Transformation and bioaccessibility of lead during physiologically based extraction test: effects of phosphate amendment and extract fluid components

Abstract

In this study, a physiologically based extraction test (PBET) was used to determine the effects of the fluid components and soluble phosphate amendment on the transformation and bioaccessibility of Pb(NO₃)₂ and PbCO₃(s) during the extraction. In the gastric phase, phosphate amendment reduced up to 95% bioaccessible Pb in both Pb(NO₃)₂ and PbCO₃(s) within 10 min, with the decrease being continued with time. The decrease in Pb bioaccessibility was attributed to the formation of insoluble Pb phosphate precipitates (i.e., Pb₅(PO₄)₃Cl(s) and PbHPO₄(s)), which were verified and quantified by X-ray diffraction. The continuous decrease of Pb bioaccessibility resulted from the lasting transformation of Pb phosphate precipitates over the extraction time. In the intestinal phase, the decrease of Pb bioaccessibility with P amendment was over 60%, but it increased slightly during the extraction process. This was probably due to dissolution of formed Pb carbonates or phosphates through competitive complexation of the fluid-inherent pepsin or organic acid with Pb under neutral condition. Complexation of free Pb²⁺ with organic components in the fluid, including pepsin, organic acid, pancreatin or bile inhibited the formation of insoluble Pb phosphate and resulted in an increase of Pb bioaccessibility in both gastric and intestinal phases. Results from this study indicated that both phosphate amendment and fluid components impacted the transformation and bioaccessibility of Pb, and that the one-time measurement using PBET assay may not accurately estimate Pb bioaccessibility in the P-rich soil or the contaminated soil amended with P.

---

1. Introduction

Incidental ingestion of Pb-contaminated soil via hand-to-mouth activity is a primary pathway of Pb exposure in young children.¹ Assessment of Pb bioavailability in contaminated soil has received considerable attention since Pb has adverse effects on children's cognitive development.^2,3 Animal-based in vivo evaluation of Pb bioavailability is difficult due to its operational and ethical considerations.⁴ Therefore, several in vitro approaches have been developed to assess Pb bioaccessibility in attempts to mimic the effects of the human digestion process in the past two decades. The common approaches include in vitro gastrointestinal (IVG),⁵ the solubility bioaccessibility research consortium (SBRC),⁶ physiologically based extraction test (PBET),⁷ and unified BARGE method (UBM) assays.⁸ Among them, PBET assay presents good correlations with in vivo animal results on Pb bioavailability.⁷ It turns out to be a classic and mature in vitro method to assess Pb bioaccessibility in soils in the past.^9–11

Generally, Pb bioaccessibility in soil depends on its solubility in human gastrointestinal environment.¹² The solubility of Pb from soil in the stimulated track may be influenced by the soil site-specific properties and physical or chemical forms of Pb. Research to date has illustrated that the Pb speciation can be affected by various organic or inorganic constituents in soils. Cao et al.¹³ reported the formation of insoluble Pb phosphate minerals (e.g., Pb₅(PO₄)₃Cl and Pb₅(PO₄)₃OH) in Pb-contaminated soil with phosphate amendment, resulting in decrease of Pb bioaccessibility. Scheckel et al.¹⁵ attributed the decreased soil Pb bioaccessibility to the formation of Pb₅(PO₄)₃Cl upon addition of phosphate-rich soft drink. In addition, Juhasz et al.¹⁴ observed that Pb relative bioavailability decreased in soils following the phosphate amendments using an in vivo mouse model. It is possible that Pb bioaccessibility could be further reduced in simulated ingestion solution containing phosphate source because insoluble Pb phosphate minerals might be formed during the extraction test. Pb speciation could be changed as it moved through the gastrointestinal tract, which may facilitate the formation of insoluble Pb phosphates. Therefore, phosphate amendments-induced speciation transformation during extracting process needs to be considered as it plays an important role in the bioaccessibility assessment from incidental soil ingestion.

In vitro methods were normally conducted with various simulated components at different pHs, which could result in the variability of Pb bioaccessibility. Oomen et al.¹⁶ compared five in vitro models and concluded that gastric pH was a major source of variability on bioaccessibility. The difference in organic constituents of stimulated fluids is another factor affecting bioaccessibility. Oomen et al.¹⁸ suggested that the formation of Pb-bile complexes increased soluble Pb concentration in intestinal phase. What is more, Smith et al.¹⁷ showed that the PBET assay which contained several organic acids and pepsin could inhibit As sorption onto iron oxide surfaces at the modified pH (pH = 1.5) in gastric phase, thus increasing As bioaccessibility compared to SBRC assay which only contained glycine. A recent research demonstrated that human gut microbiota in colon phase induced changes from inorganic As to organic As, leading to a higher As bioaccessibility.¹⁹

Identification of Pb speciation and transformation during the extraction processes is considered to be conducive to assess in vitro bioaccessibility, and bioaccessibility results would clearly influence the risk assessment of incidental soil ingestion. Therefore, this study was conducted to demonstrate the transformation and bioaccessibility of Pb during the PBET assay process for Pb(NO₃)₂ and PbCO₃(s) upon Ca(H₂PO₄)₂ amendment. The two Pb compounds were selected with the consideration of Pb(NO₃)₂ standing for soluble Pb species and PbCO₃(s) representing the less soluble Pb species. The objectives were (1) to investigate the impact of P amendment on Pb transformation of two different Pb compounds during the extraction procedure, (2) to understand the impact of Pb transformation on Pb bioaccessibility, and (3) to determine the effect of fluid components on Pb bioaccessibility in gastrointestinal phase.

2. Material and methods

2.1 Material characterization

All chemicals used in this study including Pb(NO₃)₂, PbCO₃(s), and Ca(H₂PO₄)₂·2H₂O were analytical grade and purchased from Sinopharm Chemical Reagent Co, Ltd. All solutions were prepared in the deionized water (18 MΩ cm).

2.2 In vitro Pb bioaccessibility assessment

To evaluate Pb bioaccessibility, the physiologically based extraction test (PBET) was employed in this study with a minor modification.⁷ The assay simulated process in human gastrointestinal tract, and included two phases: the gastric phase (G-P) at pH = 2 followed by the intestinal phase (I-P) at pH = 7.

Gastric extraction fluid was prepared by dissolving 1.25 g of pepsin, and four common organic acids (0.50 g of citric acid, 0.50 g of DL-malic acid, 420 μL of lactic acid, and 500 μL of acetic acid) in 1 L of Millipore water. The fluid pH was then adjusted to 2.00 ± 0.05 with concentrated HCl. A high Pb level at 4.80 mM of Pb was introduced by adding Pb(NO₃)₂ or PbCO₃(s) to a 200 mL PBET fluid in the screw-capped 250 mL bottles. The Pb concentration is equivalent to 10 000 mg kg⁻¹ in soil, which is often observed in some heavily contaminated soils in China.^20,21 In addition, the high concentration of Pb was selected in this study with the aim to meet the objective that we tried to find the Pb speciation in solid using XRD analysis which has 1% Pb (10 000 mg kg⁻¹) detection limit. The soluble P amendment (Ca(H₂PO₄)₂·2H₂O) was added with a molar ratio of P/Pb at 2 : 1, which was in excess of the stoichiometry of P/Pb = 0.6 to ensure Pb₅(PO₄)₃Cl(s) formation.¹³ The extractions rotated end-over-end at 30 rpm for the designated time of 0.17 h, 1 h, and 3 h. Note that the standard time for the PBET gastric phase is 1 h;⁷ three time points were chosen to determine the change in the speciation and bioaccessibility of Pb with time. After the extraction, 50 mL of the solution for each was filtered through 0.45 μm nitrocellulose membrane filters. The solid retained on the filter was vacuum-dried for mineral analysis.

In the intestinal phase, the remaining gastric fluid after 1 h extraction (standard time) was adjusted to pH = 7.0 ± 0.2 with saturated NaHCO₃ solution followed by the addition of 100 mg pancreatin and 350 mg bile salts. The samples were continuously tumbled at 30 rpm at pH = 7. After extraction for 1 h, 4 h and 12 h, the solution was filtered through 0.45 μm nitrocellulose membrane filters. Note that the standard time for the PBET intestinal phase is 4 h;⁷ three time points were chosen to determine the change in the speciation and bioaccessibility of Pb with time. The solid retained on the filter was vacuum-dried for mineral analysis. All gastric and intestinal phase extractions were performed in three replicates for each sample.

2.3 Effect of fluid components on Pb bioaccessibility

To examine the effects of PBET fluid components on Pb bioaccessibility, three experiments were carried out in the gastric phase: (i) the one excluded pepsin in the fluid (non-pepsin); (ii) the other excluded organic acid in the fluid (non-organic acid); and (iii) the third one excluded both pepsin and organic acid in the fluid (non-(pepsin + organic acid)). The three experiments represented the effect of pepsin, organic acid and the combined effect of pepsin and organic acid components on the bioaccessibility, respectively.

After the standard gastric extraction, the other three experiments were carried out in intestinal phase: (i) the one excluded pancreatin in the fluid (non-pancreatin); (ii) the other excluded bile salt in the fluid (non-bile salt); (iii) and the third one excluded both pancreatin and bile salt in the fluid (non-pancreatin + bile salt). The three experiments represented the effect of pancreatin, bile salt and the combined effect of pancreatin and bile salt components on the bioaccessibility, respectively.

All the experiments were carried out with three replicates at the standard time, i.e., 1 h for the gastric extraction and 4 h for the intestinal extraction. After the designated time extraction, the solution was filtered through 0.45 μm nitrocellulose membrane filters. The solid retained on the filter was vacuum-dried for mineral analysis.

2.4 Lead speciation modeling

Visual MINTEQ modeling²² was used to elucidate possible Pb speciation and transformation during the PBET assay. The Pb(NO₃)₂, PbCO₃(s), HCl and organic components (glycine, citric acid, malic acid, lactic acid, and acetic acid) were added at the initial concentrations in the model. Note that glycine was chosen to stand for pepsin in this section because no thermodynamic data of Pb-pepsin was included in the database. Glycine has the similar organic ligands as pepsin, such as carboxyl (–COOH) or amino (–NH₂).²³ The modeling can help to predict the distribution of Pb species in liquid and solid phases, and together with the XRD analysis, it was able to reveal Pb transformation during the extraction.

2.5 Analytical methods

The pH was measured using a pH/Ion 510 Bench Meter (Pte Ltd/Oakon Instruments). Lead concentration was determined using either a flame atomic absorption spectrometry (AAS; novAA700 Jena AAS) or inductively coupled plasma optical emission spectrometer (ICP-OES; 7500A, Agilent Corporation, USA), depending on its concentration. Phosphorus concentration was measured by the molybdenum–ascorbic acid method.²⁴ All results were expressed as the average of three replicates. The minerals of solid phases retained on the filter were identified using X-ray diffraction (XRD), and the XRD samples were a mixture of the three triplicates. Samples were vacuum-dried and manually pulverized to pass a no. 400 U.S. standard sieve. Step-scanned XRD data was collected by a Bruker D8 Advance X-ray Diffractometer equipped with Cu Kα radiation conducted at 40 kV and 30 mA. The data was collected between angles (2θ) of 10°and 50°. The XRD patterns were analyzed using the Jade Software (Ver. 6.0). Following identification of the mineral species present in the samples, mineral quantification was achieved using the Rietveld refinement technique.

3. Results and discussion

3.1 Pb bioaccessibility from Pb(NO₃)₂ and PbCO₃(s)

The bioaccessibility of Pb compounds would depend on their dissolution rates.^25,26 As shown in Fig. 1a, the bioaccessibility of Pb from Pb(NO₃)₂ and PbCO₃(s) rapidly increased to about 90% within 0.17 h in acidic gastric phase and kept a slow increase over the 3 h extraction period, which is generally considered as the longest residence time of food in children's stomach.⁴ The Pb bioaccessibility of Pb(NO₃)₂ and PbCO₃(s) in the gastric phase after the standard 1 h tumbling was 96.0% and 92.3%, respectively (Fig. 1a), with the predominant Pb forms in the gastric solution as organic-complex (>65%) and free Pb²⁺ (>20%) based on the modeling results (data not shown). Note that PbCO₃(s) is generally stable and less soluble (K_sp = 3.30 × 10⁻¹⁴),²⁶ it tends to be readily dissolved under the acidic conditions like the acidic gastric fluid at pH = 2. Thus, PbCO₃(s) had similar Pb bioaccessibility as Pb(NO₃)₂ (Fig. 1a). This was consistent with Palumbo-Roe et al.²⁷ who found relatively high bioaccessibility (range 20–100%) of Pb in PbCO₃-containing soil from the abandoned mining areas in UK. Nevertheless, soluble Pb(NO₃)₂ still had higher Pb bioaccessibility than less soluble PbCO₃ (Fig. 1a).

Fig. 1 Bioaccessibility of Pb for two Pb compounds in the gastric phase (G-P) (a) and intestinal phase (I-P) (b) as functions of P amendment and extracting time.

In the intestinal phase, the dissolved Pb in the acidic gastric fluid including free Pb²⁺ and Pb–organic complex was further reduced in the neutral (pH = 7) intestinal environment and decreased greatly as the extraction time increased (Fig. 1b). The Pb bioaccessibility for both Pb compounds was less than 0.1% after 4 h extraction in the intestinal phase, which may be attributed to the precipitation of PbCO₃ due to addition of NaHCO₃ to the intestinal fluid.¹⁵ Continuous formation of PbCO₃(s) resulted in the decrease in Pb bioaccessibility of Pb(NO₃)₂ and PbCO₃ as extraction time increased (Fig. 1b).

Overall, Pb(NO₃)₂ showed higher Pb bioaccessibility than PbCO₃, which was consistent with their solubility. But Pb bioaccessibility for both Pb compounds failed to reach equilibrium completely in both gastric and intestinal phases at their standard time of 1 h and 4 h, respectively.

3.2 Effect of phosphate amendment on Pb bioaccessibility

It is well documented that Pb is readily precipitated as less soluble Pb-phosphate minerals, such as Pb₅(PO₄)₃Cl or Pb₃(PO₄)₂ when adequate amount of P is present.²⁸ Compared to the test without P, the addition of P greatly reduced Pb bioaccessibility of both Pb compounds in the gastric phase, and the Pb bioaccessibility decreased with increasing time (Fig. 1a). It was assumed that acidic gastric fluid would induce dissolution of Pb particles, which provided more labile Pb for continuous formation of Pb-phosphate minerals. It was also possible that some conversions of Pb minerals would occur during the extraction based on their thermodynamic stability.

When P was added to Pb(NO₃)₂ in the acid gastric fluid, a white precipitate was instantly formed. After 0.17 h, the molar ratio of changed P and Pb (ΔP/ΔPb) in solution was 1 (Fig. 2), equivalent to the stoichiometric value of 1 for PbHPO₄. With the extraction time increasing to 3 h, the molar ratio of ΔP/ΔPb gradually decreased to 0.59 (Fig. 2), close to the stoichiometric value of 0.6 for Pb₅(PO₄)₃Cl(s). Presumably, the dissolved Pb was prone to PbHPO₄ formation initially because H₃PO⁰₄ was the dominate species at pH < 2.12,²⁹ and later PbHPO₄(s) might be gradually converted to more stable Pb₅(PO₄)₃Cl(s) based on the thermodynamic equilibrium. The formation of Pb–P minerals was confirmed by the results of XRD which showed peaks of PbHPO₄(s) at 2θ ∼ 26.8° and Pb₅(PO₄)₃Cl(s) at 2θ ∼ 30.2° (Fig. 3a). The changes in peak intensities of those Pb phosphate minerals were not obvious over time from 0.17 h to 3 h, therefore, it was hard to identify the mineral transformation from Fig. 3a. However, refinement analysis of the XRD patterns could support the transformation occurrence. The percentage of PbHPO₄ decreased from 61.1% to 21.3% when the time increased from 0.17 h to 3 h, whereas the percentage of Pb₅(PO₄)₃Cl(s) increased from 34.9% to 76.4% in the same period (Table 1). Eventually, the precipitated Pb in gastric phase increased from 96.0% to 97.7% (Table 1), as a result, Pb bioaccessibility of Pb(NO₃)₂ decreased from 4.1% to 2.7% when the time increased from 0.17 h to 3 h (Fig. 1b).

Fig. 2 The molar ratio of reduced P/Pb (ΔP/ΔPb) for two Pb compounds with P amendment at 0.17 h, 1 h, and 3 h during the gastric phase.

Fig. 3 X-ray diffraction patterns of Pb(NO₃)₂ (a) and PbCO₃ (c) with P amendment in the gastric phase (G-P) and the patterns of Pb(NO₃)₂ (b) and PbCO₃ (d) with P amendment in intestinal phase (I-P) as functions of extracting time. Minerals with peaks labeled: 1 = Pb₅(PO₄)₃Cl; 2 = PbHPO₄; 3 = PbCO₃.

Table 1 Percentage (%) of solid Pb species among total Pb in the PBET process of gastric phase (G-P) and intestinal phase (I-P) as a function of extracting time calculated by XRD refinement

Original Pb compounds	Solid Pb formed	G-P			I-P
		0.17 h	1 h	3 h	1 h	4 h	12 h
Pb(NO3)2	PbHPO4(s)	61.1	51.0	21.3	21.5	20.7	20.1
	Pb5(PO4)3Cl(s)	34.9	45.9	76.4	77.5	77.1	77.5
	PbCO3(s)	—	—	—	0.7	1.1	1.1
	Total	96.0	96.9	97.7	99.7	98.9	98.7
PbCO3(s)	PbHPO4(s)	5.2	1.9	1.7	2.7	2.6	2.6
	Pb5(PO4)3Cl(s)	40.1	45.9	48.9	46.3	46.3	47.1
	PbCO3(s)	50.9	50.2	47.5	50.8	50.1	49.3
	Total	96.2	98.0	98.1	99.8	99.0	99.0

---

Similarly, the addition of P reduced Pb bioaccessibility of PbCO₃(s) in the gastric fluid and Pb bioaccessibility decreased from 3.7% to 1.7% as the time increased from 0.17 h to 3 h (Fig. 1a). The decrease in Pb bioaccessibility was attributed to the formation of PbHPO₄(s) and Pb₅(PO₄)₃Cl(s) which was evidenced by XRD analysis showing the peaks of PbHPO₄(s) and Pb₅(PO₄)₃Cl(s) at 2θ ∼ 29.8° and 2θ ∼ 30.2° (Fig. 3c), respectively. Those peaks were broad (Fig. 3c) indicating that the formed Pb phosphate minerals were not well crystallized. Furthermore, the molar ratios of ΔP/ΔPb were much less than stoichiometric value of insoluble Pb–P minerals, such as PbHPO₄(s) or Pb₅(PO₄)₃Cl(s) (Fig. 2). All the observations implied that transformation of PbCO₃ into PbHPO₄(s) or Pb₅(PO₄)₃Cl(s) was limited. PbCO₃(s) dissolution was mainly transport-controlled and dissolved Pb ions were accumulated to form a saturated layer adjacent to the mineral surface.²⁶ The local concentration gradient may favor precipitation of Pb–P minerals on particle surfaces,³⁰ which would constrain further dissolution within inert matrices and resulted in limited PbCO₃(s) transformation. Similar results were observed by Scheckel and Ryan¹⁵ who indicated that Pb dissolution from paint was limited due to precipitation of pyromorphite on the surfaces of Pb-paint particles after addition of phosphoric acid-containing soft-drink. The computational refinement of XRD patterns (Table 1) suggested that PbCO₃(s) decreased from 50.9% to 47.5% as the time increased from 0.17 h to 3 h, correspondingly, the new precipitate, PbHPO₄(s) and Pb₅(PO₄)₃Cl(s) transformed from 5.2% and 40.1% to 1.7% and 48.9%, respectively (Table 1). Eventually, the precipitated Pb in the gastric phase increased from 96.2% to 98.1% (Table 1), accordingly, Pb bioaccessibility of PbCO₃(s) decreased from 3.9% to 1.7% when the time increased from 0.17 h to 3 h (Fig. 1a).

In the intestine phase, P amendment reduced the Pb bioaccessibility to less than 0.04%, much lower than 0.06–0.11% without P amendment (Fig. 1b). However, the Pb bioaccessibility in the intestinal phase slightly increased during the extraction. This was probably due to dissolution of formed Pb carbonates or phosphates through the complexation of the fluid-inherent organic matter with Pb in the neutral intestinal medium. The hypothesis was confirmed by the decrease in solid Pb minerals (Table 1). Total solid Pb in Pb(NO₃)₂ and PbCO₃(s) reduced from 99.7% and 99.8% to 98.7% and 99.0%, respectively, when the time increased from 1 h to 12 h (Table 1).

The results indicated that if Pb-contaminated soil was rich in P or amended with P materials, Pb speciation would change in the gastro-intestinal phase with the formation of insoluble Pb–P minerals, resulting in reduction of the Pb bioaccessibility. On the other hand, Pb bioaccessibility changed with the extraction time as it did not reach equilibrium within the experimental time. That means sampling at the one-time period using PBET method (1 h for gastric phase and 4 h for intestinal phase) may not provide a complete and definite picture of the Pb transformation and bioaccessibility in P-rich soil or the contaminated soil amended with phosphate.

3.3 Effect of the fluid components on Pb bioaccessibility

The primary function of digestive enzymes is to decompose proteins (pepsin) and carbohydrates (pancreatin), while bile salt emulsifies lipids in human's gastrointestinal system.⁴ As the main site of digestion, stomach mainly contains HCl, pepsin, and organic acid such as citric acid, malic acid, lactic acid, and acetic acid. Previous studies have demonstrated that pH in the gastric fluids is a major factor for controlling Pb bioaccessibility.^7,31 Besides fluid pH, digestive fluids also play an important role in controlling Pb bioaccessibility.^4,18

With or without pepsin or organic acid in the gastric fluid, little change was observed in Pb bioaccessibility of Pb(NO₃)₂ and PbCO₃(s) (Fig. 4a), indicating that the effect of the two organic components on Pb bioaccessibility was not limited. It demonstrated again that the pH of the gastric fluids was the major factor responsible for Pb bioaccessibility.³¹ At gastric fluids pH of about 2, most of organic acids were predominant with protonation with less available ligands for free Pb²⁺ (Fig. 5).

Fig. 4 Bioaccessibility of Pb for two Pb compounds with or without phosphate amendments in gastric phase (G-P) (a) and in intestinal phase (I-P) (b) as a function of digestive components process at standard 1 h.

Fig. 5 Percentage of fluid components distribution as a function of pH predicted by Visual MINTEQ.

In the presence of P amendment, the gastric Pb bioaccessibility for both Pb compounds was significantly reduced in the absence of pepsin or organic acid or both, compared to the standard fluid with all presences (Fig. 4a). For example, Pb bioaccessibility for Pb(NO₃)₂ and PbCO₃(s) in the gastric fluid without pepsin was reduced from 3.03% and 1.94% to 1.24% and 0.52%, respectively, compared to the standard process (Fig. 4a). The maximum reduction was obtained at the gastric fluids with only HCl, i.e., without both pepsin and organic acid (Fig. 4a). The presence of carboxyl (–COOH) in pepsin or organic acid may competitively complex with free Pb²⁺, inhibiting formation of insoluble Pb–phosphate precipitates and thus increased Pb bioaccessibility. In addition, pepsin also contains N-functional groups with negative charge, such as aromatic amino acid, which are favorable for Pb²⁺ adsorption via electrostatic interactions.³² Therefore, a lower Pb bioaccessibility was observed in the system without pepsin than that without organic acid (Fig. 4a). According to MINTEQ modeling, about 70% Pb existed as complexes of PbH–glycine²⁺ and PbH₂–(glycine)₂²⁺ in the standard PBET fluids, with only about 21% Pb being present as free Pb²⁺ (Table 2). Without organic acids in the fluid, as much as 70% Pb was still present as complexes of PbH–glycine²⁺ or PbH₂–(glycine)₂²⁺ for both Pb(NO₃)₂ and PbCO₃(s) (Table 2). From these observations we could infer that Pb preferred to complex with pepsin than organic acid. Organic acid had less capacity of binding with Pb²⁺ probably because H⁺ ions competed with Pb²⁺ ions for binding sites.³³ This was further evidenced by the fact that no complex was observed in the absence of pepsin or both organic acid and pepsin, while as high as about 72% Pb or about 75% Pb was present as free Pb²⁺ in the absence of pepsin or both organic acid and pepsin, respectively (Table 2). As a result, precipitation of Pb–P minerals was enhanced, and correspondingly, Pb bioaccessibility was reduced in the absence of pepsin or organic acid or both, compared to the standard fluid (Fig. 4a).

Table 2 Percentage (%) of soluble Pb species among the total Pb as a function of digestive composition in gastric phase (G-P) process at standard 1 h predicted by Visual MINTEQ

	Pb(NO3)2				PbCO3(s)
	Standard	Non-organic acid	Non-pepsin	Non-(organic acid + pepsin)	Standard	Non-organic acid	Non-pepsin	Non-(organic acid + pepsin)
Pb^2+	21.1	21.3	72.4	77.3	20.6	20.7	71.9	75.1
PbCl^+	2.9	2.93	16.6	17.9	2.85	2.87	16.9	17.4
PbCl2 (aq.)	0.04	0.04	0.3	0.33	0.04	0.04	0.31	0.32
PbNO3^+	0.46	0.46	2.71	—	—	—	—	—
Pb(NO3)2 (aq.)		—	0.02	—	—	—	—	—
PbH2–citrate^+	0.21	—	1.07	—	0.21	—	1.09	—
PbH–citrate (aq.)	0.02	—	0.1	—	0.02	—	0.1	—
Pb–malate (aq.)	—	—	0.01	—	—	—	0.01	—
Pb–lactate^+	0.43	—	2.15	—	0.42	—	2.18	—
Pb–acetate^+	0.05	—	0.27	—	0.05	—	0.28	—
PbH2–(glycine)2^2+	4.4	4.44	—	—	4.3	4.34	—	—
PbH–glycine^2+	65.9	66.4	—	—	64.4	64.9	—	—

---

In the intestinal extraction, Pb bioaccessibility for Pb(NO₃)₂ and PbCO₃(s) was reduced in the absence of pancreatin or bile salt or both, compared to the standard process. The same case was true in the presence of P (Fig. 4b). Pancreatin bound Pb could be formed metal–proteins in the neutral intestinal fluid (pH = 7) and then the metal ions were stable.³¹ Bile salt could decrease the surface tension on its surfactant properties, then forming complexes with metals as well.¹⁸ The pancreatin or bile salt mainly contains carboxyl (–COOH). The O-containing function groups may readily be disassociated into COO⁻ under neutral intestinal (pH = 7) condition (Fig. 5) followed by complexing with Pb²⁺, inhibiting formation of sparely soluble Pb–phosphate precipitates. Therefore, the intestinal Pb bioaccessibility for both Pb compounds was higher in the presence of pancreatin and bile salt than that in their absence (Fig. 4b).

4. Conclusions

The Pb bioaccessibility of Pb compounds largely depends on their solubility, which is in turn determined by their chemical species. Our study showed that soluble Pb(NO₃)₂ had higher Pb bioaccessibility than less soluble PbCO₃(s).

In the presence of phosphate amendment, insoluble Pb–phosphate precipitates (i.e., Pb₅(PO₄)₃Cl(s) and PbHPO₄(s)) were formed, resulting in decrease of Pb bioaccessibility of Pb(NO₃)₂ and PbCO₃(s) in the gastro-intestinal tract based on PBET assay. The continuous decrease of Pb bioaccessibility in the gastric phase resulted from the continuous transformation of Pb phosphate during the extraction. In the neutral intestinal tract, the competitive complexation of the fluid-inherent organic matter with Pb was contributed to the slightly increase of Pb bioaccessibility during the extraction. The results indicated that if Pb-contaminated soil was rich in P or amended with P materials, Pb speciation in the gastro-intestinal phase could be changed and insoluble Pb–P minerals would be formed, thus reducing Pb bioaccessibility. On the other hand, Pb bioaccessibility changed with extraction time as it did not reach equilibrium. That means the sampling with one-time period based on PBET method (1 h for gastric tract and 4 h for intestinal tract) may be not accurate to estimate the Pb bioaccessibility in P-rich soil or Pb-contaminated soil amended with P.

In terms of fluid components, the presence of pepsin, organic acid, pancreatin or bile salt in gastric and intestinal phases increased Pb bioaccessibility, probably due to their competitive complexation with Pb²⁺, inhibiting formation of insoluble Pb-phosphate precipitates. It implies that Pb bioaccessibility is affected by the extraction fluid components, varying with different methods. Therefore, such influence may be crucial and valuable in evaluating Pb bioaccessibility at specific contaminated sites. What is more, it can provide an insight into the chemical and physical factors on Pb bioaccessibility in vitro methods, and a screening mechanism on Pb bioavailability in vivo studies for future work.

Acknowledgements

This work was supported in part by National Natural Science Foundation of China (No. 21377081, 21428702) and Shanghai Education Commission (No. 14ZZ026).

References

1. S. Zahran, M. A. Laidlaw, S. P. McElmurry, G. M. Filippelli and M. Taylor, Environ. Sci. Technol., 2013, 47, 2839–2845.
2. S. A. Counter, L. H. Buchanan and F. Ortega, J. Toxicol. Environ. Health, Part A, 2009, 72, 1150–1159.
3. P. J. Surkan, A. Zhang, F. Trachtenberg, D. B. Daniel, S. McKinlay and D. C. Bellinger, Neurotoxicology, 2007, 28, 1170–1177.
4. E. Deshommes, R. Tardif, M. Edwards, S. Sauve and M. Prevost, Chem. Cent. J., 2012, 6, 138.
5. R. R. Rodriguez, N. T. Basta, S. W. Casteel and L. W. Pace, Environ. Sci. Technol., 1999, 33, 642–649.
6. M. E. Kelley, S. E. Brauning, R. A. Schoof and M. V. Ruby, Assessing Oral Bioavailability of Metals in Soil, Battelle Press, Columbus, OH, 2002.
7. M. V. Ruby, A. Davis, R. Schoof, S. Eberle and C. M. Sellstone, Environ. Sci. Technol., 1996, 30, 422–430.
8. S. Denys, J. Caboche, K. Tack, G. Rychen, J. Wragg, M. Cave, C. Jondreville and C. Feidt, Environ. Sci. Technol., 2012, 46, 6252–6260.
9. L. Carrizales, I. Razo, J. I. Téllez-Hernández, R. Torres-Nerio, A. Torres, L. E. Batres, A. C. Cubillas and F. Díaz-Barriga, Environ. Res., 2006, 101, 1–10.
10. X. S. Luo, S. Yu and X. D. Li, Appl. Geochem., 2012, 27, 995–1004.
11. J. Sialelli, G. J. Urquhart, C. M. Davidson and A. S. Hursthouse, Environ. Geochem. Health, 2010, 32, 517–527.
12. M. H. Zia, E. E. Codling, K. G. Scheckel and R. L. Chaney, Environ. Pollut., 2011, 159, 2320–2327.
13. X. Cao, A. Wahbi, L. Ma, B. Li and Y. Yang, J. Hazard. Mater., 2009, 164, 555–564.
14. A. L. Juhasz, D. Gancarz, C. Herde, S. McClure, K. G. Scheckel and E. Smith, Environ. Sci. Technol., 2014, 48, 7002–7009.
15. K. G. Scheckel and J. A. Ryan, Sci. Total Environ., 2003, 302, 253–265.
16. A. G. Oomen, J. Tolls, A. J. A. M. Sips and M. A. G. T. Van den Hoop, Arch. Environ. Contam. Toxicol., 2003, 44, 107–115.
17. E. Smith, K. Scheckel, B. W. Miller, J. Weber and A. L. Juhasz, Sci. Total Environ., 2014, 473, 171–177.
18. A. G. Oomen, C. J. M. Rompelberg, E. V. d. Kamp, D. P. K. H. Pereboom, L. L. D. Zwart and A. J. A. M. Sips, Arch. Environ. Contam. Toxicol., 2004, 46, 183–188.
19. N. Yin, Z. Zhang, X. Cai, H. Du, G. Sun and Y. Cui, Environ. Sci. Technol., 2015, 49, 10675–10681.
20. H. Deng, Z. H. Ye and M. H. Wong, Environ. Pollut., 2006, 141, 69–80.
21. X. Li and C. Huang, Environ. Geol., 2006, 52, 1631–1637.
22. J. P. Gustafsson, Visual MINTEQ, (version 3.0), Department of Land and Water Resources Engineering, The Royal Institute of Technology, Stockholm, Sweden, 2010, http://www.lwr.kth.se/english/OurSoftWare/Vminteq/.
23. M. O. Barnett, V. A. Loganthan, D. W. Kilgour, O. K. Hartzog, J. Zhuang, Y. Wang and P. M. Jardine, Environ. Eng. Sci., 2011, 28, 719–724.
24. S. Olsen and L. Sommers, in Methods of Soil Analysis, part 2, ed. A. L. Page, American Society of Agronomy, Madison, WI, 1982, phosphorus, pp. 407–414.
25. X. Cao, L. Q. Ma, S. P. Singh and Q. Zhou, Environ. Pollut., 2008, 152, 184–192.
26. M. V. Ruby, R. Schoof, W. Brattin, M. Goldade, G. Post, M. Harnois, D. E. Mosby, S. W. Casteel, W. Berti, M. Carpenter, D. Edwards, D. Cragin and W. Chappell, Environ. Sci. Technol., 1999, 33, 3697–3705.
27. B. Palumbo-Roe, J. Wragg, M. R. Cave and D. Wagner, Environ. Sci. Pollut. Res., 2013, 20, 7699–7710.
28. P. Miretzky and A. Fernandez-Cirelli, Environ. Chem. Lett., 2008, 6, 121–133.
29. K. G. Scheckel, G. L. Diamond, M. F. Burgess, J. M. Klotzbach, M. Maddaloni, B. W. Miller, C. R. Partridge and S. M. Serda, J. Toxicol. Environ. Health, Part B, 2013, 16, 337–380.
30. M. V. Ruby, A. Davis, J. H. Kempton, J. W. Drexler and P. D. Bergstrom, Environ. Sci. Technol., 1992, 26, 1242–1248.
31. S. Bruce, B. Noller, V. Matanitobua and J. Ng, J. Toxicol. Environ. Health, Part B, 2007, 70, 1700–1711.
32. Y. Li, W. Demisie and M. K. Zhang, Environ. Sci. Pollut. Res., 2013, 20, 4993–5002.
33. M. V. Ruby, A. Davis, T. E. Link, R. Schoof, R. L. Chaney, G. B. Freeman and P. Bergstrom, Environ. Sci. Technol., 1993, 27, 2870–2877.

---