Luchun Duan

Effects of ageing and soil properties on the oral bioavailability of benzo[a]pyrene using a swine model

Oral bioavailability of benzo[a]pyrene (B[a]P) was studied in a swine model using eight spiked soil samples after incubation for 50 and/or 90 days. Silica sand was used as a reference material and the relative bioavailability (RB) of B[a]P in soils was calculated as the quotient of the area under the plasma B[a]P curve (AUC) for soil and AUC for the silica sand. Significantly reduced RB was observed in all study soils after 90 days ageing, ranging from 22.1±0.4% to 62.7±10.1%, except for one very sandy soil (sand content 87.6%) where RB was unchanged (108.1±8.0%). Apart from this, bioavailability decreased during ageing with the decrease (from day 50 to day 90) being only significant for a clayey soil containing expandable clay minerals. Statistical analyses of B[a]P RB at day 90 (eight soils) and soil properties showed no direct correlation between RB and specific soil properties such as total organic carbon (TOC) and clay content which were commonly linked to organic contaminant sequestration. However, strongly significant relationships (p<0.001) were found between RB and the fine particle associated carbon (FPAC) defined as (Silt+Clay)/TOC, and between RB and the soil mesopore (<6nm; p<0.001) fraction, after two samples with high pH and high EC being excluded from the analyses. The bioaccessibility estimated by four in vitro extraction methods: dichloromethane/acetone sonication (DCM/Ace), butanol vortex (BuOH), hydroxypropyl-β-cyclodextrin extraction (HPCD) and Milli Q water leaching methods at different sampling time (1 day, 50 days and 90 days after spiking) also showed a decreasing trend. Significant correlations were found between B[a]P RB and DCM/Ace (R(2)=0.67, p<0.05) extractable fraction and BuOH (R(2)=0.75, p<0.01) extractable fraction.

Effect of ageing on benzo(a)pyrene extractability in contrasting soils

Changes in benzo[a]pyrene (B[a]P) extractability over 160 days ageing in four contrasting soils varying in organic matter content and clay mineralogy were investigated using dichloromethane: acetone 1:1 (DCM/Ace), 60 mM hydroxypropyl-β-cyclodextrin (HPCD) solution, 1-butanol (BuOH) and Milli-Q water. The B[a]P extractability by the four methods decreased with ageing and a first-order exponential model could be used to describe the kinetics of release. Correlation of the kinetic rate constant with major soil properties showed a significant effect of clay and sand contents and pore volume fraction (<6 nm) on sequestration of the desorbable fraction (by HPCD) and the water-extractable fraction. Analysis of (14)C-B[a]P in soils after ageing showed a limited loss of B[a]P via degradation. Fractionation of B[a]P pools associated with the soil matrix was analysed according to extractability of B[a]P by the different extraction methods. A summary of the different fractions is proposed for the illustration of the effect of ageing on different B[a]P-bound fractions in soils. This study provides a better understanding of the B[a]P ageing process associated with different fractions and also emphasises the extraction capacity of the different methods employed.

Pyrogenic carbon in Australian soils

Pyrogenic carbon (PyC), the combustion residues of fossil fuel and biomass, is a versatile soil fraction active in biogeochemical processes. In this study, the chemo-thermal oxidation method (CTO-375) was applied to investigate the content and distribution of PyC in 30 Australian agricultural, pastoral, bushland and parkland soil with various soil types. Soils were sampled incrementally to 50cm in 6 locations and at another 7 locations at 0-10cm. Results showed that PyC in Australian soils typically ranged from 0.27-5.62mg/g, with three Dermosol soils ranging within 2.58-5.62mg/g. Soil PyC contributed 2.0-11% (N=29) to the total organic carbon (TOC), with one Ferrosol as high as 26%. PyC was concentrated either in the top (0-10cm) or bottom (30-50cm) soil layers, with the highest PyC:TOC ratio in the bottom (30-50cm) soil horizon in all soils. Principal component analysis - multiple linear regression (PCA-MLR) suggested the silt-associated organic C factor accounted for 38.5% of the variation in PyC. Our findings suggest that PyC is an important fraction of the TOC (2.0-11%, N=18) and chemically recalcitrant organic C (ROC) obtained by chemical C fractionation method accounts for a significant proportion of soil TOC (47.3-84.9%, N=18). This is the first study comparing these two methods, and it indicates both CTO-375 and C speciation methods can determine a fraction of recalcitrant organic C. However, estimated chemically recalcitrant organic carbon pool (ROC) was approximately an order of magnitude greater than that of thermally stable organic carbon (PyC).

Residual hydrophobic organic contaminants in soil: Are they a barrier to risk-based approaches for managing contaminated land?

Risk-based approaches to managing contaminated land, rather than approaches based on complete contaminant removal, have gained acceptance as they are likely to be more feasible and cost effective. Risk-based approaches aim to minimise risks of exposure of a specified contaminant to humans. However, adopting a risk-based approach over alternative overly-conservative approaches requires that associated uncertainties in decision making are understood and minimised. Irrespective of the nature of contaminants, a critical uncertainty is whether there are potential risks associated with exposure to the residual contaminant fractions in soil to humans and other ecological receptors, and how they should be considered in the risk assessment process. This review focusing on hydrophobic organic contaminants (HOCs), especially polycyclic aromatic hydrocarbons (PAHs), suggests that there is significant uncertainty on the residual fractions of contaminants from risk perspectives. This is because very few studies have focused on understanding the desorption behaviour of HOCs, with few or no studies considering the influence of exposure-specific factors. In particular, it is not clear whether the exposure of soil-associated HOCs to gastrointestinal fluids and enzyme processes release bound residues. Although, in vitro models have been used to predict PAH bioaccessibility, and chemical extractions have been used to determine residual fractions in various soils, there are still doubts about what is actually being measured. Therefore it is not certain which bioaccessibility method currently represents the best choice, or provides the best estimate, of in vivo PAH bioavailability. It is suggested that the fate and behaviour of HOCs in a wide range of soils, and that consider exposure-specific scenarios, be investigated. Exposure-specific scenarios are important for validation purposes, which may be useful for the development of standardised methods and procedures for HOC bioaccessibility determinations. Research is needed to propose the most appropriate testing methods and for assessing potential risks posed by residual fractions of HOCs. Such investigations may be useful for minimising uncertainties associated with a risk-based approach, so that consideration may then be given to its adoption on a global scale. This review critically appraises existing information on the bioavailability of HOC residues in soil to establish whether there may be risks from highly sequestered contaminant residues.

A meta-analysis to correlate lead bioavailability and bioaccessibility and predict lead bioavailability

Defining the precise clean-up goals for lead (Pb) contaminated sites requires site-specific information on relative bioavailability data (RBA). While in vivo measurement is reliable but resource insensitive, in vitro approaches promise to provide high-throughput RBA predictions. One challenge on using in vitro bioaccessibility (BAc) to predict in vivo RBA is how to minimize the heterogeneities associated with in vivo-in vitro correlations (IVIVCs) stemming from various biomarkers (kidney, blood, liver, urinary and femur), in vitro approaches and studies. In this study, 252 paired RBA-BAc data were retrieved from 9 publications, and then a Bayesian hierarchical model was implemented to address these random effects. A generic linear model (RBA (%)=(0.87±0.16)×BAc+(4.70±2.47)) of the IVIVCs was identified. While the differences of the IVIVCs among the in vitro approaches were significant, the differences among biomarkers were relatively small. The established IVIVCs were then applied to predict Pb RBA of which an overall Pb RBA estimation was 0.49±0.25. In particular the RBA in the residential land was the highest (0.58±0.19), followed by house dust (0.46±0.20) and mining/smelting soils (0.45±0.31). This is a new attempt to: firstly, use a meta-analysis to correlate Pb RBA and BAc; and secondly, estimate Pb RBA in relation to soil types.

Comparison of oral bioavailability of benzo[a]pyrene in soils using rat and swine and the implications for human health risk assessment.

BACKGROUND There are many uncertainties concerning variations in benzo[a]pyrene (B[a]P) soil guidelines protecting human health based on carcinogenic data obtained in animal studies. Although swine is recognised as being much more representative of the human child in terms of body size, gut physiology and genetic profile the rat/mice model is commonly used in practice. OBJECTIVES We compare B[a]P bioavailability using a rat model to that estimated in a swine model, to investigate the correlation between these two animal models. This may help reduce uncertainty in applying bioavailability to human health risk assessment. METHODS Twelve spiked soil samples and a spiked silica sand (reference material) were dosed to rats in parallel with a swine study. B[a]P bioavailability was estimated by the area under the plasma B[a]P concentration-time curve (AUC) and faecal excretion as well in the rats. Direct comparison between the two animal models was made for: firstly, relative bioavailability (RB) using AUC assay; and secondly, the two assays in the rat model. RESULTS Both AUC and faecal excretion assays showed linear dose-response for the reference material. However, absolute bioavailability was significantly higher when using faecal excretion assay (p<0.001). In aged soils faecal excretion estimated based on solvent extraction was not accurate due to the form of non-extractable fraction through ageing. A significant correlation existed between the two models using RB for soil samples (RBrat=0.26RBswine+17.3, R(2)=0.70, p<0.001), despite the regression slope coefficient revealing that the rat model would underestimate RB by about one quarter compared to using swine. CONCLUSIONS In the comparison employed in this study, an interspecies difference of four in RB using AUC assay was identified between the rat and swine models regarding pharmacokinetic differences, which supported the body weight scaling method recommended by US EPA. Future research should focus on the carcinogenic competency (pharmacodynamics) used in experiment animals and humans.

Variation in arsenic bioavailability in rice genotypes using swine model: An animal study

An in vivo assay using swine was used to measure the absolute bioavailability (AB) of As from cooked rice of twelve genotypes commonly grown in Bangladesh. An assessment of both total As in rice and its bioavailability is crucial for estimating human exposure following dietary intake by the local community. Average As concentrations in each rice genotype ranged from 108±4μg/kg to 580±6μg/kg. Arsenic speciation shows that most of the rice genotype contains 73 to 100% inorganic As. Swine were administered with As orally and via intravenous method, i.e. injection and fed certain common Bangladeshi rice genotypes (cooked). Swine blood As levels were measured to calculate As bioavailability from rice. Pilot studies shows that for As(III) and As(V), 90.8±12.4% and 85.0±19.2% of the administered oral dose was absorbed from the gastrointestinal tract whereas organic As was poorly absorbed resulting in low bioavailability values 20.2±2.6% (MMA) to 31.2±3.4% (DMA), respectively. These studies demonstrates that rice genotypic characters influenced As bioavailability in rice grown in As-contaminated areas and the bioavailability varied between 25% and 94%. Arsenic in salt tolerant rice genotypes Binadhan-10 and BRRI dhan47 as well as brown rice genotypes Kheali Boro and Local Boro has lower bioavailability (<50%) compared to other rice genotypes. The most commonly cultivated and consumed variety (BRRI dhan28) has As bioavailability of 70%, which poses a significant risk to consumers. Calculation of maximum tolerable daily intake (MTDI) for humans due to consumption of rice based on bioavailability data was higher than those calculated based on inorganic and organic As concentration in rice genotypes.

Issues raised by the reference doses for perfluorooctane sulfonate and perfluorooctanoic acid

On 25th May 2016, the U.S. EPA released reference doses (RfDs) for Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoic Acid (PFOA) of 20ng/kg/day, which were much more conservative than previous values. These RfDs rely on the choices of animal point of departure (PoD) and the toxicokinetics (TK) model. At this stage, considering that the human evidence is not strong enough for RfD determination, using animal data may be appropriate but with more uncertainties. In this article, the uncertainties concerning RfDs from the choices of PoD and TK models are addressed. Firstly, the candidate PoDs should include more critical endpoints (such as immunotoxicity), which may lead to lower RfDs. Secondly, the reliability of the adopted three-compartment TK model is compromised: the parameters are not non-biologically plausible; and this TK model was applied to simulate gestation and lactation exposures, while the two exposure scenarios were not actually included in the model structure.

Abiotic factors controlling bioavailability and bioaccessibility of polycyclic aromatic hydrocarbons in soil: Putting together a bigger picture

The bioavailability and bioaccessibility of polycyclic aromatic hydrocarbons (PAHs) in soil underpin the risk assessment of contaminated land with these contaminants. Despite a significant volume of research conducted in the past few decades, comprehensive understanding of the factors controlling the behaviour of soil PAHs and a set of descriptive soil parameters to explain variations in PAH bioavailability and bioaccessibility are still lacking. This review focuses on the role of source materials on bioavailability and bioaccessibility of soil PAHs, which is often overlooked, along with other abiotic factors including contaminant concentration and mixture, soil composition and properties, as well as environmental factors. It also takes into consideration the implications of different types of risk assessment (ecological and human health) on bioavailability and bioaccessibility of PAHs in soil. We recommend that future research should (1) account for the effects of source materials on bioavailability and bioaccessibility of soil PAHs; (2) adopt non-disruptive methods to analyse soil components controlling PAH sequestration; (3) integrate both natural organic matter (NOM) and xenobiotic organic matter (XOM) while evaluating the influences of soil organic matter (SOM) on the behaviour of PAHs; and (4) consider the dissimilar desorption scenarios in ecological risk assessment and human health risk assessment while assessing PAH bioavailability and bioaccessibility.

A Pooled Data Analysis to Determine the Relationship between Selected Metals and Arsenic Bioavailability in Soil

Chronic exposure to arsenic (As) is a global concern due to worldwide exposure and adverse effects, and the importance of incorporating bioavailability in the exposure assessment and risk assessment of As is increasing acknowledged. The bioavailability of As is impacted by a number of soil properties, such as pH, clay and metal concentrations. By retrieving 485 data from 32 publications, the aim of this study was to determine the relationship between selected metals (Fe and Al) and As bioavailability. In present study, the bioaccessibility (BAC) data measured by in vitro approaches were converted into bioavailability data based on the previously determined relationship between BAC and bioavailability. The As relative bioavailability (RBA) was summarized to be 24.36 ± 18.49%, which is in the range previously reported. A significant association between Fe concentration and the bioavailability of As was observed while this association varied for different types of RBA data. This disparity may suggest a non-reliable association between Fe and As bioavailability. The correlations between logarithmically transformed total content of Fe + Al and As bioavailability is then outlined: RBA = (−8.40 ± 1.02) × Ln(Fe + Al) + (58.25 ± 4.09), R2 = 0.25, p < 0.001, n = 212. Jackknife resampling was also applied to validate the relation between total content of (Fe + Al) and As bioavailability, which suggested that the relation is robust. This is the first pooled study to address the relations between selected metal concentrations and As bioavailability, which may provide some implications to establish soil properties-based RBA prediction for As.