Peroxidase catalyzed co-polymerization of pentachlorophenol and
a potential humic precursor
Kengo Morimoto, Kenji Tatsumi*, Ken-Ichi Kuroda
Hydrospheric environmental protection department, national institute for resources and environment, 16-3 Onogawa, Tsukuba, Ibaraki 3058569, Japan
Accepted 24 January 2000
Abstract
The transformation of xenobiotics in the soil was studied in a model experimental system using pentachlorophenol (PCP), an enzyme, and a potential humic precursor. It has been shown that potential humic precursors such as p-coumaric acid greatly accelerated the peroxidase-catalyzed reaction of PCP (Morimoto K. and Tatsumi K, 1997. Eect of humic substances on the enzymatic formation of OCDD from PCP. Chemosphere 34, 1277±1283). About 90% of the PCP reacted but almost no transformation products were detected by GC/MS analysis. Most of the transformed PCP was detected in the polymeric products by the determination of the adsorbable organic halide (AOX) in gel permeation chromatography (GPC) fractions. Pyrolysis gas chromatography/mass spectrometry (py-GC/MS) of the polymers showed that tetrachlorophenol (TeCP) was the major pyrolysis product. This accords with the ion chromatography observation that about 20% of the PCP chlorine was released as chloride ion. However, nitrobenzene oxidation produced only a small amount of PCP and no TeCP. Based on the selectivity of the two degradation methods, it is concluded that TeCP is bound to the polymer by a diphenylether linkage, and PCP by an ether linkage to the aliphatic side chain. Thus, it was concluded that the major part of the PCP was incorporated into the polymer fromp-coumaric acid by the release of a chlorine atom from the PCP molecule to form a diphenylether linkage with thep-hydroxyl group of thep-coumaric acid polymer.72000 Elsevier Science Ltd. All rights reserved.
Keywords:PCP;p-Coumaric acid; Humic substance; Peroxidase; Pyrolysis-GC/MS; Alkaline nitrobenzene oxidation
1. Introduction
Large amounts of toxic chemicals have been applied as agricultural chemicals or discharged as industrial euent into the environment. Their fate and the possi-bility of their bioremediation have been an active con-cern.
The transformation of phenolic substances by white rot fungi may provide a solution for part of this pro-blem. The fungi excrete extracellular phenol-oxidizing enzymes such as laccase or peroxidase, which bring about the transformation. Peroxidase is a common extracellular enzyme of white rot fungi but also occurs
elsewhere, e.g. in lignin synthesis, in plants or in the reduction of superoxide in various organisms (Everse et al., 1991). It is known to produce phenoxy radicals by dehydrogenation of phenolic substances in the pre-sence of hydrogen peroxide, leading to coupling reac-tions which form phenolic polymers such as lignin.
This enzyme also has an important role in humi®ca-tion reachumi®ca-tions in soils (Ladd and Butler, 1975; Sul®ta and Bollag, 1981), a complex process including both degradation and synthesis. The ability of the enzyme to cause phenolic coupling is also likely to allow the immobilization of toxic phenolic substances such as chlorophenols (Bollag et al., 1987; Ruttimann-Johnson and Lamar, 1997). In any study of such immobiliz-ation, it is easier to use the isolated enzyme rather than fungal cultures, with their attendant growth restrictions. If the process is to become an instrument
0038-0717/00/$ - see front matter72000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 3 8 - 0 7 1 7 ( 0 0 ) 0 0 0 1 6 - X
www.elsevier.com/locate/soilbio
* Corresponding author.
for bioremediation, it will also be safer to use the iso-lated enzyme rather than whole micro-organisms.
Pentachlorophenol (PCP) has commonly been used as a herbicide and it is still used for wood preservation in some countries (Tanjore and Viraraghavan, 1994). It is well known that peroxidase catalyzes the trans-formation of PCP in the presence of hydrogen per-oxide. This reaction is thought to involve the polymerization of PCP but also yields octachlorodi-benzo-p-dioxin (OCDD) as a by-product (Oberg et al. 1990; Oberg and Rappe, 1992). However, this result was obtained in an experiment where no humic sub-stances were present. To assess the fate of PCP in soil, it is necessary to consider the participation of humic substances in any transformation process (Tatsumi et al., 1994a, 1994b; Roper et al. 1995; Ruttimann-John-son and Lamar, 1996). Model experiments (Morimoto and Tatsumi, 1997) have shown that the oxidation of PCP in the presence of potential humic precursors such as p-coumaric acid or p-hydroxybenzoic acid can remove up to 90% of the PCP. This reaction also sig-ni®cantly reduced the yields of chlorinated quinones or dimers, such as chlorinated dioxins or phenoxyphenols. However, the details of the PCP reaction pathway are not clearly understood.
We studied the dechlorination and co-polymeriz-ation of PCP with p-coumaric acid, and the resulting chlorine balance. From analyses using alkaline nitro-benzene oxidation and pyrolysis GC/MS, the chemical structure of the polymeric product is discussed.
2. Materials and methods
2.1. Materials
Pentachlorophenol and p-coumaric acid were pur-chased from Nacalai Tesque (Kyoto, Japan). Nitro-benzene was obtained from Aldrich Chemical. (Milwaukee, WI, USA.). Horseradish peroxidase (HRP) was obtained from Wako Pure Chemical Indus-try (Osaka, Japan).
2.2. Reaction conditions
To observe any PCP decrease and dechlorination, the reaction solution contained 50 mM PCP, 200 mM p-coumaric acid, 1 unit HRP mlÿ1, 0.5 mM H2O2. The
pH was adjusted to 5, and the mixture was reacted at 288C under aerobic conditions using a water bath sha-ker. PCP and chloride ion concentrations in the sol-ution were monitored during the reaction.
To obtain sucient polymeric product for structure analysis, the concentrations of the reactants were raised in a second experiment to 150 mM PCP, 300
mM p-coumaric acid, 6 units HRP mlÿ1, 0.9 mM
H2O2. To dissolve this quantity of PCP, acetone was added (25% v/v) to the solution.
2.3. Sample preparation
The reaction solution was fractionated by gel per-meation chromatography (GPC) after 0, 5 and 30 min incubation. Fractions (15 ml) were collected and the adsorbable organic halide (AOX, see below) of each fraction was measured as an index of the total organic chloride.
In a further reaction experiment (500 ml volume), 50 ml samples were taken at ®me points 0, 5, 15, 30, 60,120 and 240 min. The pH of each sample was raised to 9 with NaOH to stop the reaction and to dissolve the transformation products. Then 50 ml samples from these solutions were used to determine AOX, and to obtain the polymeric products and the remainder was dialyzed against water in cellulose tubing (mean pore diameter 2.4 nm). The AOX of the dialyzed reaction solution was measured to evaluate the quantity of PCP that was incorporated into the enzymatic polymeriz-ation, and the remaining solution was taken to dryness and analyzed by pyrolysis-GC/MS (Meier and Faix, 1992). A further 2 l of solution was similarly reacted to obtain 20 mg of polymer products for alkaline nitrobenzene oxidation analysis.
2.4. Analytical methods
PCP in the reaction solution was monitored by HPLC using a Jasco PU-980 (Japan Spectroscopic, Tokyo, Japan) provided with a UV detector and an integrator. A reverse phase column, Cosmosil 5C18-SL(4.6 mm i.d. 25 cm, Nacalai tesque, Kyoto, Japan), was used with a 10:90 methanol/0.08% phos-phate buer mobile phase (¯ow rate 1.0 ml minÿ1
). The concentration of chloride ions released from PCP was measured in a DIONEX series 4000i ion chromatograph using a HPIC AS12A column with 0.3 mM NaHCO3/2.7 mM Na2CO3as eluant.
GPC was carried out on Biogel P-30 polyacrylamide gel (Bio-Rad, California, USA) packed in a 2.550 cm column using 50 mM NaOH/50 mM Na2B4O7 as eluant.
To evaluate the quantity of total organic chlorine, the reaction solution was injected into a 3 40 mm column of activated carbon powder (40 mg). After washing the column repeatedly with KNO3 to remove inorganic halide ion, the total chloride concentration of this column was measured by a TOX-10S instru-ment (Mitsubishi Kasei, Tokyo, Japan). The halide concentration measured as above is called adsorbable organic halide (AOX) that is used as an index of total organic halide.
structure was analysed by pyrolysis-GC/MS and by nitrobenzene oxidation, using syringaldehyde as an in-ternal standard. About 100mg of the material was pyr-olyzed at 5008C for 5 s in a JHP-3 curie point pyrolyzer (Japan Analytical Industry Co. Ltd., Mizuho, Tokyo, Japan). The pyrolyzate was directly injected into an HP5972 / HP5890 series II GC/MS (Hewlett Packard, Palo Alto, CA, USA) equipped with a HP-1 fused silica capillary column (0.2 mm25 m). The temperature of the injector and the detector trans-fer line was kept at 3008C. The column oven tempera-ture was held at 508C for 5 min and then raised to 3008C at a rate of 88C minÿ1
.
The polymer (20 mg) was also oxidised in alkaline nitrobenzene at 1708C for 2.5 h in an oil bath shaker (Chen, 1992). The resultant mixture was ®ltered and the acidi®ed ®ltrate was extracted with diethylether. The ether-soluble products were analyzed by GC/MS as described above.
3. Results and discussion
3.1. Incubation analysis
Morimoto and Tatsumi (1997) have shown that horseradish peroxidase can eectively catalyze the transformation of PCP when reacted with p-coumaric acid. Almost no chlorinated monomeric or dimeric products were detected. This suggests that the PCP was incorporated into polymeric structures or further degraded by dechlorination and ring cleavage. To quantify the degradation of PCP, its concentration during the reaction was determined, along with the chloride ion released (Fig. 1). This chloride was expressed as a percentage of the total chlorine con-tained in the PCP in the initial solution. On this basis, more than 80% of the PCP disappeared within the ®rst 5 min of reaction and more than 90% within 20 min. Along with the PCP decrease, 18% of the chlor-ine initially contachlor-ined in PCP was released as chloride
ion in the ®rst 5 min, but thereafter the process slowed with the maximum being near 20%, corresponding to the release of one chlorine atom from each PCP mol-ecule. However, about 70% of the chlorine in the in-itial PCP was not detected in the reaction solution. Its incorporation in the polymeric products is the only remaining explanation for its fate.
3.2. Molecular weight distribution of the chlorinated reaction products
Dialysis and GPC were used to detect co-polymeriz-ation of PCP as a possible fate of the unknown 70% chlorine. To obtain more products for fractionation, the concentration of PCP in the reaction solution was raised three-fold. This decreased the transformation yield and 15% of the initial PCP was left unchanged for 30 min after the reaction started.
Fig. 2 shows the AOX change in the reaction sol-ution and the dialysis residue along with the residual PCP concentration. The AOX of the reaction solution was about 20±30% less than the initial concentration of PCP, corresponding to a dechlorination of nearly 20%. The AOX of the dialysis residue was slightly more than 50%. Since this fraction cannot pass through the 2.4 nm pore-sized dialysis tubing, these molecules should contain at least 5 benzene rings. The dierence in AOX between the reaction solution and the dialysis residue should correspond to the amount of organic chlorine (including that of residual PCP) in products smaller than 2.4 nm.
Dialysis showed most of the transformed PCP was coupled with p-coumaric acid to form oligomeric or polymeric products. GPC was used to determine the molecular size distribution of these products in the reaction solution and the AOX of each GPC fraction is shown in Fig. 3. Most of the organic chlorine was detected in two peaks, corresponding to the exclusion limit and low molecular weight peaks. The latter peak was found to be residual PCP by comparing AOX with the PCP concentration of the same fractions
Fig. 2. Adsorbable organic halide (AOX) change of the reaction sol-ution and the dialyzed reaction solsol-ution during incubation.
measured by HPLC, using the AOX content of 150
mM PCP solution as a standard. Another peak of almost the same AOX content appeared at the exclu-sion volume, which corresponds to a 40,000 Da pro-tein standard. Allowing for the physical dierences between phenolic polymers and protein molecules, this peak suggests that the polymer contained more than 200 monomer units, and the detection of AOX from this fraction indicates the incorporation of PCP into that polymer, since PCP was the only chlorine source in the solution mixture.
HPLC analysis showed the total absence of p-cou-maric acid after 2 min of reaction, at the point where PCP just started to decrease. In addition, preliminary experiments showed that when PCP was reacted with the polymer resulting from a 10 min previous reaction ofp-coumaric acid, the PCP reduction rate was almost the same or even a little faster than with the p-couma-ric acid monomer. These results suggest that p-couma-ric acid ®rst polymerized and then the PCP was bound to that polymer, a mechanism that also explains the absence of dimeric or trimeric products including chlorine. No oligomers or dimers were present even in the very early stages of the reaction, despite several attempts to obtain dimeric or trimeric product.
3.3. Characterization of the polymerized product
Chromatograms of the nitrobenzene oxidation pro-ducts from the 30 min-reaction product are shown in Fig. 4(a) and the pyrogram is shown in Fig. 4(b). The only chlorinated substance produced by nitrobenzene oxidation was PCP and the yield was 2% of the initial solution. In contrast, pyrolysis produced a consider-able amount of tetrachlorophenol (TeCP) and also a small amount of trichlorophenol (TrCP) and PCP, the PCP is almost the same yield as in the nitrobenzene oxidation. The detection of TeCP means one chlorine
atom of PCP was lost at the binding site where PCP forms a linkage with the polymer.
Fig. 5 shows the variation of each chlorinated pyrol-ysis product in the dialyzed reaction solution while Fig. 2 shows the AOX variation in the same incu-bation. The total chlorophenol recovery at 120 min was about 18% of the initial solution and 35% of the dialysis residue AOX. This yield is appropriate consid-ering that the recovery of phenolic substances from lig-nin in general is about 30% (Lam et al., 1990; Kuroda et al., 1994).
TeCP was the major organohalogen detected, corre-sponding to the observation that nearly 20% of the chlorine was released as chloride ion from the trans-formed PCP. The concentration of pyrolysis products increased for the ®rst 2 h of reaction, but decreased afterwards. However the AOX of the dialysis residue showed no such decline (Fig. 2). One possible expla-nation is that the polymer became more resistant to pyrolysis as the reaction proceeded, indicating a con-tinuous condensation within the polymer structure even after most of the monomers were consumed.
Phenol oxidizing enzymes such as peroxidase pro-duce phenoxy radicals through dehydrogenation of the hydroxyl group (Courteix and Bergel, 1995a, 1995b). This radical is resonance-stabilized as shown in Fig. 6 for p-coumaric acid. Here radical coupling occurs at the p-hydroxyl group, the m-position of the benzene ring and theb-carbon of the side-chain.
Fig. 7 shows the probable linkage in an analogous lignin structure and the possibility of their cleavage by nitrobenzene oxidation and pyrolysis. L2 and L4 are C±C linkages, presumed to form enzymatically, between PCP and p-coumaric acid. These linkages are not cleaved either by nitrobenzene oxidation or by py-rolysis. Among the other probable linkages (L1, L3 and L5), nitrobenzene oxidation would cleave ether linkages between aliphatic and aromatic structures (L1) but not diphenylether bonds (L3 and L5 Ð Viller et al., 1997). Thus, the PCP in both the nitrobenzene oxidation and the pyrolysis products would have origi-nated from the aliphatic side-chain. As stated earlier most of thep-coumaric acid might have been polymer-ized before the incorporation of the PCP. In this case, the side-chain of p-coumaric acid would have been transformed beforehand. However, some aliphatic car-bons in the side-chains or in bridges between benzene rings would still be capable of radical coupling with PCP as exempli®ed by the L1 linkage.
On the other hand, would TeCPs detected only by pyrolysis were probably derived from a diphenylether linkage, because if they were on the side-chain they must also be cleaved and detected by nitrobenzene oxi-dation. Pyrolysis of the diphenylether linkage is thought to be dicult and has not been studied in detail. However, when pyrolyzed, some lignin dipheny-Fig. 3. AOX of gel chromatography fractions (Bio-Gel P-30) from
lethers were cleavable to a remarkable extent. In our hands, 52% of 2-(4-carboxy-2-methoxyphenoxy)-4,5-dimethoxybenzoic acid was cleaved at 5008C, and around 32% of 3-(4-carboxy-2-methoxyphenoxy)-4,5-dimethoxybenzoic acid. Diphenylethers including a polychlorinated phenol such as PCP may cleave readily
due to the reduction of the ring aromaticity by the chlorine substituents.
The oxygen of the ether linkage connecting TeCP to the polymer must have come from p-coumaric acid or its polymer, because any linkage through the hydroxyl group of PCP will cleave to form PCP and not TeCP.
The phenolic p-hydroxyl group (L5) is the only point where p-coumaric acid or its polymer can supply the phenoxy radical to form a diphenylether. This corre-sponds with the observation (Morimoto and Tatsumi, 1997) that the enzymatic transformation of PCP was highly aected by a methoxyl group at the meta pos-ition of the humic precursors. This can be explained by the steric hindrance between them-methoxyl groups and any incoming PCP molecule approaching the p-hydroxyl group.
Thus, we can conclude that the major part of the PCP enzymatically transformed with p-coumaric acid was bound to the polymer through diphenylether lin-kages at thep-hydroxyl groups of p-coumaryl subunits and the remaining PCP was bound to their aliphatic side chains by an ether linkage.
From the direct measurement of chemicals in soil, it is known that the major part of PCP applied to farm-land is retained in the soil and does not easily leach out (Weiss et al., 1982). However the mechanism of retention is not known but adsorption to the soil or-ganic matter is thought to be one of the major causes (Christodoulatos et al., 1994; Christodoulatos and Mohiuddin, 1996). The enzymatic binding of xeno-biotics into humic substances has also been highlighted (Bollag et al., 1992), but in the case of PCP, enzyme catalyzed association with ferulic or syringic acid was quantitatively not so eective. Ruttimann-Johnson and
Lamar (1996) reported on the ability of manganese peroxidase to polymerize PCP with ferulic acid, but lit-tle attention was paid to p-coumaric acid although it is the most abundant humic precursor in various kinds of soils and sediments (Shindo et al., 1978; Matsumoto and Hanya, 1980). Our research shows that peroxidase can eectively catalyze the linkage of PCP to p-couma-ric acid polymer, principally through a diphenylether linkage between the p-hydroxyl group of p-coumaric acid and the benzene ring of PCP by dechlorination at the point of linkage. These results reinforce the im-portance of enzymatic coupling of PCP with humic substances in soil as an immobilization process.
Acknowledgements
We thank Dr Hiroshi Oi (Tsukuba University) for his comments about the interpretation of chromato-Fig. 5. The amounts of pyrolysis products from the dialyzed reaction
solution and their change with incubation.
Fig. 7. Structural analysis of the polymeric product fromp-coumaric acid and PCP. PCP detected by nitrobenzene oxidation and pyrolysis GC/MS; TeCP detected only by pyrolysis GC/MS.
graphic results and Dr Hamish Anderson of the Macanley Land Use Research Institute, Aberdeen for his generous help with the preparation of the ®nal ver-sion of our paper.
References
Bollag, J.M., Chen, C.M., Sarkar, J.M, Michael, J.L, 1987. Extraction and puri®cation of a peroxidase from soil. Soil Biology & Biochemistry 19, 61±67.
Bollag, J.M., Myers, C.J, Minard, R.D, 1992. Biological and chemi-cal interactions of pesticides with soil organic matter. Science of the Total Environment 123/124, 205±217.
Chen, C.L, 1992. Nitrobenzene and cupric-oxide oxidations. In: Stephen, Y.Lin, Carlton, W.Dence (Eds.), Methods in Lignin Chemistry. Springer Verlag, Berlin, pp. 301±321.
Christodoulatos, C., Kor®atis, G.P., Talimcioglu, N.M, Mohiuddin, M, 1994. Adsorption of pentachlorophenol by natural soils. Journal of Environmental Science and Health. Part A: Environmental Science and Engineering 29, 883.
Christodoulatos, C, Mohiuddin, M, 1996. Generalized models for prediction of pentachlorophenol adsorption by natural soils. Water Environment Research 68, 370±378.
Courteix, A, Bergel, A, 1995a. Horseradish peroxidase catalyzed hy-droxylation of phenol. Part 1: Thermodynamic analysis. Enzyme and Microbial Technology 17, 1087±1093.
Courteix, A, Bergel, A, 1995b. Horseradish peroxidase - catalyzed hydroxylation of phenol. Part 2: Kinetic Model. Enzyme and Microbial Technology 17, 1094±1100.
Everse, J., Everse, K.E., Grisham, M.B. (Eds.), 1991. Peroxidases in Chemistry and Biology, vols. I and II. CRC Press, Boca Raton. Kuroda, K., Yamaguchi, A, Sakai, K, 1994. Analysis of Sugi wood
and its lignin preparations by pyrolysis-gas chromatography. Mokuzai Gakkaishi 40, 987±995.
Ladd, J.N, Butler, J.H.A, 1975. Humus-enzyme systems and syn-thetic organic polymer-enzyme analogs. In: Paul, E.A., McLaren, A.D. (Eds.), Soil Biochemistry, vol. 4. Marcel Dekker, New York, pp. 143±194.
Lam, T.B.T., Iiyama, K, Stone, B.A, 1990. Lignin in wheat inte-nodes. Part 2: Alkaline nitrobenzene oxidation by wheat straw lignin and its fractions. Journal of the Science of Food and Agriculture 51, 493±506.
Matsumoto, G., Hanya, T., 1980. Gas chromatographic-mass spec-trometric identi®cation of phenolic acids in recent sediments. Journal of Chromatography 93, 89±94.
Meier, D, Faix, O, 1992. Pyrolysis-gas chromatography-mass spec-trometry. In: Stephan, Y.L., Carlton, W.D (Eds.), Methods in Lignin Chemistry, vol. 34. Springer Verlag, Berlin, pp. 177±199. Morimoto, K, Tatsumi, K, 1997. Eect of humic substances on the
enzymatic formation of OCDD from PCP. Chemosphere 34, 1277±1283.
Oberg, L.G., Glas, B., Swanson, S.E., Rappe, C, Paul, K.G, 1990. Peroxidase-catalyzed oxidation of chlorophenols to polychlori-nated dibenzo-p-dioxins and dibenzofurans. Archives of Environmental Contamination and Toxicology 19, 930±938. Oberg, L.G, Rappe, C, 1992. Biochemical formation of PCDD/Fs
from chlorophenols. Chemosphere 25, 49±52.
Roper, J.C., Sarkar, J.M., Dec, J, Bollag, J.M, 1995. Enhanced enzy-matic removal of chlorophenols in the presence of co-substrates. Water Research 29, 2720±2724.
Ruttimann-Johnson, C, Lamar, R.T, 1996. Polymerization of penta-chlorophenol and ferulic acid by fungal extracellular lignin-degrading enzymes. Applied and Environmental Microbiology 62, 3890±3893.
Ruttimann-Johnson, C, Lamar, R.T, 1997. Binding of pentachloro-phenol to humic substances in soil by the action of white rot fungi. Soil Biology & Biochemistry 29, 1143±1148.
Shindo, H., Ohta, S, Kuwatsuka, S, 1978. Behavior of phenolic sub-stances in the decaying process of plants. IX Distribution of phe-nolic acids in soils of paddy ®elds and forests. Soil Science and Plant Nutrition 24, 223±243.
Sul®ta, J.M, Bollag, .M, 1981. Polymerization of phenolic com-pounds by a soil-enzyme complex. Soil Science Society of America Journal 45, 297±302.
Tanjore, S, Viraraghavan, T, 1994. Pentachlorophenol-water pol-lution impacts and removal technologies. International Journal of Environmental Studies 45, 155±164.
Tatsumi, K., Freyer, A., Minard, R.D, Bollag, J.M, 1994a. Enzyme-mediated coupling of 3,4-dichloroanilline and ferulic acid: a model for pollutant binding to humic materials. Environmental Science and Technology 28, 210±215.
Tatsumi, K., Freyer, A., Minard, R.D, Bollag, J.M, 1994b. Enzymatic coupling of chloroanilines with syringic acid, vanillic acid and protocatechuic acid. Soil Biology & Biochemistry 26, 735±742.
Viller, J.C., Caperos, A, Garcia-Ochoa, F, 1997. Oxidation of hard-wood kraft-lignin to phenolic derivatives. Nitrobenzene and cop-per oxide as oxidants. Journal of Wood Chemistry and Technology 17, 259±285.
Weiss, U.M., Scheunert, I., Klein, W, Korte, F, 1982. Fate of pentachlorophenol-14C in soil under controlled conditions.
