methylxanthine (A) nzyme Microb. Technol., 1998, vol. 22, April 357

4 Papers 2 M = 200 s (A) ci M = 200 s ( B) M = ( Cl Figre 3 Dimethylxanthines degradation by the seven selected strains. nitial sbstrates were: theophylline (A); theobromine (B); and paraxanthine (C). Strains tested were: C28825 (]; V12A25 (+; C25A35 (A); V14A35 (x); V29A25 (*; C7A25 (+; and V33A25 (H) amonts of paraxanthine and theobromine also present in the cltdre medim can be de to an nspecific reaction of the,$ame enzyme which catdyzes the transformation of caffeine into theophylline. t cold also be de to specific enzymes which cold be weakly expressed or that present a weak affinity for the sbstrate. n bacteria, the major intermediaries in caffeine degradation are theobromine, 7-methylxanthine, and ~anthine'~-'~ which is another difference between bacterial and fngal systems. Althogh prodced at a slow rate, theobromine and paraxanthine can be sbseqently degraded by all strains. Theobromine is even more effectively degraded than theophylline. Theobromine and theophylline degradation leads to 3-methylxanthine formation (7- and 1 -demethylation, respectively). Paraxanthine degradation leads to formation of 1- and 7-methylxanthine ( 7- and l-demethylation). Formation of trimethylric acid and dimethy- lric acids were never detected regardless of the sbstrates sed. From all these experiments, we can conclde that the first steps of caffeine degradation in filamentos fngi consist of demethylation reactions. Demethylations in position 1 and 7 can occr each time a methyl grop is present in this position bt the 7-demethylation seems to be preferentially sed. ha7 identified xanthine as a degradation prodct of caffeine by a strain of A. niger. t wold be interesting to frther identify the prodcts reslting from caffeine degradation of or seven strains. t is difficlt to say whether caffeine degradation in filamentos fngi is de to one or more enzymes and whether these enzymes are the same as thoste present in bacteria. n bacteria, caffeine degradation is lead by at least two enzymes and one of these enzymes has been prified." Answers to these qestions cold come from attempts presently made to prify the enzyme(s) implicated in the first steps of caffeine degradation. Acknowledgments The present work was performed as part of a cooperative agreement between the Consejo Nacional de Ciencia y. Tecnologia (CNACyT, Mexico) and the nstitt Franpis!(., de Recherche Scientifiqe por le Développement en Cooperation (RSTM, France) within a specific program ndertaken at the Universidad Atonoma Metropolitana (Mexico). References Zlaga. J. Contribtion à 'étde de la composition chimiqe de la plpe de café (Coffea arabicn L.). Thèse de Doctorat es Science. Faclté des Sciences, Université de Nechatel, Sisse, Perrad-Gaime, 1. Cltres mixtes en milie solide de bactéries lactiqes et de champignons filamentex por la conservation et la decafeination de la plpe de café. Thèse de doctorat, Université de Montpellier. France, lias, L. G. Composición qímica de la plpa de café y otros sbprodctos. n: Prrlp de Café: Co~nposición, Tecnología y Utilizacióiz (Braham, J.R. and Bressani, R., ds.). NCAP Gatemala City 1978; Zlaga, J.,Bonilla, C.. and Qijano. R. M. Contribción al estdio y tilización de la plpa de café. n: 7ème Col. ntern. Qim. Café, Hambrg (ASC Costes, A d.). Paris, 1975, Rossos, S., Aqiáhatl, M. A., Trejo-Hernandez, M. R., Gaime!!' Perrad,., Favela,., Ramakrishna, M., Raimbalt, M.. and -+' Viniegra-Gonzáles, G. Biotecnological managment of coffee plp - isolation, screening, characterisation, selection of caffeine-degrading fngi and natral microtlora present in coffee plp and hsk. Appl. Microbiol. Biotechizol. 1995, 42, Rossos, S., Hannibal,.L., Aqiáhatl, M. A., Trejo-Hernandez, M. R., and Marakis. S. Caffeine degradation by Penicillirrm verrcos~~m in solid state fermentation of coffee plp: Critical effect of additional inorganic and organic nitrogen sorce. J. Food Sci. Techwl. 1994, 31 (4), na, K. Biochemical stdies of caffeine. Degradatio'n of caffeine by. mold. Nippo!~ Nogeikngakn Knishi 1971, 45 (8) Schwimmer. S.. Krtzman. R. H., and Heftmann,. Caffeine metabolism by Penicil/irrr~ roy~reforti. Arch. Binchem. Biophjs. 1971, 147, Starvic. B. Methylxanthines: Toxicity to hmans. 2. Caffeine. Fd. Chein. Tos (7), Aqiahatl, M. A.. Rainibalt. M.. Rossos, S., and Trejo, M. R. Coffee plp detoxification by solid state fermentation: solation, 1 ' 358 nzyme Microb. Technol., 1998, vol. 22, April

5 ~~ ~.b- the i Caffeine degradation by fngi: M. Hakil et al. identification, and physiological stdies. n: Solid State Fermerira- theobromine from caffeine. Biosci. Biotecknol. Biochem. 1993, 57 tion in Bioconversion of Agroindstrial Raw Materials (Raimbalt, (8) M., d.) RSTM, Montpellier, France, Blecher, R. and Lingens, F. The metabolism of caffeine by a Krtzman, R. H. and Schwimmer, S. Caffeine removal from growth Psedomonas ptida stain. Hoppe-Seyler's 2. Physiol. Cliein. 1977, media by microorganisms. xperiencia 1971, 127 (4), , Bchanan, R. L. and Fletcher, A. M. Methylxanthine inhibition of 16. Glück, M. and Lingens, F. Stdies on the microbial prodction of aflatoxin prodction. J. Food Sci. 1978, 43 (2), theobromine and heteroxanthine from caffeine. Appl. Microbiol. Hohnloser, W., sswald, B., and Lingens, F. nzymological aspects Biotechnol. 1987, 25, of caffeine demethylation and formaldehyde oxidation by Psedo- 17. Asano, Y., Komeda T., and Yamada, H. nzymes involved in nionas ptida C1. Hoppe-Seyler's 2. Pliysiol. Clienz , theobromine prodction from caffeine by Psedornonas prid No Biosci. Biotechnol. Biocliem. 1994, 58 (12), Asano, Y., Komeda, T., and Yamada, H. Microbial prodction of A,,.. ' í i t! nzyme Microb. Technol., 1998, vol. 22, April 359

6 . %: