Further Investigation of Microbial Degradation of Microcystin Using the

Jan 15, 2009 - Present affiliation: Department of Food Technology, Technological Federal University of Parana, P.O. Box 157, Zip Code 85660-000, Dois ...
0 downloads 15 Views 1019KB Size
Download PDF
Chem. Res. Toxicol. 2009, 22, 391–398

391

Further Investigation of Microbial Degradation of Microcystin Using the Advanced Marfey Method Elisabete Hiromi Hashimoto,†,‡ Hajime Kato,† Yoshito Kawasaki,† Yuriko Nozawa,† Kiyomi Tsuji,§ Elisa Yoko Hirooka,| and Ken-ichi Harada*,† Graduate School of EnVironmental and Human Science and Faculty of Pharmacy, Meijo UniVersity, Tempaku, Nagoya 468-8503, Japan, Kanagawa Prefectural Institute of Public Health, Shimomachiya, Chigasaki, Kanagawa 253-0087, Japan, and Department of Food Technology, State UniVersity of Londrina, P.O. Box 6001, Zip Code 86051-990, Londrina, Parana´, Brazil ReceiVed September 22, 2008

It is known that microcystin (MC) is subject to microbial degradation to provide three types of products, linearized MCLR (Adda-Glu-Mdha-Ala-Leu-MeAsp-Arg), tetrapeptide Adda-Glu-Mdha-Ala, and Adda. They can be readily detected by the usual HPLC, because they commonly have an Adda moiety with a diene and an absorption maximum at 238 nm as the chromophore. However, no other degradation products without such a chromophore have been isolated to date. In this study, cell preparation of a bacterium B-9 that can degrade MC and detection of the degradation products were devised. First, we regulated the B-9 hydrolytic activity by washing with sodium chloride solution to obtain a desired cell preparation, which permitted an additional intermediate and the final products of MCLR to be obtained. Second, the resulting products could be firmly identified using the advanced Marfey method with the aid of log D. As a result of these experiments, the following degradation products were further identified: a tetrapeptide, AddaGlu-Mdha-Ala, tripeptides Adda-Glu-Mdha, Glu-Mdha-Ala, and Arg-MeAsp-Leu, a dipeptide, Glu-Mdha, and amino acids Adda, Arg, and methylamine derived from Mdha. The present study expands the hydrolytic activity of the B-9 strain, which can hydrolyze not only cyanobacterial cyclic peptides but also MC to the intermediates and final products. The established characterization method composed of the advanced Marfey method and log D would be a standard technique for the structural characterization of a mixture of amino acids and peptides. Introduction Microcystin (MC),1 the cyclic heptapeptide toxin produced by freshwater cyanobacteria such as Microcystis, Anabaena, Nostoc, and Planktothrix, shows potent hepatotoxicity and tumor-promoting activity through inhibition of protein phosphatases 1 and 2A (1, 2). They consist of a cyclo(Ala-R1MeAsp-iso-R2-Adda-Glu-iso-Mdha) structure, and over 80 variants have been reported to date (3, 4). Among the variants fully identified, microcystin-LR (MCLR; R1 ) Leu; R2 ) Arg) is the most toxic and has been used as the standard variant in various experiments (1). MCs are very stable and resistant to enzymatic hydrolysis by the usual proteases such as trypsin, probably due to their cyclic structure (5-7). A toxic incident of 50 people’s deaths occurred in Brazil in 1996 due to MC in the water used for hemodialysis (8, 9). Now MCs are threatening human health and life, and many problems associated with MC * To whom correspondence should be addressed. Phone: +81-52-8392720. Fax: +81-52-834-8090. E-mail: [email protected]. † Meijo University. ‡ Present affiliation: Department of Food Technology, Technological Federal University of Parana, P.O. Box 157, Zip Code 85660-000, Dois Vizinhos, Parana´, Brazil. § Kanagawa Prefectural Institute of Public Health. | State University of Londrina. 1 Abbreviations: MC(s), microcystin(s); LC/ITMS, liquid chromatography/ion trap tandem mass spectrometry; MeAsp, erythro-β-methyl-D-aspartic acid; Adda, (2S,3S,8S,9S)-3-amino-9-methoxy-2,4,6-trimethyl-10-phenyldeca-4(E),6(E)-dienoic acid; Mdha, N-methyldehydroalanine; MCLR, microcystin-LR; ESI, electrospray ionization; EDTA, ethylenediaminetetraacetic acid; SIM, selected ion monitoring; PMSF, phenylmethanesulfonyl fluoride; FDLA, 1-fluoro-(2,4-dinitrophenyl)-5-leucinamide; DLA, (2,4dinitrophenyl)-5-leucinamide.

remain unsolved. To avoid these tragic incidents, we must develop an effective method for regulation of the occurrence of cyanobacteria and their toxins. In the environment, there are many bacteria with degradation activity against MC. Such an MC-degrading bacterium was first isolated and identified to be one of the Sphingomonas strains (ACM-3962) in 1994 (10). Phenotypically, similar bacteria capable of degrading MC have been subsequently reported (11-16). The strain B-9, which is 99% similar to the Sphingosinicella microcystiniVorans strain Y2 on the basis of the 16S rDNA sequence (GenBank accession no. AB084247) (17) showed a promising potential for the degradation of MCLR, MCRR, 3-desmethyl-MCLR, dihydro-MCLR, and nodularin (18) and nontoxic cyanobacterial cyclic peptides (19). In a pilotscale study, the ACM-3962 effectively degraded MCLR in slow sand filtration (20), and a feasible bioreactor using a B-9 strain immobilized in polyester resin rapidly removed MC from lake water (21). A molecular study of Sphingomonas strain ACM-3962 revealed the presence of three hydrolytic enzymes involved in MCLR degradation. Microcystinase (MlrA) catalyzes the initial ring opening of MCLR at the Adda-Arg peptide bond to give linearized MCLR, which is further degraded to a tetrapeptide by MlrB. The third enzyme MlrC hydrolyzes the tetrapeptide to smaller peptides and amino acids. The putative protein MlrD provided the transport of MCLR and its degradation products across the bacterial cell wall (22, 23). Effective use of an appropriate protease inhibitor could accumulate the following degradation intermediates: the use of EDTA (ethylenediamine-

10.1021/tx8003517 CCC: $40.75  2009 American Chemical Society Published on Web 01/15/2009

392

Chem. Res. Toxicol., Vol. 22, No. 2, 2009

tetraacetic acid) accumulated the linearized MCLR (Adda-GluMdha-Ala-Leu-MeAsp-Arg) and tetrapeptide Adda-Glu-MdhaAla, which allowed classification of MlrA and MlrC as two metalloproteases, whereas the use of 1,10-phenanthroline characterized MlrB as a possible serine protease (22, 23). Further, intact Adda was first isolated from MCLR by degradation using a B-9 strain as one of the final products (14). As mentioned above, there have been three intermediates of MC, linearized MC, tetrapeptide, and Adda, and a degradation scheme has been proposed (14). They can be readily detected by the usual HPLC, because they commonly have an Adda moiety with a diene and an absorption maximum at 238 nm as the chromophore (24, 25). To date, no other degradation products without such a chromophore have been isolated. There are two problems to be overcome for the complete identification of the microbial degradation products. One is to devise a cell preparation of the MC-degrading bacterium that permits control of the degradation, and another is to develop an analysis method that can detect and identify the degradation products without unusual amino acid and/or peptide standards. In this study, three cell preparations of the B-9 strain were devised, and the advanced Marfey method was applied for the detection and identification of the degradation products, because this involves nonempirical liquid chromatography/mass spectrometry (LC/MS) to determine the absolute configuration of an amino acid in a peptide without a standard (26, 27). Using both the devised cell preparation and the advanced Marfey method, the MCLR degradation behavior was carefully investigated.

Experimental Procedures Chemicals. Acetonitrile, formic acid, HPLC-grade methanol, NaCl, and Tris-Cl buffer (pH 7.6) were purchased from Nacalai Tesque (Kyoto, Japan). Ultrapure (distilled) water used in the mobile phases was purchased from Wako Pure Chemical (Osaka, Japan) and Kanto Chemical (Tokyo, Japan). As a derivatization reagent, 1-fluoro-(2,4-dinitrophenyl)-5-L-leucinamide (L-FDLA) was purchased from Kokusan Chemicals (Tokyo, Japan). Amino acid standards D-Ala, L-Arg, D-Glu, L-Leu, and MeAsp were purchased from Sigma (St. Louis, MO), Wako Pure Chemical, Tokyo Chemical Industry (Tokyo, Japan), and Nacalai Tesque. Purified Adda was produced by MCLR degradation using the cell extract of the B-9 strain (14). MCLR was isolated and purified (>95% on HPLC) from lyophilized cyanobacterial cells collected from Laguna de Bay, Philippines. Preparation of the Degradation Assay System from an MC-Degrading Bacterium. The bacterial strain B-9 (isolated from Lake Tsukui, Kanagawa, Japan) was inoculated in 100 mL of Sakurai medium (0.2 g of peptone, 0.1 g of yeast extract, and 0.05 g of glucose) and incubated at 27 °C for 3 days at 200 rpm (21). The cells were harvested (3000g, 30 min, room temperature), and a pellet was suspended in 50 mM Tris-Cl buffer (pH 7.6) and again centrifuged (3000g, 30 min). The degradation assay was carried out with cell pellets as (1) a crude B-9 cell (CB9), MCLR added directly to the freshly harvested pellets, (2) a washed B-9 cell (WB9), MCLR added into the pellet previously rinsed three times with 50 mM Tris-Cl buffer (3000g, 30 min), followed by sequential rinsing with 0.9% NaCl at 7500g, 12000g, and 15000g, 5 min (three repetitions each), and (3) a lyophilized B-9 cell (LB9), MCLR added into the crude B-9 cell pellet lyophilized and dissolved in 50 mM Tris-Cl buffer. The viability of CB9 and WB9 was examined by inoculating 10 µL of cell suspension in 10 mL of Sakurai medium at 27 °C for 24-48 h. A biodegradation experiment for CB9, WB9, and LB9 performed without MCLR was used as the negative control. Advanced Marfey Method. The advanced Marfey method used a 50 µM concentration of each of the pure D-, L-, and DL-amino

Hashimoto et al. acids Ala, Arg, Glu, Leu, and MeAsp and Adda (26, 27). The detection limit was established by comparing 0.05, 0.1, 0.5, 5, and 50 µM concentrations of the amino acid pools D-Glu, D-Ala, MeAsp, L-Arg, L-Leu, and Adda in 50 mM Tris-Cl buffer or H2O. Each amino acid pool was added to 20 µL of 1 M NaHCO3 and 50 µL of 1% L-FDLA in acetone, followed by vortexing and incubating at 37 °C for 60 min. The reaction was quenched by adding 20 µL of 1 M HCl and the material dried-centrifuged. The residue was dissolved in 300 µL of acetone and centrifuged to precipitate NaHCO3 (three repetitions, 15000g, 2 min). The collected supernatant (900 µL) was dried, and the residue was dissolved in acetonitrile (1 mL). A 5 µL sample of the solution was screened by TLC (thin-layer chromatography; 20 × 20 cm silica gel 60F254, 1 mm, Merck, Frankfurt, Germany) using chloroform/methanol (1: 1) as the mobile phase, while it was analyzed by LC/ITMS (liquid chromatography/ion trap mass spectrometry). The performance of the advanced Marfey method in the analysis of MCLR biodegradation products was evaluated using 100 µg/mL and 1 mg/mL MCLR incubated with the lyophilized B-9 cell (LB9) at 27 °C in the dark. A 120 µL volume of the mixture was sampled after 0, 24, 48, and 72 h, followed by addition of 120 µL of methanol to quench the degradation reaction and then stored at -20 °C. A 40 µL sample of the solution was used for HPLC-PDA analysis at 238 nm, while the remaining 200 µL was dried, dissolved in 50 µL of H2O, derivatized with L-FDLA, and analyzed by LC/ITMS as described above. MCLR Biodegradation Assay. Biodegradation of 1 mg/mL MCLR in H2O was carried out using 200 µL of bacterial strain B-9 (crude cell, CB9; washed cell, WB9; lyophilized cell, LB9) with incubation at 27 °C in the dark. An aliquot of 120 µL was sampled after 0, 6, 12, 24, 36, 48, 72, and 96 h and filtered by an Ultrafree-MC membrane centrifuge-filtration unit (hydrophilic PTFE, 0.20 µm, Millipore, Bedford, MA). Biodegradation was quenched with 120 µL of methanol. A 40 µL sample was analyzed by HPLC-PDA (238 nm) for biodegradation monitoring, followed by identification of the biodegradation products by LC/ITMS (238 nm). The remaining 200 µL was dried, and the residue was dissolved in 50 µL of H2O to perform the advanced Marfey derivatization with L-FDLA as previously described. The derivatized residue was dissolved in acetone and centrifuged, and the dried supernatant residue was dissolved in 100 µL of acetonitrile for LC/ ITMS analysis (340 nm). High-Performance Liquid Chromatography. The MCLR degradation process was monitored by HPLC-PDA (238 nm). The system consisted of a pair of LC 10AD VP pumps, a DGU 12A degasser, a CTO 6A column oven, an SPD 10A VP photodiode array detector, and an SCL 10A VP system controller (Shimadzu, Kyoto, Japan). The sample was filtered through the Ultrafree-MC membrane centrifuge-filtration unit (hydrophilic PTFE, 0.20 µm, Millipore), and 5 µL of filtrate was loaded onto a TSK-gel Super ODS column (2.0 µm, 2.0 × 100 mm, TOSOH, Tokyo, Japan) at 40 °C. The mobile phase was 0.1% formic acid in water (A) and 0.1% formic acid in methanol (B). The gradient conditions were initially 40-90% B at 20 min, and the flow rate was 200 µL/min. Liquid Chromatography/Ion Trap Mass Spectrometry. Prior to L-FDLA derivatization, cyclic MCLR, linear MCLR, tetrapeptide Adda-Glu-Mdha-Ala, and Adda were confirmed by LC/MS (238 nm). The sample, column, mobile phase, and gradient conditions were the same as those for HPLC analysis. The L-FDLA-derivatized sample was analyzed by LC/ITMS. The LC separation was performed by an Agilent 1100 HPLC system (Agilent Technologies, Palo Alto, CA). A 5 µL volume of the filtrated sample was loaded onto a TSK-gel Super ODS column (2.0 µm, 2.0 × 100 mm) at 40 °C. The mobile phase was 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). The flow rate was 200 µL/min with UV detection at 340 nm. The gradient conditions were initially 10-90% B at 40 min. The entire eluate was directed into the mass spectrometer where it was diverted to waste for 2.5 min after injection to avoid any introduction of salts into the ion source. The MS analysis was performed using a Finningan LCQ Deca XP plus ITMS instrument (Thermo Fischer Scientific, San Jose, CA)

Microbial Degradation of Microcystin

Figure 1. Analysis procedure for degradation of MCLR, in which three kinds of B-9 cell preparations, a crude B-9 cell (CB9), a lyophilized B-9 cell (LB9), and a washed B-9 cell (WB9), and the advanced Marfey method for WB9 were included.

equipped with an electrospray ionization (ESI) interface. The ESI conditions in the positive ion mode were as follows: capillary temperature, 300 °C; sheath gas flow rate, 35 (arbitrary units); ESI source voltage, 5000 V; capillary voltage, 43 V; tube lens offset, 15 V. Various scan ranges were used according to the molecular weights of the FDLA derivatives of possible amino acids and peptides from the MCLR biodegradation. Prediction of log D Using ACD Laboratories LogD Software. The log D values were calculated using the software ACD Laboratories LogD (Advanced Chemistry Development ACD/ Laboratories Research, Ontario, Canada). See http://www.acdlabs. com/products/phys_chem_lab/logd/tech.html.

Results Selection of a Favorable Degradation Condition. To eliminate contaminants, particularly L-amino acids and small peptides, from the Sakurai medium used, the B-9 crude cells were rinsed three times with 50 mM Tris-Cl, followed by 0.9% NaCl. As will be shown below, this cell preparation of the crude mycelium (designated WB9) was effective for obtaining intermediates of MCLR. In this study, the conventional cell preparations, crude B-9 cell (CB9) and lyophilized B-9 cell (LB9), were compared with WB9 concerning the degradation behavior (Figure 1). Further, CB9 and WB9 were confirmed to be still alive by checking their viability. The comparison among the three cell preparations was performed with a solution at 1 mg/mL MCLR, and the degradation process was carefully monitored by HPLC-PDA and LC/MS in both pre and post L-FDLA derivatization. Figure 2 shows the MCLR biodegradation process of CB9, LB9, and WB9, which were analyzed by HPLC at 238 nm. Although the linearized MCLR and tetrapeptide Adda-Glu-Mdha-Ala were commonly monitored as the known products in addition to MCLR, different degradation behavior was observed among the three preparations. MCLR samples were similarly decreased, and the formed linearized MCLR was converted rapidly to the tetrapeptide in the three preparations. While LB9 and CB9 showed a similar biodegradation profile in which the tetrapeptide peaks were mainly observed at 24 h, followed by their rapid degradation, the tetrapeptide peak continued and did not disappear during the experiment with WB9. These results suggested a possibility that WB9 can provide new intermediates such as tri- and dipeptides, because the degradation speed is slower than that of the conventional preparations. Application of the Advanced Marfey Method. We applied the advanced Marfey method to all MCLR degradation experi-

Chem. Res. Toxicol., Vol. 22, No. 2, 2009 393

ments, in which the three preparations with MCLR were always compared with the negative control without MCLR using selected ion monitoring (SIM), because standards for the unusual small peptides and amino acids were not available. Although known linearized MCLR, tetrapeptide, and Adda could be detected before the FDLA derivatization, probable tetrapeptides, tripeptides, dipeptides, and amino acids were examined by SIM using the mass range of putative products after the derivatization. To confirm the capability of this method for our purpose, it was applied to methylamine that is derived from one of the constituent amino acids, Mdha. The derivatized standard methylamine appeared at 13.2 min in the SIM profile at m/z 323.6-326.6. While the negative control showed no desired peak at each sampling time in the SIM profiles at m/z 323.6-326.6, the desired peak began to appear at 36 h, and the most intense peak was observed at 96 h in the SIM profile with MCLR (Supporting Information Figure 1). These results indicated that the desired degradation products could certainly be observed from the hydrolysate by this method and that the degradation process could be demonstrated on the basis of the SIM profiles. MCLR biodegradation products derived from the tetrapetide (Adda-Glu-Mdha-Ala) and tripeptide (Leu-MeAsp-Arg) moieties were detected by SIM. These precursors can be detected at m/z 908.8 and 16.7 min and at m/z 711.1 and 10.5 min, respectively. From the former, the two tripeptides Adda-Glu-Mdha and GluMdha-Ala appeared at m/z 839 and 15.5 min and at m/z 595.6 and 12.0 min, respectively. While Adda-Glu from the tetrapeptide moiety and Leu-MeAsp from the counterpart tripeptide were not observed (data not shown), dipeptides Glu-Mdha, MeAspArg, and Mdha-Ala were found at 12.4, 16.8, and 16.9 min. Although Adda, Arg, and Ala appeared at 18.8, 10.1, and 13.3 min, respectively, Leu, Glu, and MeAsp could not detected. Due to the inherent characteristic of the Mdha residue, which was spontaneously converted to methylamine (28), it was detected as [CH3NH - L-DLA + H]+ at m/z 325.7 from 48 to 96 h (Supporting Information Figure 2). Figure 3 shows the biodegradation process of the representative intermediates from 0 to 96 h on the basis of the results by the advanced Marfey method. The production of the tetrapeptide Adda-Glu-Mdha-Ala (Figure 3A) increased from 36 to 48 h, and that of its degradation products, tripeptides Adda-Glu-Mdha (Figure 3B) and Glu-Mdha-Ala (Figure 3E), reached maxima at 72 or 96 h. Although the corresponding dipepides were not detected, the amount of methylamine similarly increased in the final stage (Figure 3D). The production of the counterpart tripeptide Leu-MeAsp-Arg (Figure 3H) reached a maximum at 24 h and then decreased slowly to the end of the experiment. The detected Arg from this tripeptide showed behavior similar to that of the tripeptide (Figure 3J). These results suggested the tetrapeptide was mainly cleaved at Adda-Glu and Mdha-Ala peptide bonds by the B-9 strain. Adjustment of Retention Times Using log D. As mentioned above, we detected the following microbial degradation products from MCLR by WB9: tetrapeptide Adda-Glu-Mdha-Ala, tripeptides Adda-Glu-Mdha, Glu-Mdha-Ala, and Leu-MeAsp-Arg, dipeptides Glu-Mdha, Mdha-Ala, and MeAsp-Arg, and amino acids Adda, Arg, and methylamine derived from Mdha. These products could be identified by SIM using the mass range of putative products after the derivatization with FDLA. The derivatization can provide the hydrophobicity and resolve the degradation products under the usual reversed-phase conditions. However, it is ambiguous whether they are real degradation products. To determine which compound is a real product, we

394

Chem. Res. Toxicol., Vol. 22, No. 2, 2009

Hashimoto et al.

Figure 2. Microbial degradation process of MCLR using (A) CB9, (B) LB9, and (C) WB9. MCLR, linearized MCLR, and tetrapeptide AddaGlu-Mdha-Ala were analyzed by HPLC.

Table 1. log D and Predicted and Observed Retention Times (min) of Constituent Peptides and Amino Acids from Microcystin-LR retention time (min) peptide/amino acid

log D

predicted

observed

Adda-Glu-Mdha-Ala Adda-Glu-Mdha Adda-Glu Glu-Mdha-Ala Glu-Mdha Mdha-Ala Leu-MeAsp-Arg Leu-MeAsp MeAsp-Arg Adda Glu methylamine Ala Leu MeAsp Arg

6.05 6.7 6.87 1.81 2.51 2.91 1.65 5.35 0.33 8.93 3.09 4.79 4.09 5.5 4.07 0.88

15.6 16.3 16.4 11.3 12 12.4 11.2 14.8 10 18.4a 12.6 14.4 13.6a 15 13.5 10.5a

16.7 15.5 NDc 12 12.4 16.8b 10.5 ND 16.9b 18.8 ND 13.2 ND ND ND 10.1

a Available amino acids were used as the standard sample to estimate the retention time on the basis of the relationship between the retention time and log D. b These were excluded due to the large difference between the predicted and observed retention times. c ND ) not detected.

introduced log D, an index of hydrophobicity, and estimated the retention time of a putative degradation product (29, 30). First, the relationship between the hydrophobicity using log D and the retention time of available degradation products Adda, Ala, and Arg derivatized with FDLA was estimated. Second, the retention times of unknown degradation products were extrapolated on the basis of the estimated times of the three products. Table 1 summarizes log D and the predicted and observed retention times of the putative degradation products. Although the dipeptides MeAsp-Arg and Mdha-Ala could be detected in the SIM experiments, the retention times predicted using log D were too early compared with the detected times, suggesting that they were not real products from MCLR. Therefore, these could be excluded, and the remaining degradation products were almost fitted with the estimated times as shown in Table 1. After these operations, a total of eight degradation products including the tetrapeptide were found to be additional degradation products.

Discussion Degradation Behavior Using the Selected Conditions. Bourne et al. reported the microbial degradation behavior of MC in the conventional manner including the use of an enzyme inhibitor (22). The degradation of MCLR to the linearized MCLR by a microcystin hydrolytic enzyme, MlrA, and the tetrapeptide to small peptides or amino acids was inhibited by a known metalloprotease inhibitor, EDTA. PMSF (phenylmethanesulfonyl fluoride), a serine protease inhibitor, inhibited the degradation of the linearized MCLR to the tetrapeptide by MlrB (23). In our previous study, the degradation of MCLR to Adda by a B-9 cell extract without any inhibitor was too quick to detect other intermediates. In a previous study, we described that B-9 can hydrolyze peptides other than MC (19). This indicates that B-9 also degrades various oligopeptides in the medium, accumulating low molecular weight peptides similar to those investigated in this study. It was suggested that an accurate analysis would be laborious due to these peptides. Such interference was overcome by repetitive washing of B-9 cells with sodium chloride solution to eliminate amino acids and peptides derived from the medium components. The cell viability was demonstrated by checking B-9 washed cell (WB9) growth after 24 h. An additional advantage of WB9 was found to be the slower degradation behavior of MCLR compared with that of CB9 and LB9, which can improve monitoring of the biodegradation process. The repetitive washing process probably retarded the transportation of the biodegradation products into the cells, mainly for the tetrapeptide (Figure 2). Therefore, the WB9 cell preparation was chosen for subsequent experiments. Application of the Advanced Marfey Method and log D for Identification of the Real Degradation Products. There have been three intermediates of MC, linearized MC, tetrapeptide, and Adda, that can be readily detected by ordinary HPLC, because they commonly have an Adda moiety with a diene as the chromophore with an absorption maximum at 238 nm (24, 25). To date, no other degradation products without such a chromophore have been isolated. For the complete identification of the degradation products, the following should be overcome: (1) how to control the microbial degradation; (2) how to differentiate real products such as amino acids and low molecular weight peptides from impurities; (3) how to detect and separate the desired amino acid and peptides; (4) how to obtain the unusual amino acid and/or peptide standards.

Figure 3. Microbial degradation process of (A) Adda-Glu-Mdha-Ala, (B) Adda-Glu-Mdha, (C) Adda, (D) methylamine, (E) Glu-Mdha-Ala, (F) Glu-Mdha, (G) Mdha-Ala, (H) Leu-MeAsp-Arg, (I) MeAspArg, and (J) Arg as determined on the basis of the results by the advanced Marfey method.

Microbial Degradation of Microcystin Chem. Res. Toxicol., Vol. 22, No. 2, 2009 395

396

Chem. Res. Toxicol., Vol. 22, No. 2, 2009

Hashimoto et al.

Figure 4. Renewal degradation scheme for MCLR using WB9. The solid arrows indicate the degradation route, and the dotted arrows indicate the degradation to undetected peptide or amino acid.

As mentioned above, the microbial degradation using an MCdegrading microorganism has been controlled using an enzyme inhibitor (22). In this study, we found that washing an MCdegrading microorganism showed an interesting behavior in which this could slow the degradation. To overcome the remaining problems 2-4, the advanced Marfey method was applied. In this method, an amino acid or peptide is derivatized with FDLA and the resulting derivative possesses a DLA chromophore that can be detected at 350 nm and separated under the usual reversed-phase conditions. Further, a desired derivative can be detected by SIM using the putative m/z value (26, 27). An effective application of the advanced Marfey method would resolve all three of these problems. The advanced Marfey derivatization was carried out in all MCLR biodegradation experiments as well as the corresponding negative control. The method can detect and separate a desired amino acid from impurities and determine its absolute configuration (Supporting Information Figures 1 and 2). Due to unavailability of the unusual small peptide standards, biodegradation products were tentatively identified using SIM of the negative control (B-9) versus the B-9 strain + MCLR. As a result of these experiments, 10 products could be tentatively identified as shown in Table 1. To further ensure the results, we used log D, an index of hydrophobicity considering ionic forms depending on the pH, to predict the retention time of the

DLA derivatives. After these operations, a total of 8 of the 10 degradation products including the tetrapeptide were found to be additional degradation products. Therefore, we can also depict the biodegradation process of the representative intermediates on the basis of the results of these operations (Figure 3). The combination of the advanced Marfey method with log D would provide more reliable structural information on a mixture composed of amino acids and small peptides. The detailed results and discussion on log D will be reported elsewhere. A Renewal Degradation Scheme for Microcystin. Figure 4 summarizes the microbial degradation scheme of MCLR verified in this study. The degradation was started at the Arg-Adda peptide bond with the ring opening to produce a linearized MCLR (Adda-Glu-Mdha-Ala-Leu-MeAsp-Arg). Hydrolysis at the Ala-Leu peptide bond synchronously provided the tetrapeptide Adda-Glu-Mdha-Ala and tripeptide Leu-MeAspArg. The successive cleavage of the tetrapeptide formed mainly Adda and a tripeptide (Glu-Mdha-Ala), and another tripeptide (Adda-Glu-Mdha) was detected as a minor product. Although these tripeptides were probably cleaved to each amino acid through the corresponding dipeptides, they were not detected. Finally, Adda and methylamine derived from Mdha were identified as the amino acids. On the other hand, the counterpart tripeptide Leu-MeAsp-Arg was probably hydrolyzed to each

Microbial Degradation of Microcystin

amino acid by way of the dipeptides Leu-MeAsp and MeAspArg, but only Arg was detected. Bourne et al. reported four enzymes associated with MC degradation, MlrA, MlrB, MlrC, and MlrD (23). Saito et al. described that other MC-degrading bacteria possess MlrA in a genetic study (31). Bourne et al. described the functions of these enzymes in the ACM-3962 strain as follows: MlrA hydrolyzes MCLR to linearized MCLR, MlrB hydrolyzes the linearized tetrapeptide, MlrC hydrolyzes these peptides to smaller peptides or amino acids, and MlrD is a transporter. The mlrA gene, which codes the enzyme MlrA synthesis, is conserved in at least three different bacterial species (31), and it has been detected in MCdegrading bacteria but not in all Sphingomonas, suggesting that the MC degradation gene was acquired during evolution of Sphingomonas (31). Although these enzymes have not yet been detected, the mlrA gene was preliminarily detected in the B-9 strain. Indeed, B-9 shows almost the same degradation behavior as ACM-3962. A previous study demonstrated that the B-9 strain can degrade nontoxic cyanobacterial cyclic peptides (19), and further microbial degradation of MC can be clearly shown in the present study, indicating that MC-degrading bacteria possess additional genes and the corresponding enzymes for the degradation of peptides.

Conclusion To further understand the MCLR-degrading ability of the B-9 strain, isolation of the MC-degrading enzymes and detection of related genes are naturally significant. Identification of the intermediates and final products of MCLR is also important to understand the very unique hydrolytic activity of this strain. In this study, we regulated B-9 hydrolytic activity by washing with sodium chloride solution to obtain a desired cell preparation, which permits the intermediate and final products of MCLR to be obtained. The resulting products could be firmly identified using the advanced Marfey method with the aid of log D. The present study expands the hydrolytic activity of the B-9 strain, which can hydrolyze not only cyanobacterial cyclic peptides but also MC to the intermediates and final products. In a preliminary experiment, we confirmed that B-9 can also hydrolyze other types of cyclic peptides produced by microorganisms. Explanation of the function and role of B-9 and other MC-degrading microorganisms in the natural environment is thus highly required. The B-9 strain may contribute to the detoxification of dangerous compounds such as MC and other peptides in the ecosystem, and better understanding of the mechanism of this process would facilitate overcoming pollution problems. Supporting Information Available: Time course for the SIM profiles of methylamine-DLA derived from Mdha [(A) MCLR + WB9 and (B) WB9 (negative control)] and SIM profiles of putative degradation products [(A) Adda-Glu-Mdha-Ala, (B) Adda-Glu-Mdha, (C) Adda, (D) Glu-Mdha-Ala, (E) Glu-Mdha, (F) Leu-MeAsp-Arg, (G) Arg, (H) methylamine, (I) MeAspArg, and (J) Mdha-Ala; (a) WB9 + MCLR and (b) WB9 as the negative control]. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Dawson, R. M. (1998) The toxicology of microcystins. Toxicon 36, 953–962. (2) Imanishi, S., and Harada, K.-I. (2004) Proteomics approach on microcystin binding proteins in mouse liver for investigation of microcystin toxicity. Toxicon 43, 651–659.

Chem. Res. Toxicol., Vol. 22, No. 2, 2009 397 (3) Botes, D. P., Kruger, H., and Viljoin, C. C. (1982) Isolation and characterization of four toxins from the blue-green alga Microcystis aeruginosa. Toxicon 20, 945–954. (4) Hoeger, S. J., Hitzfeld, B. C., and Dietrich, D. R. (2005) Occurrence and elimination of cyanobacterial toxins in drinking water treatment plants. Toxicol. Appl. Pharmacol. 203, 231–242. (5) Jones, G. J., and Orr, P. T. (1994) Release and degradation of microcystin following algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein phosphatase inhibition assay. Water Res. 28, 871–876. (6) Tsuji, K., Naito, S., Kondo, F., Ishikawa, N., Watanabe, F. M., Suzuki, M., and Harada, K.-I. (1994) Stability of microcystin from cyanobacteria: Effect of light on decomposition and isomerization. EnViron. Sci. Technol. 28, 173–177. (7) Mazur, H., and Plinski, M. (2001) Stability of cyanotoxins, microcystin-LR, microcystin-RR and nodularin in seawater and BG-11 medium of different salinity. Oceanologia 43, 329–339. (8) Jochimsen, E. M., Carmichael, W. W., An, J. S., Cardo, D. M., Cookson, S. T., Holmes, C. E. M., Antunes, M. B., de Melo Filho, D. A., Lyra, T. M., Barreto, V. S. T., Azevedo, S. M. F. O., and Jarvis, W. R. (1998) Liver failure and death after exposure to microcystins at a hemodialysis center in Brazil. N. Engl. J. Med. 338, 873–878. (9) Pouria, S., de Andrade, A., Barbosa, J., Cavalcanti, R. L., Barreto, V. T. S., Ward, C. J., Preiser, W., Poon, G. K., Neild, G. H., and Codd, G. A. (1998) Fatal microcystin intoxication in haemodialysis unit in Caruaru, Brazil. Lancet 352, 21–26. (10) Jones, G. J., Bourne, D. G., Blakeley, R. L., and Doelle, H. (1994) Degradation of the cyanobacterial hepatotoxin microcystin by aquatic bacteria. Nat. Toxins 2, 228–235. (11) Park, H. D., Sasaki, Y., Maruyama, T., Yanagisawa, E., Hiraishi, A., and Kato, K. (2001) Degradation of the cyanobacterial hepatotoxin microcystin by a new bacterium isolated from a hypertrophic lake. EnViron. Toxicol. 16, 337–343. (12) Saitou, T., Sugiura, N., Itayama, T., Inamori, Y., and Matsumura, M. (2003) Degradation characteristics of microcystins by isolated bacteria from Lake Kasumigaura. J. Water SRTsAqua 52, 13–18. (13) Ishii, H., Nishijima, M., and Abe, T. (2004) Characterization of degradation process of cyanobacterial hepatotoxins by a gramnegative aerobic bacterium. Water Res. 38, 2667–2676. (14) Harada, K.-I., Imanishi, S., Kato, H., Mizuno, M., Ito, E., and Tsuji, K. (2004) Isolation of Adda from microcystin-LR by microbial degradation. Toxicon 44, 107–109. (15) Rapala, J., Berg, K. A., Lyra, C., Niemi, R. M., Manz, W., Suomalainen, S., Paulin, L., and Lahti, K. (2005) Paucibacter toxiniVorans gen. nov., sp. nov., a bacterium that degrades cyclic cyanobacterial hepatotoxins microcystins and nodularin. Int. J. Syst. EVol. Microbiol. 55, 1563–1568. (16) Valeria, A. M., Ricardo, E. J., Stephan, P., and Alberto, W. D. (2006) Degradation of microcystin-RR by Sphingomonas sp.CBA4 isolated from San Roque reservoir (Co´rdobasArgentina). Biodegradation 17, 447–455. (17) Maruyama, T., Park, H. D., Ozawa, K., Tanaka, Y., Sumino, T., Hamana, K., Hiraishi, A., and Kato, K. (2006) Sphingosinicella microcystiniVorans gen. nov., sp. nov., a microcystin-degrading bacterium. Int. J. Syst. EVol. Microbiol. 56, 85–89. (18) Imanishi, S., Kato, H., Mizuno, M., Tsuji, K., and Harada, K.-I. (2005) Bacterial degradation of microcystins and nodularin. Chem. Res. Toxicol. 18, 591–598. (19) Kato, H., Imanishi, S. Y., Tsuji, K., and Harada, K.-I. (2007) Microbial degradation of cyanobacterial cyclic peptides. Water Res. 41, 1754– 1762. (20) Bourne, D. G., Blakeley, R. L., Riddles, P., and Jones, G. J. (2006) Biodegradation of the cyanobacterial toxin microcystin LR in natural water and biologically active sand filters. Water Res. 40, 1294–1302. (21) Tsuji, K., Asakawa, M., Anzai, Y., Sumino, T., and Harada, K.-I. (2006) Degradation of microcystins using immobilized microorganism isolated in an eutrophic lake. Chemosphere 65, 117–124. (22) Bourne, D. G., Jones, G. J., Blakeley, R. L., Jones, A., Negri, A. P., and Riddles, P. (1996) Enzymatic pathway for the bacterial degradation of the cyanobacterial cyclic peptide toxin microcystin LR. Appl. EnViron. Microbiol. 62, 4086–4094. (23) Bourne, D. G., Riddles, P., Jones, G. J., Smith, W., and Blakeley, R. L. (2001) Characterization of a gene cluster involved in bacterial degradation of the cyanobacterial toxin microcystin LR. EnViron. Toxicol. 16, 523–534. (24) Lawton, L. A., Beattie, K. A., Hawser, S. P., Campbell, D. L., and Codd, G. A. (1994) Evaluation of assay methods for the determination of cyanobacterial hepatotoxicity. In Special PublicationsDetect Methods Cyanobacterial Toxins (Codd, G. A., Jeffries, T. M., Keevil, C. W., and Potter, E., Eds.) Vol. 149, pp 111-116, The Royal Society of Chemistry, Cambridge, U.K.

398

Chem. Res. Toxicol., Vol. 22, No. 2, 2009

(25) McElhiney, J., and Lawton, L. A. (2005) Detection of the cyanobacterial hepatotoxins microcystins. Toxicol. Appl. Pharmacol. 203, 219– 230. (26) Fujii, K., Ikai, Y., Mayumi, T., Oka, H., Suzuki, M., and Harada, K.-I. (1997) A non-empirical method using LC/MS for determination of the absolute configuration of constituent amino acids in a peptide: Elucidation of limitations of Marfey’s method and of its separation mechanism. Anal. Chem. 69, 3346–3352. (27) Fujii, K., Ikai, Y., Oka, H., Suzuki, M., and Harada, K.-I. (1997) A nonepimetical method using LC/MS for determination of absolute configuration of constituent amino acids in a peptide: Combination of Marfey’s method with mass spectrometry and its practical application. Anal. Chem. 69, 5146–5151. (28) Harada, K.-I., Matsuura, K., Suzuki, M., Watanabe, M. F., Oishi, S., Dahlem, A. M., Beasley, V. R., and Carmichael, W. W. (1990) Isolation and characterization of the minor components associated with

Hashimoto et al. microcystins LR and RR in the cyanobacterium (blue-green algae). Toxicon 28, 55–64. (29) Kadar, E. P., Wujcik, C. E., Wolford, D. P., and Kavetskaia, O. (2008) Rapid determination of the applicability of hydrophilic interaction chromatography utilizing ACD labs log D suite: A bioanalytical application. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 863, 1–8. (30) Dohta, Y., Yamashita, T., Horiike, S., Nakamura, T., and Fukami, T. (2007) A system for LogD screening of 96-well plates using a waterplug aspiration/injection method combined with high-performance liquid chromatography-mass spectrometry. Anal. Chem. 79, 8312–8315. (31) Saito, T., Okano, K., Park, H.-D., Itayama, T., Inamori, Y., Neilan, B. A., Burns, B. P., and Sugiura, N. (2003) Detection and sequencing of the microcystin LR-degrading gene mlrA, from new bacteria isolated from Japanese lakes. FEMS Microbiol. Lett. 229, 271–276.

TX8003517