Waste wood recycling as animal bedding and development of bio-monitoring tool using the CALUX assay

Environment International 30 (2004) 639 – 649 www.elsevier.com/locate/envint

Waste wood recycling as animal bedding and development of bio-monitoring tool using the CALUX assay Misuzu Asari a,*, Hiroshi Takatsuki a, Michifumi Yamazaki b, Tomonori Azuma b, Hidetaka Takigami c, Shin-ichi Sakai c b

a Environment Preservation Center, Kyoto University, Kyoto 606-8501, Japan Hokkaido Forest Products Research Institute, 1-10 Nishikagura, Asahikawa, Hokkaido 071-0198, Japan c National Institute for Environmental Studies, Onogawa 16-2, Tsukuba, Ibaraki 305-8506, Japan

Received 2 October 2003; accepted 3 December 2003

Abstract Animal bedding made of waste wood samples from seven different plants in Japan were chemically analyzed in terms of persistent organic pollutants (POPs) including polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/DFs), coplanar polychlorinated biphenyls (Co-PCBs), drin compounds, chlordane compounds and various inorganic toxic compounds (Cr, Cu, As, B, Cd and Pb) to investigate the chemical characteristics and levels of contamination. Further investigation was conducted to determine the success of applying the Chemically Activated Luciferase Expression (CALUX) bioassay to the waste wood samples in combination with a cleanup procedure for the detection of dioxin-like compounds in order to develop the CALUX bioassay as a rapid and cost-effective screening/monitoring method and a contributive tool to risk management in the waste wood recycling process. For the cleanup procedure, crude extracts from wood samples were prepared by dimethylsulfoxide (DMSO)/n-hexane extraction, and then the extracts were processed by silica gel – 44% sulfuric acid reflux treatment at 70 jC for 60 min to yield the bioassay fractions. The presence of POPs and inorganic toxic compounds were confirmed in most of the litter samples. In particular, Co-PCBs in one sample (litter dust) showed a high concentration level (1,200,000 pg/g, 240 pg TEQ/g), suggesting the potential for contamination from demolition waste. The CALUX assay-determined TEQs (CALUX-TEQs) were significantly high in the sample after DMSO/n-hexane extraction, probably due to labile aryl hydrocarbon receptor (AhR) ligands such as PAHs; however, they were remarkably reduced through a single silica gel – 44% sulfuric acid reflux treatment. The ratio between CALUX-TEQ values and WHO toxicity equivalent values (WHO-TEQ) obtained by congener-specific chemical analysis ranged from 0.058 to 22 and show comparatively good agreement. Underestimation in some samples, however, was observed where WHO-TEQ values of Co-PCBs contributed greatly to total WHO-TEQ values. Reasons for this gap could be lower CALUX assay-determined relative potencies (REPs) than the WHO-TEFs for these congeners or AhR-antagonistic effects of non dioxin-like PCBs which coexist at higher concentration than Co-PCBs. The CALUX assay is proposed as a promising application in the recycling process of wooden materials. D 2004 Elsevier Ltd. All rights reserved. Keywords: Wood; CALUX; Persistent organic pollutants

1. Introduction A large amount of waste wood has been and will continue to be released each year. Instead of waste incineration and landfill, wood recycling has accelerated over the years. In

* Corresponding author. Tel.: +81-75-753-7709; fax: +81-75-7537710. E-mail address: [email protected] (M. Asari). 0160-4120/$ - see front matter D 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2003.12.002

addition to paper and paperboard, fiberboard products, and particleboard, other items made of recovered waste wood could be included in mulch and animal bedding. Although forestry by-products such as sawdust and bark are commonly used for animal bedding in addition to agricultural waste such as husk and hay, a lack of litter has arisen even in some areas of Hokkaido, which is located in northern Japan and known as a major area in forestry production, because of the increase in the number of cattle contrary to the decrease of forest products. Waste

640

M. Asari et al. / Environment International 30 (2004) 639–649

wood from construction has become one of the promising alternative resources. The amount of waste wood litter produced as by-products from construction totaled 110,000 tons in 2000, which corresponds to 35% of waste wood chips from construction in Hokkaido (Horie, 2002; Seino, 2002). However, the use of waste wood as animal bedding poses a potential concern regarding insufficient removal of hazardous contaminants from waste flows such as demolition wastes. Some contaminants are potentially included in wood preservatives, such as chromated copper arsenate (CCA), creosote consisting of various polycyclic aromatic hydrocarbons (PAHs), chlorophenols (CPs) and polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/DFs) as impurities in CPs, and organochlorine insecticides such as drin compounds and chlordane compounds (Yasuhara et al., 2003; Sakai et al., 2001). Since animal bedding is used in direct contact with livestock and then thrown into pastures or farm soil possibly after being composted, potential contaminants pose a risk of dermal, inhalant and oral exposure. There is a high possibility of some contaminants, potentially including wood preservatives, accumulating in the body of cattle, resulting in hazardous effects even on humans through consumption of meat and dairy products. PCDD/DFs are potentially included in waste wood and are one of the highly prioritized persistent pollutants that have raised concerns about health effects on humans and wildlife. Pentachlorophenol (PCP) of a technical grade has been used mainly as an agricultural chemical (herbicide) in Japan, especially during the period from 1960 to 1972. Although we did not obtain quantitative data, application of CPs including PCP for wood preservation seems to have been relatively common during that period (Inoue, 1972; Suzuki, 1998). In the United States, the US Environmental Protection Agency (EPA) estimated that the use of PCP as a wood preservative between 1970 and 1995 amounted to 400,000 metric tons, and that PCDD/DFs between 4800 and 36,000 g TEQ were incorporated annually into treated wood (EPA, 2000). There have been some contributions to the estimation of considerable PCDD/DFs release from this treated wood (Winters et al., 1999; Lorber et al., 2002). PCP-treated wood has also been identified as one of the potential sources for PCDD/ DFs release in Europe (Eduljee and Dykke, 1996; Dyke et al., 1997). In terms of the profile characteristics for PCDD/ DF congeners in technical-grade PCP, the prominence of highly chlorinated dioxins is well known (Masunaga et al., 2001). Similar profiles in PCP-treated wood or waste wood suggest that the residual PCDD/DFs are impurities of PCP (Sakai et al., 2001; Asari et al., 2002; Lorber et al., 2002). The levels of PCP and PCDD/DFs in PCP-treated wood or waste wood have been varied. Fries et al. (2002) collected wood samples from livestock facilities, analyzed them for PCP and PCDD/DFs, and classified the results by the relationship between the chemical concentrations and the methods of PCP application: low-level application to

prevent fungus growth (PCP: 2500– 82,000 ng/g, PCDD/ DFs: 0.016 –2.4 ng TEQ/g), intermediate-level application by surface treatment (PCP: 110,000 –880,000 ng/g, PCDD/ DFs: 1.6 – 53 ng TEQ/g) and high-level application by pressure treatment (PCP: 1,600,000 – 8,500,000 ng/g, PCDD/DFs: 27 – 310 ng TEQ/g), according to a US Department of Agriculture standard. This classification could also be applied to waste wood research in Japan (Asari et al., 2002). Furthermore, the presence of waste wood treated with PCP including 10 ppb TEQ levels of PCDD/DFs in addition to high levels of PAHs in creosote was found in that research, although some practices of CPs and creosote in combination were reported, but without detailed data (Engwall et al., 1999; Wan and Oostdam, 1995). Therefore, PCP-treated wood has also been hypothesized as an important source of dioxin in milk and beef in cases where it is used around livestock facilities (Fries et al., 1999; Huwe et al., 1999). To control its contamination in the food chain, limitsetting for the presence of dioxin in food and feed came into force in EU (IP/01/1698, IP/01/1670) and an integrated and systematic approach coupled with monitoring has been designed in Europe (EC Press Releases, 2002; Hoogenboom, 2002; Verstraete, 2002). The objectives of this study were to investigate the chemical characteristics of waste wood samples and to develop the Chemically Activated Luciferase Expression (CALUX) bioassay (Sanderson et al., 1996; Garrison et al., 1996) as a rapid and cost-effective screening/monitoring method and a contributive tool to risk management in the waste wood recycling process. The CALUX bioassay has been applied to various matrices in humans, the environment, feed/food and waste recycling, and has shown good applicability as a bio-monitoring tool for dioxin-like compounds (Legler et al., 2002; Murk et al., 1996; Hoogenboom, 2002; Behnisch et al., 2002; Takigami et al., 2002). Waste wood samples including litter were selected and firstly analyzed in terms of PCDD/DFs, coplanar polychlorinated biphenyls (Co-PCBs), drin compounds, chlordane compounds and various inorganic toxic compounds (Cr, Cu, As, B, Cd and Pb). Further investigation was conducted to determine the success of applying the CALUX bioassay to waste wood samples in combination with a cleanup procedure especially for the detection of chemically stable and persistent dioxinlike polyhalogenated aromatic hydrocarbons (PHAHs).

2. Materials and methods 2.1. Sample collection and preparation In December 2002, litter samples made of waste wood were collected from seven different plants (A – G) in Hokkaido. Generally, waste wood litter is chipped in two steps. Foreign material such as metal fragments, rubble and litter dust is often removed during chipping. A detailed description of each plant, origin of the waste

M. Asari et al. / Environment International 30 (2004) 639–649

wood, type of operation for sorting and dust treatment is summarized in Table 1. Waste wood chips aimed at energy recovery, litter dust, litter made of waste plywood and that of virgin sawdust were also sampled by random collection from several areas of each stockyard and mixed well for preparation. The samples were dried for about two weeks at room temperature (around 15 jC) under ambient pressure in a well-ventilated clean draft chamber. Then, the samples were broken into pieces ( < 5 mm) using a crusher. Waste wood chips used for manufacturing particleboard were obtained for reference. To design and develop a cleanup procedure of waste wood samples for the CALUX assay, a waste wood sleeper that contained high concentration levels of PCP, PCDD/DFs and PAHs (Asari et al., 2002) and untreated scraps for children’s handicrafts and toys were used. The sleeper sample was a cross-section of used railroad sleeper pressure-treated with PCP in combination with creosote. Wood subsamples were collected from several grams of sawn materials. Table 1 Description of the plants from which litter samples were obtained Plant

Origin of waste wood

Pollutants removal

Dust treatment

Commercial products

A

Waste wood from other’s demolition Packaging

No operation for treated wood Magnetic selector Treated wood judged by appearance Magnetic selector Treated wood judged by appearance

Removal

Fuel chips (main)

B

C

Waste wood from own demolition Demolition from others Waste wood from other’s demolition Packaging

D

Waste wood from other’s demolition Crates

E

Waste wood from other’s demolition Packaging

F

Waste wood from own demolition Waste wood from other’s demolition Waste wood chips from others

G

Litter Removal

Litter

Mixing

Fuel chips

Treated wood judged by appearance Magnetic selector Treated wood judged by appearance Magnetic selector Treated wood judged by appearance Magnetic selector

Removal

Chips for litter (Plant G) Litter

Mixing

Litter

Removal

Litter

No operation for treated wood Magnetic selector

Removal

Litter

641

2.2. Chemical analysis 2.2.1. PCDD/DFs and Co-PCBs For analysis of PCDD/DFs and Co-PCBs, wood samples were Soxhlet-extracted with toluene for 16 h. Crude extracts were added to 13C12-labeled internal standards and concentrated. The samples were then processed by silica gel – 44% sulfuric acid reflux treatment (hexane and silica gel – sulfuric acid were mixed with the crude extracts) at 70 jC for 60 min. The hexane layers were concentrated and cleaned up using multilayer silica gel column chromatography (AgNO3 – silica/H2SO4 – silica/KOH – silica). This procedure was conducted in accordance with the Japanese Industrial Standard (JIS) method K 0312 (JIS KO312, 1999). The eluates were concentrated and further fractionated using activated carbon dispersed silica gel column chromatography; samples were eluted with hexane (Fr.1: this fraction was discarded), followed by 25% dichloromethane – hexane (Fr.2: containing mono – ortho PCBs) and toluene (Fr.3: containing non-ortho PCBs and PCDD/Fs). Each fraction was concentrated and a recovery standard was added. PCDD/DFs and Co-PCBs were identified and quantified by high-resolution capillary column gas chromatography (HP6890; Agilent Technologies, Palo Alto, USA) coupled with high-resolution mass spectrometry (AutoSpec Ultima E; Micromass, Manchester, UK); GC columns for 4– 6, 7– 8 CDD/Fs and PCBs were SP-2331 (Spelco, Bellfonte, USA), DB-17 (J&W Scientific, Folson, USA) and DB-5 (J&W Scientific), respectively. 2.2.2. Organochlorine insecticides For analysis of drin and chlordane compounds, the interim manual on exogenous endocrine disrupting chemicals (Japanese Ministry of the Environment, 1998) was adopted. An internal standard (13C12-p,pV-DDT) was added to the samples, which were then ultrasonically extracted with acetone. The acetone layers were concentrated and 5% NaCl – water was added. The acetone – water layers were extracted with hexane. The extracts were concentrated and cleaned up using Florisil column chromatography (eluted with 20% diethylether –hexane). The eluates were further cleaned up using a SEP CARTRIDGE CARBOGRAPH (GL Sciences, Tokyo, Japan) and a Sep-Pak Plus NH2 (Waters, Milford, USA); samples were eluted with 15% acetone– hexane. The eluates were concentrated and a recovery standard (d10-fluoranthene) was added. Aldrin, dieldrin, endrin, trans-chlordane, cis-chlordane, trans-nonachlor, cis-nonachlor and oxychlordane were identified and quantified by a high-resolution gas chromatograph (HP6890) coupled with a quadrupole mass spectrometer (HP5973; Agilent Technologies); a GC column was HP-5ms (Agilent Technologies). 2.2.3. Inorganic toxic compounds For analysis of inorganic toxic compounds, samples were wet-digested with nitric acid/sulfuric acid by microwave:

642

M. Asari et al. / Environment International 30 (2004) 639–649

MLS-1200MEGA (Milestone, Monroe, USA). The digests were concentrated to dryness and then dissolved with hydrochloric acid/nitric acid. For cadmium, chromium, copper and lead, the digests were filtered and the residues were further dissolved using alkali fusion with sodium carbonate. Metals in the dissolutions were quantified by inductively coupled plasma –atomic emission spectrometer (ICP-AES): OPTIMA3000XL (Perkin-Elmer, Wellesley, USA). For arsenic and boron, the digests were filtered and the filtrate was analyzed. Arsenic was quantified using a hydride generation atomic adsorption spectrometer (HGAAS). Boron was quantified by ICP-AES. 2.3. CALUX assay 2.3.1. Cleanup strategy Crude extract from 10 g of each wood sample was prepared by Soxhlet-extraction with toluene (16 h) following dimethylsulfoxide (DMSO)/n-hexane extraction (Larsen et al., 1991). The extracts were processed by silica gel – 44% sulfuric acid reflux treatment at 70 jC for 60 min. The effectiveness of this method was previously reported for samples of fly ash (Behnisch et al., 2002) and PCBcontaining mineral oil (Takigami et al., 2002). For the sleeper sample and the untreated sample, silica gel –44% sulfuric acid reflux treatment was repeated four times and each fraction was collected separately in addition to the crude extracts. Based on a comparison of the CALUX results (i.e., CALUX-TEQ) of each fraction with the chemical analysis results (i.e., WHO-TEQ), the optimal number of reflux times was determined. Then, for the litter samples, assay fractions were obtained after DMSO/n-hexane extraction and silica gel –44% sulfuric acid reflux treatment the optimized number of times. All the fractions were evaporated and replaced with a small volume of DMSO for the CALUX bioassay. 2.3.2. CALUX assay The recombinant rat hepatoma H4IIE cell line, stably transfected with the aryl hydrocarbon receptor (AhR)-controlled luciferase-cDNA construct (Garrison et al., 1996) (DR-CALUXR cell line), was obtained from Bio Detection Systems (Amsterdam, The Netherlands) and the CALUX assay was carried out as previously described (Behnisch et al., 2002). The cells were plated in 96-well plastic plates in 0.1-ml a-minimal essential medium supplemented with 10% fetal bovine serum per well. Following 24-h incubation, the cells with at least 95% confluence were dosed in triplicate by adding 0.1 ml of the above medium supplemented with the chemical or extract to be tested dissolved in DMSO (0.4%). After 24-h treatment, the dose medium was removed and the exposed wells were filled with 0.1 ml PBS containing 1 mM calcium and magnesium ions. PBS was renewed and 0.1 ml LucLiteR assay substrate (PerkinElmer) was added per well. Thirty minutes after the addition

of the assay substrate, luciferase activity was measured by luminometer for 10 s/well. The TCDD standard dose – response curve was fitted using a cumulative fit function of Slide Write Plus Ver. 6.00 (Advanced Graphics Software). CALUX assay-determined TEQs (CALUX-TEQs) for the tested fractions were obtained from their dilutions so that luciferase activities were in the reproducible lower part of the linear range corresponding to 1 –4 pM in TCDD. The limit of quantification (LOQ) was calculated as the luciferase activity elicited by the solvent control plus ten times the standard deviation. In general, the LOQ of CALUX assay is almost below 1 pM (60 fg/well) and the limit of detection (LOD) is below 0.3 pM (20 fg/well) in TCDD (Behnisch et al., 2003). For selectivity in general, CALUX bioassay should be able to determine TEQ values selectivity as the sum of PCDDs, PCDFs and dioxin-like PCBs comparable to WHO-TEQ values (Behnisch et al., 2003). The percent standard deviation should not be above 10% in a triplicate determination for each sample dilution and not above 30% between three independent experiments.

3. Results and discussion 3.1. Chemical concentrations in samples 3.1.1. PCDD/DFS and Co-PCBs Chemical concentrations of PCDD/DFs and Co-PCBs in the litter samples and other wood samples are shown in Table 2. The concentrations of PCDD/DFs showed significantly high values (21,000 pg TEQ/g) in the sleeper sample compared with that in the untreated wood sample (0.30 pg TEQ/g), possibly due to pressure treatment with PCP in combination with creosote. The level is in the range of highlevel application by pressure treatment (27,000 – 310,000 pg TEQ/g) according to the classification for PCP-treated wood (Fries et al., 2002) and the predominance of OCDD to congener profiles of PCDD/DFs (see Fig. 1) represent the characteristics of PCP impurities, as reported previously (Asari et al., 2002). Concentrations of PCDD/DFs in waste wood litter samples varied between 0.0052 and 0.73 pg TEQ/g. Most of them were one order of magnitude greater than that in natural sawdust. Their levels, however, did not reflect the presence of PCP-treated waste wood in the litter samples, in contrast to a sample from a wooden livestock facility treated with PCP (Fries et al., 2002), and their congener profiles (not shown here) also did not resemble those in the PCP-treated samples. Litter dust B contained a high concentration of CoPCBs (1,200,000 pg/g, 240 pg TEQ/g). The concentration of Co-PCBs in Litter B, Litter C and Fuel chip C also showed relatively high values (37,000 pg/g, 5.6 pg TEQ/g; 18,000 pg/g, 8.6 pg TEQ/g; 4800 pg/g, 3.6 pg TEQ/g, respectively). Congener and homologue profiles showed a

Table 2 Chemical concentrations in litter and other wood samples Origin

Plant B

Plant C

Fuel chip A

Litter A

Litter B

Litter dust B

Fuel chip C

PCDDs TEQ PCDFs TEQ PCDD/DFs TEQ PCBs Co-PCBs TEQ Total TEQ Aldrin Dieldrin Endrin trans-Chlordane cis-Chlordane trans-Nonachlor cis-Nonachlor Oxychlordane Cr Cu As B Cd Pb

pg/g pg TEQ/g pg/g pg TEQ/g pg/g pg TEQ/g pg/g pg/g pg TEQ/g pg TEQ/g ng/g ng/g ng/g ng/g ng/g ng/g ng/g ng/g Ag/g Ag/g Ag/g Ag/g Ag/g Ag/g

280 0.39 16 0 290 0.39 110,000 2700 0.36 0.76 ND < 2 * ND < 3 1.9 1.2 0.99 0.34 ND < 20 5.5 4.1 0.32 14 ND < 2.0 4.5

490 0.32 75 0 570 0.32 96,000 3200 0.42 0.74 ND < 5 * ND < 3 18 11 9.4 3.4 ND < 10 38 21 6.0 18 ND < 2.0 61

68 0.0052 60 0 130 0.0052 350,000 37,000 5.6 5.6 ND < 2 * ND < 30 4.2 2.4 0.2 1.0 ND < 10 8.5 12 1.7 8.8 ND < 2.0 14

420 0.32 270 0.43 690 0.73 16,000,000 1,200,000 240 240 ND < 5 * ND < 50 9.5 4.8 0.4 1.9 ND < 20 21 61 2.6 15 ND < 2.0 110

190 0.015 N.D. 0 190 0.015 110,000 4800 3.6 3.6 ND < 40 * ND < 30 13 5.3 0.8 1.5 ND < 50 3.1 12 6.5 11 ND < 2.0 13

Plant D

Plant E

Plant F

Plant G

Litter C

Litter D

Litter E

Litter F

Litter G

Plywood litter G

120 0.010 N.D. 0 120 0.010 800,000 18,000 8.6 8.6 ND < 1 * ND < 10 ND < 0.5 0.65 0.53 ND < 0.5 ND < 1 ND < 2.0 ND < 4.0 1.2 9.0 ND < 2.0 ND < 4.0

1400 0.85 200 0.0095 1600 0.86 120,000 1600 0.22 1.1 ND < 1 * ND < 5 14 8.9 7.3 2.6 ND < 2 < 2.0 < 4.0 1.4 17 ND < 2.0 < 4.0

460 0.032 28 0 480 0.032 140,000 1700 0.24 0.27 ND < 8 * ND < 2 21 11 0.7 4.1 ND < 20 14 6.8 6.6 7.0 < 2.0 8.4

340 0.031 48 0 390 0.031 250,000 1400 0.20 0.23 ND < 5 * ND < 10 130 84 5.7 22 ND < 30 6.1 7.7 15 15 ND < 2.0 4.6

86 0.0066 8.4 0 94 0.0066 150,000 1300 0.19 0.19 ND < 2 * ND < 60 ND < 0.5 15 ND < 0.5 ND < 1 ND < 10 5.1 ND < 4.0 2.9 18 ND < 2.0 ND < 4.0

7.2 0 2.4 0 9.6 0 100,000 1900 0.26 0.26 ND < 2 * ND < 5 ND < 0.5 ND < 0.5 ND < 0.5 ND < 0.5 ND < 50 ND < 2.0 ND < 4.0 ND < 0.3 18 ND < 2.0 ND < 4.0

Sawdust

Sleeper

Waste wood chips

Untreated wood

34 0 3.2 0 37 0 120,000 1300 0.19 0.19 ND < 5 * ND < 30 4.9 2.7 2.3 ND < 10 ND < 50 ND < 2.0 ND < 4.0 1.3 5.3 ND < 2.0 ND < 4.0

11,000,000 17,000 980,000 4000 12,000,000 21,000 57,000 6700 0.93 21,000 ND < 150 * ND < 100 ND < 1 ND < 1 ND < 1 ND < 1 ND < 5 13 16 0.41 8.7 ND < 2.0 46

210 0.016 120 0.92 330 0.94 130,000 4100 0.51 1.4 ND < 10 * ND < 10 12 6.2 0.6 3.4 ND < 2 ND < 2.0 ND < 4.0 0.53 15 ND < 2.0 ND < 4.0

73 0.0060 55 0 130 0.0060 140,000 2200 0.30 0.30 ND < 4 * ND < 1 ND < 0.5 ND < 0.5 ND < 0.5 ND < 0.5 ND < 5 ND < 2.0 ND < 4.0 ND < 0.3 13 ND < 2.0 ND < 4.0

M. Asari et al. / Environment International 30 (2004) 639–649

Plant A

Sample

Unit wet-base

*: Not determined (unable to be quantitated due to insufficient separation of peaks). ND for PCDD/DFs and Co-PCBs: Not detected at the limit of detection; 2 pg/g for TeCDD/DFs and PeCDD/DFs, 6 pg/g for HxCDD/DFs, HpCDD/DFs and Co-PCBs, 10 pg/g for OCDD/DF. ND for PCBs: Not detected at the limit of quantification; 40 pg/g. ND for drin, chlordane and metals: Not detected at the limit of detection. TEQ: WHO-TEQ (1998) values were calculated assuming that the amounts of congeners below the detection limit were zero.

643

644

M. Asari et al. / Environment International 30 (2004) 639–649

Fig. 1. Congener profiles of PCDD/DFs in the sleeper sample.

large contribution of PCB #77, #105, #118 and #156 to total TEQ, and TeCBs, PeCBs and HxCBs to total PCBs (Fig. 2). This suggests contamination from commercial

PCBs because PCB #77, #105 and #118 are known to be predominant congeners in commercial PCBs (Takasuga et al., 1995; Alcock et al., 1998). These products may

Fig. 2. Congener and homologue profiles of Co-PCBs in Litter B, Litter Dust B, Fuel chip C and Litter C.

M. Asari et al. / Environment International 30 (2004) 639–649

cause contamination to wooden structures during demolition. For example, stabilizers in fluorescent lights or home appliances, elastic sealant material with technical mixtures of PCBs or paints and agents containing PCBs, and PCBcontaminated wood by these routes would be potential contaminants. The Ordinance on the Management of Waste Wood (Waste Wood Ordinance), which came into force on 1 March 2003 in Germany, drew attention to insulating board and sound-insulating board treated with agents containing PCBs. Besides such products, other sources such as oil used during the chipping process or sample handling might also be considered as PCB pollution sources. 3.1.2. Organochlorine pesticides The chemical concentrations of aldrin and endrin in all the samples were either below the detection limits or not determined. However, trans-chlordane, cis-chlordane, transnonachlor, cis-nonachlor and oxychlordane were detected; the levels of concentration are shown in Table 2. Although all of them were below the detection limits in Plywood Litter G, trans-chlordane, cis-chlordane and transnonachlor were detected in the virgin sawdust litter. This pesticide contamination could be caused by the waste wood litter (Litter A) kept nearby in the same stockyard of the livestock facility. The concentration of trans-chlordane in Litter F (134 ng/ g) was one order of magnitude higher than that in other samples and similar to the average value in background soil samples (Meijer et al., 2003). This level was, however, two orders of magnitude lower than that in waste wood (12,000 ng/g) sprayed with chlordane, which was collected from underneath the floor of a Japanese house where chlordane had been applied as an anti-termite agent (Yasuhara et al., 2003). Therefore, it could not be clarified whether or not the sample was contaminated by a specific resource. 3.1.3. Inorganic toxic compounds Concentrations of chromium, copper, arsenic, boron, cadmium and lead are shown in Table 2. Especially for Litter dust B, the concentration of lead (110 Ag/g) was one or more orders of magnitude greater than that in other samples, and contamination seemed to originate from some foreign material. Concentrations of CCA compounds (Cr, Cu, As) and boron were two or more orders of magnitude lower than levels of waste wood treated with CCA- or boron-based preservatives (Sakai et al., 2001). These levels, however, could not be discounted compared to the set values in the Waste Wood Ordinance in Germany. In this ordinance, limit values for wood chips used in the manufacture of derived timber products must be kept at less than 30, 20, 2, 2 and 30 Ag/g for chromium, copper, arsenic, cadmium and lead, respectively. Compared with these limit values, Litter A and Litter dust B investigated in this study exceeded the limits in Cr, Cu, As and Pb for Litter A and in Cu and As for Litter dust B.

645

3.1.4. Different production operations and hazardous chemical concentrations As for Plant A, fuel chips are the main product and litter is the by-product made by collecting small particles during the chipping process. In comparing Fuel chip A with Litter A, the chemical concentrations of all the compounds determined were 1– 20 times higher in Litter A than in Fuel chip A. Similarly, as for Plant B where litter is the main product, the chemical concentrations in Litter dust B were 2 –140 times higher than those in Litter B. These represent the transfer of chemicals to small particles through the chipping operation and also suggest that removal of dust may be effective in eliminating chemical contamination. Plant C adopted a different sorting system of litter materials from that employed in other plants, where untreated wood materials were selected one by one by manually checking their appearance to remove colored materials (i.e., CCA- or creosote-treated materials). Indeed, concentrations of inorganic compounds including CCA, PCDD/DFs and chlordane compounds were higher in Fuel chip C made of waste wood during the sorting process, compared to Litter C made of the sorted material, indicating the effect of sorting. However, concentrations of Co-PCB congeners in Litter C were 2.4– 6 times higher than those in Fuel chip C. This suggests the difficulty in completely avoiding contamination of hazardous chemicals. Litter made of waste plywood apparently has no issues in terms of the investigated chemical contamination. Comparison of the different plants showed no clear relationship between sorting processes (time, cost and labor) and chemical concentrations determined in litter products. The results revealed the difficulty in controlling chemical contamination and its risk by only sorting the materials by appearance. In conclusion, this study clarified the presence of persistent organic pollutants (POPs) and inorganic toxic compounds in most of the investigated litter samples. However, the source of this contamination is not clearly known. The relationship between chemical levels in litter should be discussed considering their risk. 3.2. Optimization of cleanup procedure of waste wood samples for the CALUX assay To design a cleanup procedure of waste wood samples for the CALUX assay, the sleeper sample and the untreated wood sample were used. The WHO toxicity equivalent (WHO-TEQ) value obtained by congener-specific chemical analysis in the sleeper sample was significantly high (21,000 pg TEQ/g) and that in the untreated wood sample was low (0.3 pg TEQ/g) as shown before (Table 2). Fig. 3 shows CALUX-TEQ values at each level of the cleanup procedure for the sleeper and untreated wood samples. WHO-TEQ values for the two samples are indicated by the broken lines. The CALUX-TEQ value

646

M. Asari et al. / Environment International 30 (2004) 639–649

Fig. 3. CALUX-TEQ values at each level of cleanup procedure for the sleeper and untreated wood samples compared with WHO-TEQ values (broken lines). All samples were analyzed three times in independent experiments with the CALUX bioassay; the CV of all these tests had a mean value of 6.6% (2.5 – 11%).

was remarkably reduced through a single silica gel – 44% sulfuric acid reflux treatment after which the values became stable in spite of repeated reflux treatment. In the case of the sleeper sample, after DMSO/n-hexane extraction, the CALUX-TEQ value was 1.5  1011 pg/g although the WHO-TEQ value was 21,000 pg/g. This shows that a considerable amount of AhR agonists might exist in the sample. Creosote components remaining in the sleeper include various PAHs with relatively high molecular weight such as benzo[k]fluoranthene or dibenz[a,h]ananthracene (Asari et al., 2002), which demonstrate high CALUX activity (Machala et al., 2001). These PAHs could remain in the sample after DMSO/n-hexane extraction and contribute significantly to such high induction. However, the value was reduced to 23,000 pg CALUX-TEQ/g after just a single silica gel – 44% sulfuric acid reflux treatment. Also, for the untreated wood sample, the high CALUXTEQ value (1000 pg/g) for crude fraction obtained after DMSO/n-hexane extraction dropped significantly to 0.97 pg/g through a single silica gel –44% sulfuric acid reflux treatment. The ratio between the CALUX-TEQ value and WHOTEQ value (CALUX-TEQ/WHO-TEQ ratio) in the sleeper sample was 7,100,000 after DMSO/n-hexane extraction, and this became 1.1 after a single silica gel – 44% sulfuric acid reflux treatment. The CALUX-TEQ/WHO-TEQ ratio for the untreated wood sample became 3.2 through a single silica gel–44% sulfuric acid reflux treatment. The CALUXTEQ values for the four refluxed samples prepared stepwise all agreed well with the WHO-TEQ values. These results showed that labile AhR ligands such as PAHs in the samples were effectively removed by the proposed single reflux method and that this cleanup was available for detection devoted to PCDD/DFs and Co-PCBs in wood samples. The results were consistent with our CALUX results (good agreement with WHO-TEQ) obtained after conducting silica gel –44% sulfuric acid reflux treatment using samples of fly ash (Behnisch et al., 2002) and PCB-containing mineral oil (Takigami et al., 2002).

3.3. Bio-TEQ and chemical TEQ for litter samples CALUX-TEQ values for the fraction obtained after a single silica gel –sulfuric acid reflux treatment, WHO-TEQ values and CALUX-TEQ/WHO-TEQ ratio for the litter samples are summarized in Table 3 and Fig. 4. CALUX-TEQ values ranged from N.D. to 14 pg/g and WHO-TEQ values ranged from 0.19 to 240 pg/g, leading to CALUX-TEQ/WHO-TEQ ratios in the range from 0.058 to 22. The LOQ of the CALUX assay for these samples was around 1 pg TEQ/g using 10 g of sample materials, which was considered to show a satisfactory level of sensitivity for monitoring the TEQ levels of waste wood samples. Fur-

Table 3 CALUX-TEQ values after a single silica gel – 44% sulfuric acid reflux treatment, WHO-TEQ values and their ratio

Fuel chip A Litter A Litter B Litter dust B Fuel chip C Litter C Litter D Litter E Litter F Litter G Plywood litter G Sawdust litter

CALUXTEQ (pg/g)

WHO-TEQ (pg/g)

2.2

0.39

3.3 5.2 14

PCDD/ DFs

Co-PCBs

Total

CALUX-TEQ/ total WHO-TEQ

0.36

0.76

2.9

0.32 0.0052 0.73

0.42 5.6 240

0.74 5.6 240

6.5

0.015

3.6

3.6

1.8

2.1 11 6.0 2.2 1.2 N.D.

0.010 0.86 0.032 0.031 0.0066 0

8.6 0.22 0.24 0.20 0.19 0.26

8.6 1.1 0.27 0.23 0.19 0.26

0.24 10 22 9.7 6.1 –

N.D.

0

0.19

0.19

–

4.4 0.94 0.058

All samples were analyzed three times in independent experiments with the CALUX bioassay; the CV of all these tests had a mean value of 4.5% (2.3 – 8.5%).

M. Asari et al. / Environment International 30 (2004) 639–649

Fig. 4. Profiles of CALUX-TEQ (after a single silica gel – 44% sulfuric acid reflux treatment) and WHO-TEQ values obtained for litter samples.

thermore, LOQ could be favorably reduced to a level of 0.1 pg TEQ/g when 100 g of waste wood sample was applied. CALUX/WHO-TEQ ratios ranged from 1.8 to 20 when concentrations of PCBs in samples were below 350,000 pg/ g. On the other hand, CALUX-TEQ values were underestimated WHO-TEQ values, and CALUX/WHO-TEQ ratios ranged from 0.058 to 0.94 when concentrations of PCBs in samples were above 350,000 pg/g. This showed that the CALUX result was affected by the PCB concentration. In the underestimated samples (Litter B, Litter dust B and Litter C), WHO-TEQ values of Co-PCBs predominated in the total WHO-TEQ values (approximately 100%). This suggests that one reason for this underestimation could be lower relative potencies (REPs) of these congeners as measured in CALUX bioassay than WHO-TEFs for these congeners. Some studies showed lower REP values for several PCB congeners with CALUX or other in vitro assays (Besselink et al., 2003; Behnisch et al., 2001a,b). In Litter B, Litter dust B and Litter C, concentrations determined by chemical analysis of total PCBs were 350,000, 16,000,000 and 800,000 pg/g, respectively. This suggests that another reason for underestimation could be due to the antagonistic effect of non-planar PCBs, a remarkable amount of which are considered to exist in PCBs together with Co-PCBs. Although homologue profiles represent significant contributions of PeCBs and HxCBs for Litter B, Litter dust B and Litter C (Fig. 2), this contamination is presumably due to a mixture of Japanese technical PCBs, Kanechlor-500 (KC500) and Kanechlor-600 (KC600), which consist mainly of PeCBs and HxCBs, respectively (Takasuga et al., 1995). The antagonistic effects were obtained in the CALUX assay using KC500 and KC600; WHO-TEQ values were 16,000,000 and 7,700,000 pg/g, CALUX-TEQ values were 2,300,000 and 810,000 pg/g, resulting in CALUX-TEQ/WHO-TEQ ratios of 0.14 and 0.11, respectively (Japan Environment Cooperation, 2003). This antagonistic effect was presumed to be from nonplanar 2-4 ortho PCB congeners. The antagonistic

647

interactions between PHAHs and some nonplanar di-ortho PCB congeners, as well as certain technical PCB mixtures, on several toxicity parameters have been previously reported. Simone et al. (2000) studied the AhR-mediated induction of luciferase activity of the planar 0-1 ortho and the nonplanar 2-4 ortho fraction of the commercial PCB mixture Aroclor 1260 (corresponding to KC600) using the DR-CALUX assay, and they observed the antagonistic interaction between the 2-4 ortho-substituted PCBs and the planar congeners. To evaluate the mechanism responsible for antagonism by certain PCBs, Suh et al. (2003) studied the antagonistic activity of some di-ortho-substituted PCB congeners (PCB #47, #52, #128, #153) when present in combination with AhR agonists using three types of bioassay. They reported that the examined di-orthosubstituted PCB antagonized the induction of CYP1A1, the inhibition of IgM expression by AhR agonists and the agonist-induced AhR-DNA-binding activity in the CH12.LX murine B cell line. In addition, they suggested that PCB #52 inhibited TCDD-induced AhR-DNA-binding to a dioxin-responsive element. However, the mechanism of antagonistic interaction has not been clearly shown. A further cleanup method such as activated carbon column chromatography or gel permeation chromatography column is required to remove nonplanar PCBs. As a matter of course, however, the cleanup procedure would become more complicated by the additional step. Higher estimates of CALUX-TEQ over 10-fold CALUXTEQ/WHO-TEQ ratios seen in litter samples D –F indicated the presence of unknown dioxin-like compounds included in the complex mixture of contaminated waste wood. They may potentially include potent PCDD/DF congeners except 17 PCDD/DFs, potent PCBs except 12 Co-PCBs, and mixed PXDD/DFs, PXBs, PXNs, PXDEs (X = Br, Cl, F) (Behnisch et al., 2001a,b). Furthermore, considering the complex mixture, organic and inorganic toxic pollutants, as well as PHAHs, coexisted in waste wood. The result of chemical analysis for litter samples (Table 2) represented high concentrations of inorganic toxic compounds including CCA in addition to dioxin-like compounds. Although the interaction of such mixtures has not been clarified, it was reported that metals may modulate gene expression through signal-transduction pathways not previously associated with these metals, and that this finding suggests new directions for future studies into the biological mechanisms of toxicity and carcinogenicity of these metals (Mumtaz et al., 2002). This may also suggest new directions for studies on toxicological mechanisms of complex mixtures including both organic and inorganic pollutants. The CALUX bioassay in combination with the developed cleanup procedures using silica gel – sulfuric acid reflux treatment is found to be a promising screening method because of the level of preferable agreement with WHO-TEQ values. This method, however, requires further study to obtain better agreement of CALUX-TEQ and

648

M. Asari et al. / Environment International 30 (2004) 639–649

WHO-TEQ values in PCB-rich samples that might cause underestimation of TEQ for the CALUX assay. Targetsetting for POP monitoring and further application of the CALUX method to various types of wood samples are also required in future research. However, practical use of the CALUX assay was found to be a rapid and cost-effective screening/monitoring tool for dioxin-like POPs in the recycling process of wooden materials.

Acknowledgements We are grateful for the financial support of this study provided by the Grant-in-Aid for the Development of Innovative Technologies from the Japanese Ministry of Education, Culture, Sports, Science and Technology. We also wish to thank Prof. A. Brouwer of Bio Detection Systems.

References Alcock RE, Behnisch PA, Jones KC, Hagenmaier H. Dioxin-like PCBs in the environment—human exposure and the significance of sources. Chemosphere 1998;37:1457 – 72. Asari M, Takatsuki H, Sakai S. Polycyclic aromatic hydrocarbons, pentachlorophenol, dioxins and furans in waste wooden sleepers on the recycling market. Organohalogen Compounds 2002;59:199 – 202. Behnisch PA, Hosoe K, Sakai S. Bioanalytical screening methods for dioxins and dioxin-like compounds—a review of bioassay/biomarker technology. Environment International 2001a;27:413 – 39. Behnisch PA, Hosoe K, Sakai S. Combinatorial bio/chemical analysis of dioxin and dioxin-like compounds in waste recycling, feed/food, humans/wildlife and the environment. Environment International 2001b;27:495 – 519. Behnisch PA, Hosoe K, Shiozaki K, Ozaki H, Nakamura K, Sakai S. Low-temperature thermal decomposition of dioxin-like compounds in fly ash: combination of chemical analysis with in vitro bioassays (EROD and DR-CALUX). Environmental Science & Technology 2002;36:5211 – 7. Behnisch PA, Hosoe K, Sakai S. Brominated dioxin-like compounds: in vitro assessment in comparison to classical dioxin-like compounds and other polyaromatic compounds. Environment International 2003;29: 861 – 77. Besselink H, Jonas A, Pijnappels M, Swinkels A, Fjellanger K, Brouwer B. Comparison of DR-CALUX and HRGCMS-derived TEQs: introduction of conversion factors. Organohalogen Compounds 2003;60:203 – 6. Dyke PH, Foan C, Wenborn M, Coleman PJ. A review of dioxin releases into land and water in the UK. Science of the Total Environment 1997;207:119 – 31. Eduljee GH, Dykke P. An update of potential PCDD and PCDF emission sources in the UK. Science of the Total Environment 1996;177:303 – 21. Engwall MA, Pignatello JJ, Grasso D. Degradation and detoxification of the wood preservatives creosote and pentachlorophenol in water by photofenton reaction. Water Research 1999;33:1151 – 8. European Commission Press Releases; 2002. IP/02/959. Fries GF, Paustenbach DJ, Mather DB, Luksemburg WJ. A congener specific evaluation of transfer of chlorinated dibenzo-p-dioxins and dibenzofurans to milk of cows following ingestion of pentachlorophenol-treated wood. Environmental Science and Technology 1999;33: 1165 – 70. Fries GF, Feil VJ, Zaylskie RG, Bialek KM, Rice CP. Treated wood in

livestock facilities: relationships among residues of pentachlorophenol, dioxins, and furans in wood and beef. Environmental Pollution 2002; 116:301 – 7. Garrison PM, Tullis K, Aarts JMMG, Brouwer A, Giesy JP, Denison MS. Species-specific recombinant cell lines as bioassay systems for the detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin-like chemicals. Fundamental and Applied Toxicology 1996;30:194 – 203. Hoogenboom R. The combined use of the CALUX bioassay and the HRGC/HRMS method for the detection of novel dioxin sources and new dioxin-like compounds. Organohalogen Compounds 2002;55: 11 – 6. Horie H. Potential use of waste wood from demolition and deforestation as animal litter. Journal of the Hokkaido Forest Products Research Institute 2002;1:6 – 9. Huwe JK, Feil VJ, Larsen GL, Zaylskie RG. Pentachlorophenol levels in beef adipose tissue as an indicator of treated-wood exposure. Organohalogen Compounds 1999;43:101 – 3. Inoue K. Wood degradation and protection. Tokyo, Japan: Morikita Publishing; 1972. Japan Environment Cooperation. Report of research project on chemical properties of stockpile PCB wastes [Tokyo, Japan (in Japanese)] Japan Environment Cooperation. Japanese Ministry of the Environment. The interim manual on exogenous endocrine disrupting chemicals; 1998. JIS (Japan Industrial Standards) K0312. Method for determination of tetrathrough octa-chlorodibenzo-p-dioxins, tetra- through octa-chlorodibenzofurans and coplanar polychlorobiphenyls in industrial water and waste water; 1999. Larsen B, Tilio R, Kapila S. A DMSO-based cleanup procedure for determination of PCBs in waste oil. Chemosphere 1991;23:1077 – 84. Legler J, Jonas A, Lahr J, Vethaak AD, Brouwer A, Murk AJ. Biological measurement of estrogenic activity in urine and bile conjugates with in vitro ER-CALUX reporter gene assay. Environmental Toxicology and Chemistry 2002;21:473 – 9. Lorber MN, Barton RG, Winters DL, Bauer KM, Davis M, Palauski J. Investigation of the potential release of polychlorinated dioxins and furans from PCP-treated utility poles. Science of the Total Environment 2002;290:15 – 39. Machala M, Vondracek J, Blaha L, Ciganek M, Neca J. Aryl hydrocarbon receptor-mediated activity of mutagenic polycyclic aromatic hydrocarbons determined using in vitro reporter gene assay. Mutation Research 2001;497:49 – 62. Masunaga S, Takasuga T, Nakanishi J. Dioxin and dioxin-like PCB impurities in some Japanese agrochemical formulations. Chemosphere 2001;44:873 – 85. Meijer SN, Shoeib M, Jantunen LMM, Jones KC, Harner T. Air-soil exchange of organochlorine pesticides in agricultural soils: 1. Field measurements using a novel in situ sampling device. Environmental Science & Technology 2003;37:1292 – 9. Mumtaz MM, Tully DB, EI-Masri HE, Rosa CTD. Gene induction studies and toxicity of chemical mixtures. Environmental Health Perspectives 2002;110:947 – 56. Murk AJ, Legler J, Denison MS, Giesy JP, Guchte C, Brouwer A. Chemical-activated luciferase gene expression (CALUX): a novel in vitro bioassay for Ah receptor active compounds in sediments and pore water. Fundamental and Applied Toxicology 1996;33:149 – 60. Sakai S, Asari M, Takastuki H. Waste wood treated with preservatives and behavior of POPs. Journal of Japan Society for Safety Engineering 2001; 40(6):411 – 9. Sanderson JT, Aarts JMMJG, Brouwer A, Froese KL, Denison MS, Giesy JP. Comparison of Ah receptor-mediated luciferase and ethoxyresorufinO-deethylase induction in H4IIE cells: implication for their use as bioanalytical tools for the detection of polyhalogenated aromatic hydrocarbons. Toxicology and Applied Pharmacology 1996;137:316 – 25. Seino S. Present condition and subject for recycling of waste wood from demolition. Journal of the Hokkaido Forest Products Research Institute 2002;10:6 – 7.

M. Asari et al. / Environment International 30 (2004) 639–649 Simone AP, Sundberg H, Berg H, Scheu G, Wester P, Jensen S, et al. Contribution of planar (0-1 ortho) and nonplanar (2-4 ortho) fractions of aroclor 1260 to the induction of altered hepatic foci in female Sprague – Dawley rats. Toxicology and Applied Pharmacology 2000; 169:255 – 68. Suh J, Kang JS, Yang KH, Kaminski NE. Antagonism of aryl hydrocarbon receptor-dependent induction of CYP1A1 and inhibition of IgM expression by di-ortho-substituted polychlorinated biphenyls. Toxicology and Applied Pharmacology 2003;187:11 – 21. Suzuki K. Wood preservation. Wood Industry 1998;53:512 – 5. Takasuga T, Inoue T, Ohi E. All congener specific analytical method for polychlorinated biphenyls (PCBs) with various chromatographic cleanup and HRGC/HRMS. Journal of Environmental Chemistry 1995;5: 647 – 75. Takigami H, Hosoe K, Behnish PA, Shiozaki K, Mizukami H, Ohno M, et al. Validation study for practical bio-monitoring of waste PCB samples during their destruction treatment using DR-CALUX assay and PCB immunoassay. Organohalogen Compounds 2002;58:397 – 400.

649

US EPA. Exposure and human health reassessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Part 1: estimating exposure to dioxin-like compounds. Sources of Dioxin-Like Compounds in the United States. Draft final report. National center for environmental assessment, United States Environmental Protection Agency 2000;2. EPA/600/P-00/001Bb. Verstraete F. Development and implementation of an EC strategy on dioxins, furans and dioxin-like PCBs in food and feed. Organohalogen Compounds 2002;55:1 – 4. Wan MT, Oostdam JV. Utility and railway rights-of-way contaminants: dioxins and furans. Journal of Environmental Quality 1995;24: 257 – 65. Winters DL, Barton RG, Boggess KE, Davis M, Alburty DS, Lorber MN. A field study to evaluate the potential for the release of dioxins from pentachlorophenol-treated utility poles. Organohalogen Compounds 1999;43:1 – 5. Yasuhara A, Katami T, Shibamoto T. Formation of PCDDs, PCDFs, and coplanar PCBs from incineration of various woods in the presence of chlorides. Environmental Science & Technology 2003;37:1563 – 7.