Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 5.
Published in final edited form as: J Hazard Mater. 2024 Jan 26;467:133633. doi: 10.1016/j.jhazmat.2024.133633

Release Mechanism and Interactions of Cadmium and Arsenic Co-contaminated Ferrihydrite by Simulated In-vitro Digestion Assays

Bing Bai a, Shuqiong Kong a,*, Robert A Root b, Ruiqi Liu a, Xiaguo Wei a, Dawei Cai c, Yiyi Chen a, Jie Chen a, Zhihao Yi a, Jon Chorover b
PMCID: PMC10913812  NIHMSID: NIHMS1966674  PMID: 38335617

Abstract

Cadmium (Cd) and arsenic (As) co-contamination is widespread and threatens human health, therefore it is important to investigate the bioavailability of Cd and As co-exposure. Currently, the interactions of Cd and As by in vitro assays are unknown. In this work, we studied the concurrent Cd-As release behaviors and interactions with in vitro simulated gastric bio-fluid assays. The studies demonstrated this interaction that As bioaccessibility (2.04 to 0.18 ± 0.03%) decreased with the Cd addition compared to the As(V) single system, while Cd bioaccessibility (11.02 to 39.08 ± 1.91%) increased with the As addition compared to the Cd single system. Release of Cd and As is coupled to acidic and reductive dissolution of ferrihydrite. The As(V) is released and reduced to As(III) by pepsin. Pepsin formed soluble complexes with Cd and As. X-ray photoelectron spectroscopy shows that Cd and As formed Fe-As-Cd ternary complexes on ferrihydrite surfaces. The coordination intensity of As-O-Cd is lower than As-O-Fe, resulting in more Cd release from Fe-As-Cd ternary complexes. Our study deepens the understanding of health risks from Cd and As interactions during environmental co-exposure of multiple metal(loid)s.

Keywords: Cd-As co-contaminants, in vitro, simulated gastric bio-fluid, pepsin

Graphical Abstract

graphic file with name nihms-1966674-f0001.jpg

Release behavior and interactions of Cd and As adsorbed on ferrihydrite in an in vitro gastric simulated bio-fluid.

1. Introduction

Cadmium (Cd) and arsenic (As) are toxic trace elements that are widely distributed in natural and contaminated environments (Carlin et al., 2016; Rai et al., 2019; Xiao et al., 2024). Both Cd and As form stable complexes with the surface hydroxyl groups of ferric (hydr)oxides (Yin et al., 2024). Hence iron minerals with sorbed Cd-As are common in multiple metal(loid) contaminated dust and soil (Li et al., 2022; Xie et al., 2023; Tian et al., 2017). Incidental hand-to-mouth ingestion is considered an important exposure pathway for toxic metal(loid) to enter the human body (Zheng et al., 2022). Chronic exposure to a Cd-As co-contaminated environment causes heavy metal(loid) accumulation in living organisms and can eventually lead to cancer (Li et al., 2019). The coexistence of Cd and As has potential interactions that affect the reactivity of each metal(loid) in the environment (Li et al., 2019; Zheng et al., 2020). To unravel these interactions and potential detrimental health outcomes, it is necessary to determine the bioaccessibility and bioavailability of Cd-As co-contaminated iron minerals after ingestion. Recently, in vivo and in vitro assays have been successfully applied to elucidate Cd and As relative bioavailability (RBA) (Denys et al., 2012; Li et al., 2017; Xiao et al., 2024). However, in vivo assays are difficult to perform, costly, poorly reproducible, and ethically restricted (Calatayud et al., 2018). In vitro assays are convenient, rational, and highly repeatable, therefore, they are becoming increasingly used as an alternative to in vivo assays (Juhasz et al., 2014).

In vitro gastrointestinal (IVG) simulations are confirmed to have good correlations with in vivo models for Cd and As bioaccessibility (Juhasz et al., 2014). In vitro bioaccessibility is defined as the proportion of heavy metal(loid)s released in a simulated human bio-fluid environment (Li et al., 2017; Yin et al., 2020). For example, IVG methods use synthetic digestive juices to evaluate the gastrointestinal bioaccessibility of Cd and As and are widely utilized for their reliability and convenience (Juhasz et al., 2014; Qian et al., 2023).

The release process of Cd and As is the first stage from Cd-As sorbed iron minerals enter the human body, determining metal(loid) eventually bioavailability relative to individual Cd or As exposures. Most commonly, Cd exists as a divalent cation (Cd2+) in the environment, whereas in oxic environments, As is generally found as an oxyanion, arsenate, in the +5 oxidation state (HxAsO4x−3) (Cai et al., 2020; Jiang et al., 2023; Yin et al., 2015). Consequently, aqueous and adsorbed ions of Cd and As can interact to control their migration and transformation. For example, previous work showed that As could decrease Cd bioavailability via low solubility Cd-AsO4 formation (Li et al., 2019). Ferrihydrite (Fh) is the most common iron mineral in dust and soils and has a high affinity for Cd and As (Yin et al., 2024). Both elements form inner-sphere complexes with ferrihydrite in neutral aqueous solution (Li et al., 2021; Tiberg and Gustafsson et al., 2016).

When Cd-As-bearing particulate matter is introduced in vitro, release and subsequent reactions of Cd and As are affected by the reactivity of the sorbent minerals (e.g., Fh), the gastric bio-fluid composition, and the solid-liquid interface. Furthermore, the coexistence of Cd and As can enhance adsorption due to the formation of ternary complexes on ferric mineral surfaces (Jiang, et al., 2013; Zeng, et al., 2024). Previous work reported that the presence of common digestive enzymes affected arsenic bioaccessibility (Calatayud et al., 2018; Liu et al., 2023; Ollson et al., 2018). This mechanism was ascribed to gastric enzymes having abundant functional groups, which act as electron donors to reduce As(V) to As(III) (Liu et al., 2023). Moreover, enzymes, and dissolved organic matter (DOM) in general, can form organo-metal(loid) complexes (Cai et al., 2020; Li et al., 2019; Zhang et al., 2022). Enzymes can also form a protective layer over mineral particulate matter and inhibit metal dissolution by a physical barrier effect (Zhang et al., 2022).

Understanding synergistic enhancement or detraction of in vitro bioaccessibility of Cd and As in coexistence systems will inform health risk assessments in muti-metal(loid) contaminated sites. Whereas previous work has evaluated Cd or As bioaccessibility in single contaminant scenarios, the coupled release behavior of Cd and As in systems when they are co-sorbed to Fe(III) (hydr)oxides remains unresolved. Therefore, we designed an in vitro assay for Cd and As bioaccessibility. This study aims to (i) investigate Cd and As speciation and release in gastric bio-fluid using ferrihydrite as model surface sorbent mineral; (ii) reveal the role of pepsin in Cd and As liberation in vitro; and (III) determine the enhancing or suppressing phase partitioning behavior for Cd-As co-sorbed ferrihydrite.

2. Materials and methods

2.1. Preparation of Cd(Ⅱ) and As(V)-sorbed ferrihydrite.

Ferrihydrite was synthesized following the method described in Schwertmann and Cornell (2000). The details were in the Supporting Information (SI). Adsorbed cadmium of single Cd(II) (0.1, 0.5, and 1 mg/g), single As(V) (1, 5, and 10 mg/g), coexisting 10 mg/g As(V) with 0.1, 0.5, and 1 mg/g Cd(II), and 1 mg/g Cd(II) with 1, 5, and 10 mg/g As(V), were labeled Cd alone CdL, CdM, CdH (low, medium, high); As alone AsL, AsM, AsH; and mixtures As-CdL, As-CdM, As-CdH, and Cd-AsL, Cd-AsM, Cd-AsH. See the details of specific sorption under each condition in the Supplementary Material.

2.2. In vitro gastric simulated assay

The IVG simulated gastric bio-fluid included an acid solution of 10 g/L porcine pepsin (Porcine, catalogue No. P-7000, Sigma Chemical), and 8.77 g/L NaCl, with pH adjusted to 1.8 using 2 M HCl. The digestion time in the stomach has been adjusted from 1 hour to 3 hours to account for variations in the digestion rates of different foods. A modified IVG method was used to simulate gastric digestion (Juhasz et al, 2014). In vitro bioassay experiments were performed in an anaerobic chamber (maintain an anaerobic environment) with a shaker at 200 rpm in the dark at 37 °C for 3 hours. Reacted supernatants were isolated by filtration with a 0.45 μm membrane and stored at 4 °C until analyzed. All experiments were performed in triplicate.

2.3. Geochemical modeling

The predominance diagrams simulations were performed using the Eh-pH module within HSC Chemistry. The Eh-pH diagrams showed the stability range of the elements in the redox potential-pH coordinate, which was a good predictor of the morphology and chemistry of the elements. For modeling purposes, a simplified system including the main components of As, Cd, and Fe has been considered. As was chosen as the primary element, Cd and Fe were selected as the other elements. The reaction temperature was set at 25°C. The initial concentrations of As, Cd, and Fe were assumed to be 10−6 mol/L, 10−6 mol/L, and 10−5 mol/L respectively. Various species of As, Cd, and Fe were selected from the database as required. The aim was to describe in general terms the behavior of each species in the coexistence of As, Cd, and Fe, and the reactions that occur by reduction.

2.4. Sample Analysis

2.4.1. Aqueous phase analysis

Simulated gastric bio-fluid samples were filtered through a 0.45 μm membrane. The pH was measured by a calibrated electrode and meter (PHS-3E, LiChen, China). Dissolved Fe(Ⅱ) and total Fe were quantified using the phenanthroline spectrophotometry method with a U3900 spectrophotometer (UV-vis, U3900, Hitachi, Japan) (Hong et al., 2018). Total As, As(III), and Cd were measured by atomic fluorescence spectrophotometry (AFS-933, ThermoFisher Scientific, Beijing, China) (Hong et al., 2018). The As(V) was determined by difference as there were no organic species.

2.4.2. Solids phase analysis

Zeta potential was measured using a Zetasizer Nano-ZS90 (Malvern, UK). Mineral particle morphologies were investigated by scanning electron microscopy (SEM) equipped with an energy dispersive spectrometer (EDS) for elemental analysis (Hitachi SU8010, Japan). Molecular structure and chemical bond changes were analyzed by Fourier Transform infrared spectrometry with attenuated total reflectance for sample introduction (ATR-FTIR, ThermoFisher Scientific, USA). ATR-FTIR was operated in the 650–4000 cm−1 wavenumber range with a resolution of 4 cm−1. Surface elemental analysis was investigated with X-ray photoelectron spectroscopy (XPS) (ThermoFisher Scientific, Escalab 250XI, USA) using a monochromatic Al Kα X-ray source. Data processing for XPS was completed with the software package Avantage 5.9921.

3. Results and Discussion

3.1. In vitro bioaccessibility of Cd and As

3.1.1. Cd release

The amount and bioaccessibility of Cd released from ferrihydrite increased with increasing initial Cd adsorption. In the single Cd exposure, the released Cd was 6.4 and 110.2 μg/L for 0.1 to 1 mg/g exposure and Cd bioaccessibility increased from 6.38% to 11.02% (Table S1 and Fig. 1a). The release curves for Cd showed upward trends followed by steady release followed by a downward trend (Fig. 1b). The Cd release was attributed to acidic dissolution of iron and the formation of soluble Cd-pepsin complexes (Bao et al., 2021). The released Cd re-partitioned to the solid ferrihydrite at extended time points. Furthermore, in Cd-As co-exposure, the Cd release increased from 280.7 to 390.8 μg/L and bioaccessibility increased from 28.07% to 39.08% with the added As(V) of 1 and 10 mg/g, respectively (Table S1 and Fig. 1c). Presence of arsenic significantly accelerated Cd release. Unlike single systems, the released Cd increased rapidly and stabilized after one hour (Fig. 1d).

Fig. 1.

Fig. 1.

Open in a new tab

The released (a) Cd and As bioaccessibility in the single group and (b) Cd(II) concentrations in the single Cd group, (c) Cd and As bioaccessibility in the co-exposure group, (d) Cd concentrations in the co-exposure group, (e) As(III) and As(V) concentrations in the single As group, (f) As(III) and As(V) concentrations in the co-exposure group in simulated gastric bio-fluid. Gastric conditions: 1% pepsin, pH 1.8 ± 0.1, 37 °C, 200 rpm, anaerobic. Values are mean error bars ± SD in triplicate.

3.1.2. As release and speciation

The As(V) released from Fh increased from 5.8 to 182.7 μg/L and As bioaccessibility increased from 0.58% to 2.03% as the As adsorption increased from 1 to 10 mg/g (Fig. 1a and 1e). The released As(III) ranged from 0.00 to 20.19 μg/L after simulated gastric digestion (Fig. 1e). In the co-exposure experiment, the As(V) release decreased from 78.2 μg/L to 14.9 μg/L and As bioaccessibility decreased from 0.73% to 0.18% with Cd co-exposure increasing from 0.1 to 1 mg/g (Fig. 1b and 1f). The As(III) release decreased from 2.88 to 0.26 μg/L with Cd co-exposure (Fig. 1f). This suggested that Cd inhibited As(V) release and reduction to As(III). In all (co)exposures, the released As(V) and As(III) showed increased concentration trends for the first 15 min, after which a downward trend was observed. As(V) release was attributed to ferrihydrite acidic dissolution that was enhanced by pepsin competition at adsorption sites (Li et al., 2019). Pepsin, as an electron donor simultaneously reduced As(V) to As(III) (Zhang et al., 2022). As(V) and As(III) formed soluble As-pepsin complexes with pepsin and were subsequently adsorbed on the ferrihydrite surface (Yan et al., 2022). Reduction of As during in vitro gastric digestion increases toxicity as inorganic As(III) was much more toxic than inorganic As(V) (Zheng et al., 2022).

3.1.3. Fe release and speciation

The total Fe released was 0.70 and 0.90 mg/L in AsL and CdL, and the 0.57 and 0.73 mg/L in AsH and CdH, respectively (Table S1). Results showed that higher Cd and As concentrations resulted in less Fe release, indicating that surface complexation of Cd and As inhibited ferrihydrite dissolution. The released Fe2+(aq) ranged from 0.45–0.61 mg/L in single As experiments (Fig. S1a), ranged from 0.57–0.67 mg/L in single Cd experiments (Fig. S1b), and generally declined with increased Cd and As concentrations in co-exposure experiments (Fig. S1c,d). This indicated that Cd-As coexistence inhibited Fe(III) reduction to Fe2+(aq) under the acidic anaerobic condition.

3.1.4. The role of pepsin in As, Cd, and Fe release

To explore the role of pepsin in As, Cd, and Fe release, assays were carried out by pepsin-free simulated gastric bio-fluid. The released As(III) concentrations were 20.19 and 0.26 μg/L for AsH and As-CdH in the presence of pepsin, whereas As(III) was not detected (LoD = 0.1μg/L) in the absence of pepsin (Fig. S2a,c). The released Fe2+(aq) was 0.45, 0.57, and 0.57 mg/L for AsH, CdH, and As-CdH, respectively in the presence of pepsin, whereas Fe2+(aq) was not detected in the absence of pepsin (Fig. S2d–f). Phenol groups in pepsin provided electrons to reduce As(V) and Fe(III) to As(III) and Fe(II) in simulated gastric bio-fluid (Bao et al., 2021; Zhang et al., 2022). In addition, pepsin formed soluble Cd-pepsin, As-pepsin, and pepsin-As-Cd complexes (Li et al., 2019; Zhang et al., 2022). The aqueous Cd concentrations decreased from 110.2 to 74.8 μg/L in the absence of pepsin (Fig. S2b). This indicated that pepsin accelerated Cd and As release. Moreover, the total Fe released from AsH, CdH, and As-CdH was 19.74, 8.49, and 4.98 mg/L, respectively, corresponding to 34.63, 11.63, and 5.60 times those in the presence of pepsin. Pepsin containing to ferrihydrite surface formed a protective layer preventing ferrihydrite dissolution (Zhang et al., 2022).

3.2. Interactions of As and Cd

3.2.1. Cd inhibited the release and reduction of As(Ⅴ) compared to the single arsenic system

The concentration of As released from Fh decreased with increasing initial Cd concentration (Fig. 2a). In addition, the diminished As release rate constant from 10 mg/g As and 0 mg/g Cd to 10 mg/g As and 0.1mg/g Cd was −0.13 mg·L−1·min−1, which was higher than −0.16 mg·L−1·min−1 from 10 mg/g As and 0.5 mg/g Cd to 10 mg/g As and 1mg/g Cd (Fig. 2c), indicating the As release rate decreased with increasing Cd concentration. Moreover, As(III) concentrations were negatively correlated with Fe2+(aq) concentrations, suggesting As(III) and Fe2+(aq) competition for electrons from pepsin in simulated gastric bio-fluid (Fig. 2e). The zeta potential decreased significantly after the simulated gastric digestion. Zeta potentials followed the trend As-CdH > AsH, whereas particle size followed the trend AsH > As-CdH in reacted bio-fluid (Fig. 2f). The addition of divalent cation Cd created a positively charged interface under acidic conditions. Weakening of electrostatic repulsion leads to agglomeration of molecules and thus a reduction in particle size. Cd inhibited As release due to the formation of anion-bridged Fe-As-Cd ternary complexes on the ferrihydrite surface with As in Cd-As(V) coexistence aqueous solution depending on their charge properties (Du et al., 2021; Ollson et al., 2018). Under experimental conditions, the higher the Cd concentration favored more Fe-As-Cd ternary complexation, and the less As release from ferrihydrite.

Fig. 2.

Fig. 2.

Open in a new tab

(a) Effect of increased Cd on As release, (b) Effect of increased As concentration on Cd release, (c) As release capacity under 10 mg/g As and different adsorption Cd exposure, (d) Cd release capacity under 10 mg/g As and different adsorption Cd exposure, (e) Correlation of Fe(II) with As(III) in vitro, and (f) Zeta potential of ferrihydrite under each experimental condition, superscript 1 before the reaction, 2 after the reaction. Gastric conditions: 1% pepsin, pH 1.8 ± 0.1, 37 °C, 200 rpm, anaerobic. Values reported are the mean, and error bars are ± SD from triplicate analyses.

3.2.2. As(V) accelerated the release of Cd compared to the single cadmium system

The released Cd concentration increased with increasing As(V) concentration (Fig. 2b). In addition, the Cd release rate constant from 1mg/g Cd and 0 mg/g As to 1mg/g Cd and 1 mg/g As was 0.17 mg·L−1·min−1, higher than 0.06 mg·L−1·min−1 from 1mg/g Cd and 5 mg/g As to 1mg/g Cd and 10 mg/g As (Fig. 2d). This indicated that the Cd release rate decreased with increasing As concentration. Zeta potentials followed the trend CdH > As-CdH, whereas particle size followed the trends As-CdH > CdH in simulated gastric bio-fluid (Fig. 2f). The addition of oxyanion arsenate created a negatively charged interface under acidic conditions (Shi et al., 2021). As intermolecular repulsion decreased, zeta potential decreased, resulting in aggregation and particle size increase. The effect of As(V) accelerating Cd release could be interpreted from electrostatic interaction and the structure of the Cd complexation on the Fh surface. The formation of the As-O-Cd bond was mainly based on electrostatic interactions and belongs to an outer-sphere complex that was prone to breakage. It has been shown that Cd tends to form bidentate edge-sharing inner-sphere surface complexes on the ferrihydrite surface (Shi et al., 2021), with Cd-O bond distances of 2.28 Å and Cd-Fe bond distances of 3.23 Å (Du et al., 2018). Whereas samples also containing As(V) had a lengthened Cd-Fe distance at 3.7–3.8 Å and a Cd-As distance at about 3.5 Å (Tiberg and Gustafsson., 2016). In general, shorter bonds have stronger bond energy. Therefore, we interpret the Fe-Cd bond in a single Cd system as more stable than Fe-As-Cd in the coexistence system. Moreover, the addition of Cd reduced the coordination strength of the As-O-Fe bond, probably due to the partial formation of the As-O-Cd ternary complexes (Du et al., 2021; Li et al., 2021). Therefore, breakage of the As-O-Cd bond would be easier than As-O-Fe, resulting in more Cd released into bio-fluid.

3.3. Mineralogical analyses of solid-phase samples

3.3.1. XRD of solid-phase samples

The XRD results showed that the ferrihydrite samples did not have particularly distinct diffraction peaks, with only two broad characteristic peaks at 35°and 62°, which indicated a relatively low crystallinity of the samples (Fig. S3). The results were in agreement with the standard XRD pattern of ferrihydrite reported (Schwertmann, 2000). The As and Cd addition did not affect the crystallinity of ferrihydrite, and the resulting minerals were mainly in the amorphous state, with no crystalline secondary minerals generated.

3.3.2. Cd and As distribution on Fe mineral with SEM-EDS

Ferrihydrite particle morphology transformed from granular to crumb after 3 h reaction indicating ferrihydrite dissolution in simulated gastric bio-fluid (Fig. S4). The SEM-EDS of the Cd-containing ferrihydrite showed Cd generally co-located with Fe and O (Fig. 3a). The distribution of As and O indicated that As and O are closely related (Fig. 3b). The distribution of As was also correlated with Fe and O, whereas Cd was widely dispersed in the Cd-As system (Fig. 3c), possibly related to more Cd being released from ferrihydrite surface into aqueous solution than As after reaction.

Fig. 3.

Fig. 3.

Open in a new tab

SEM micrographs of ferrihydrite particles loaded with Cd and/or As with EDS elemental mapping after in vitro bioassay. (a) Fh-Cd, ferrihydrite-containing Cd, (b) Fh-As, ferrihydrite-containing As, and (c) Fh-As-Cd, ferrihydrite-containing Cd and As. Gastric conditions: 1% pepsin, pH 1.8 ± 0.1, 37 °C, 200 rpm, anaerobic.

3.3.3. Functional groups identification from FT-IR spectroscopy

ATR-FTIR observations confirmed the presence of As-O-Cd bonds and the evolution of pepsin functional groups (Fig. 4a). The peak at 810 cm−1 and 882 cm−1 has been assigned to the stretching vibration of the As-O-Cd and As-O group (Li et al., 2021). The peak of 822 cm−1 to 808 cm−1 on ferrihydrite was attributed to the formation of As(V)-O-Cd(II) bonds (Li et al., 2021). The As-O-Cd peak was significantly weakened after the reaction, indicating Cd and As were released from ferrihydrite into an aqueous solution. Subsequently, new C-O (951 ~ 1150 cm−1) and C=O (1390 and 1560 cm−1) groups formed after the reaction, demonstrating that pepsin had bound to the ferrihydrite surface (Ehlert et al., 2018; Xiao et al., 2020; Yin et al., 2021). The peaks of Fe-OH (~1460 cm−1) and Fe-O (~1350 cm−1) were significantly weaker after simulated gastric digestion, indicating Fe released from ferrihydrite. The peak intensity of OH group ~1630 cm−1 and ~2540 cm−1 of ferrihydrite significantly decreased after simulated gastric digestion, indicating that the phenolic group from pepsin reduced As(V) and Fe(III) to As(III) and Fe2+(aq), respectively (Bao et al., 2021; Bradham, et al., 2011; Li et al., 2014).

Fig. 4.

Fig. 4.

Open in a new tab

Spectroscopy of unreacted and in vitro reacted As/Cd ferrihydrite. (a) ATR-FTIR of control ferrihydrite and experiment Fh with As/Cd at 10 mg/g As and 1 mg/g Cd; A-Fh is before reaction, and G-Fh is after reaction. XPS spectra (b-f) of ferrihydrite with As/Cd before and after simulated gastric digestion. The XPS spectra are (b) As 3d, (c) Cd 3d, (d) Fe 2p, (e) O 1s, and (f) C 1s after and before simulated gastric digestion. Gastric conditions: 1% pepsin, pH 1.8 ± 0.1, 37 °C, 200 rpm, anaerobic.

3.3.4. Surface complexation and elements speciation by X-ray photoemission spectroscopy

The As 3d bonding energy at 44.6 and 45.4 eV corresponded to the 3d3/2 and 3d5/2 orbit, respectively (Chen et al., 2022). The intensity of the peak indicated As content decreased on the ferrihydrite. The As 3d spectra showed that As(V) at 45.4 eV decreased from 100% to 46.67% and As(III) at 44.6 eV increased from 0% to 53.33%, indicating As(V) was reduced to As(III) after simulated gastric digestion (Fig. 4b) (Chen et al., 2022). The Cd spectra of 3d3/2 at 412.4 eV and 3d5/2 at 405.1 eV indicated the existence of Cd-O bonding (Fig. 4c) (Gao et al., 2019). After the reaction, Cd was not detected, indicating Cd was released and formed a Cd-pepsin complex with pepsin. The FeIII 2p bonding energy at 725 and 711 eV corresponded to the 2p1/2 and 2p3/2 orbit, respectively. The FeII 2p bonding energy at 724 and 710 eV corresponded to the 2p1/2 and 2p3/2 orbit, respectively (Chen et al., 2022). The Fe spectra showed the percentage of Fe(II) increased from 0% to 34.3% after the reaction, indicating ferrihydrite surface reduction (Fig. 4d). The peaks at 530.2, 532.4, and 531.5 eV corresponded to metal oxide (e.g., Fe-, As-, or Cd-O), C-O, and C=O, respectively) (Fig. 4e) (Li et al., 2021; Yu et al., 2016). The O1s spectra showed a shift in the O bonding character where C-O and C=O bonds formed after the reaction and the O-H bond increased from 40.43% to 50.22%; attributed to the formation of Fe-pepsin complexes and the oxidation of phenolic groups after reaction (Liu et al., 2023). The intensity of the metal oxide peak at 530.2 eV decreased by almost 50%, indicating metals (Fe, As, and Cd) released into the simulated gastric bio-fluid (Li et al., 2021; Yu et al., 2016). After simulated gastric digestion, the C=O bond at 288.4 eV increased from 12.84% to 23.63% and C-C bonding at 284.8 eV decreased from 74.05% to 62.96% (Fig. 4f) (Gao et al., 2019). The Fe, O, and C spectra indicated phenolic and hydroxyl groups formed the Fe-pepsin complex.

3.4. Speciation of Cd and As under a wide range of environmental conditions

The Eh-pH predominance diagrams in Fig.5 showed the dissolved phases and species changes of Cd, Fe, and As; thus, these diagrams could qualitatively identify the reactions influencing Cd and As release and reduction under different pH and Eh conditions. Dissolution of ferrihydrite releases adsorbed Cd as Cd2+ under acidic pH conditions. The Cd(II) was present as Cd(OH)2 and Cd(OH)42− at higher pH. Dissolution of ferrihydrite releases adsorbed As as H3AsO4, H3AsO3, and H2AsO4− under acidic pH conditions. Iron formed mainly Fe3+ in acidic aerobic environments and Fe2+ in anaerobic environments. At higher pH values, the main forms of iron that were formed are Fe(OH)2 and Fe(OH)3. Solution pH and Eh showed decreasing trends in both single and coexisting systems (Fig. S5). The production of H+ led to a decrease in pH during the reaction. The vital activities of the microorganisms led to an Eh decrease in Eh in the simulated gastric bio-fluid (Wang et al., 2021).

Fig. 5.

Fig. 5.

Open in a new tab

Eh-pH predominance diagrams of migration transformation of Cd, Fe, and As species in As-Fe-Cd-H2O System. The molality of As, Cd, and Fe is assumed to be 10−6 mol/L, 10−6 mol/L, and 10−5 mol/L. The pentagons represent the mean pH and Eh coordinates corresponding to the reaction.

3.5. Mechanism of Cd and As(V) interactions

In vitro assays were carried out to elucidate the mechanism of the interaction of Cd and As (Fig. 6). Acid dissolution of ferrihydrite led to As(V), Cd(II), and Fe(III) released into the simulated gastric bio-fluid. Organic ligands of pepsin competed with As for ferrihydrite surface adsorption sites or formed Cd-pepsin and As-pepsin complex resulting in Cd or As released. In addition, As(V) and Fe(III) were reduced to As(III) and Fe(II) by phenol groups from the pepsin (Eqs. 2, Eqs. 3 and Eqs. 4) (Liu et al., 2023). Kinetic time series experiments showed partial soluble Cd-pepsin and As-pepsin complexes adsorbed on the ferrihydrite surface afterward.

3.5. (1)
3.5. (2)
3.5. (3)
3.5. (4)
3.5. (5)

Fig. 6.

Fig. 6.

Open in a new tab

Schematic illustration of the possible release mechanisms of As and/or Cd from ferrihydrite in simulated gastric bio-fluid under (a) As(V) only, (b) Cd only, and (c) coexisting Cd-As(V).

In Cd-As(V) coexisting system, Cd significantly decreased As(V) bioaccessibility compared to the Cd single system, whereas As(V) increased Cd bioaccessibility compared to the As(V) single system (Fig. 6). Cd and As formed Fe-As-Cd ternary complexes on ferrihydrite surfaces, inhibiting pepsin competition with As(V) for surface adsorption, simultaneously decreasing As(V) release and reduction (Eqs. 1 and Eqs. 5) (Liu et al., 2023). The formation of the As-O-Cd bond was mainly based on electrostatic interactions and belongs to an outer-sphere complex that was prone to breakage. In addition, previous Extended X-ray absorption fine structure investigations found the Cd-Fe second shell distance was about 3.3 Å in bidentate edge-sharing of oxygen atoms (Tiberg and Gustafsson., 2016). After arsenic addition, the Cd-Fe distance lengthened to about 3.7–3.8 Å, and a Cd-As distance of about 3.5 Å was observed (Li et al., 2021). The coordination intensity of As-O-Cd was lower than that of As-O-Fe (Li et al., 2021). The longer Cd-Fe distance and decreased second shell intensity indicate a weaker ligand, possibly attributed to mono-dentate Cd-Fe coordination. The lower energy of the As-O-Cd bond leads to breakage at the O ligand, leading to Cd release from Fe-As-Cd ternary complexes.

3.6. Discussion

Due to the common cadmium and arsenic co-contamination and associated health risks, Cd and As have been concerned over the past decade. However, most studies have focused on single metal contamination or bioaccessibility, whereas the underlying Cd and As interaction mechanisms related to their toxicity are not well understood. The in vitro approach could help predict pollutant toxicity and better understand the release process of Cd and As in the simulated gastric bio-fluid. The present study used the in vitro method to assess health risks associated with Cd and As co-contaminated iron minerals exposure uptake, furthermore elucidated the quantitative relationship and interaction mechanisms of Cd and As.

Acid dissolution of ferrihydrite and reduction dissolution of pepsin were the pathways of cadmium and arsenic released to simulated gastric bio-fluid. Iron and arsenic were synergistically reduced by pepsin in simulated gastric bio-fluid. The reduction of iron and arsenic led to Fe(II) and As(III) production, increasing the toxicity of arsenic in the simulated gastric bio-fluid. The formation of Fe-As-Cd ternary complexes decreased As(V) release and reduction in the Cd-As coexistence system compared to the single As system. In the simulated gastric bio-fluid, the breakage of As-O-Cd bonds increased Cd release in the Cd-As coexistence system compared to the single Cd system.

In short, Fe-As-Cd ternary complexes, pepsin-metal complexes, pepsin reduction, and ferrihydrite acid dissolution findings were used to better understand the interaction and released mechanisms of Cd and As during simulated gastric bio-fluid. The study indicated ternary complexes and pepsin may serve as involved in the interaction and released mechanisms of Cd and As, but further study was needed. The current study furthers our understanding of the interaction and released mechanisms of Cd and As that could be used for assessing toxicity following Cd and As co-contaminated exposure to humans.

4. Conclusions

Natural sources and anthropogenic activities can cause Cd-As coexistence and co-contamination of ecosystems, which threaten environmental and human health. When contaminated particles are ingested, or inhaled and cleared to the GI tract, the Cd-As coexistence bioaccessibility differed from single contaminant particles is determined through metal-metalloid interactions during the digestive release, altering co-exposure toxicity and health risk. We investigate metal(loid) release and speciation with synthetic gastric systems to elucidate in vitro interactions between Cd and As. The study demonstrates an increased bioaccessibility of cadmium in the coexisting Cd-As system compared to the single Cd system. The bioaccessibility of arsenates is reduced in the coexisting Cd-As system compared to the single As system. Furthermore, pepsin promotes metal(loid) release in simulated bio-fluid. In the presence of ferrihydrite, oxyanion As(V) can bind to divalent cation Cd to form Fe-As-Cd ternary complexes, which inhibit As(V) release and reduction, simultaneously promoting Cd release. However, the released As is accompanied by a reduction of As(V) to As(III) by phenol-OH, which would increase the arsenic toxicity. In this study, in vitro assays verified the mechanism of the interaction and release of Cd and As. This deepens our understanding of multi-metal(loid) interactions and pepsin involvement in Cd and As release. These results contribute to a more effective health risk analysis and help contribute to potential regulation strategies for multi-metal(loid) co-contaminated dust and soils. To make the results more convincing and applicable, future work could be concerned with 1) the health risks of reaction products occurring in the cadmium and arsenic co-contaminated condition; 2) the simulator of the human intestinal microbial ecosystem (SHIME) to investigate cadmium and arsenic cocontaminated bioavailability; and 3) in vivo bioavailability of cadmium and arsenic co-contamination using an animal model.

Supplementary Material

1

Highlights.

  • Enhanced bioaccessibility of cadmium in the co-existing Cd-As system can be attributed to a decrease in ligand strength compared to the Cd single system.

  • Reduced bioaccessibility of arsenate in the co-existing Cd-As system can be attributed to the formation of ternary surface complexes compared to the As single system.

  • Pepsin reduces As and Fe and forms complexes with Cd and As in solution and on minerals.

Environmental Implication.

Cadmium and arsenic, both carcinogenic trace elements, can co-occur widely in soils and ecosystems, highlighting the importance of understanding their interactions to prevent health problems and mitigate environmental impacts. The interaction of cadmium and arsenic in the human body affects their respective bioavailability. This study investigated the interaction mechanism of ferrihydrite-adsorbed cadmium and arsenic in gastric simulated bio-fluid. Our findings have important implications for the development of potential regulatory strategies for multi-metal(loid) co-contaminated dust and soil.

Acknowledgments

The work was financially supported by the National Natural Science Foundation of China (No. 42272309), and the National Institute of Environmental Health Sciences (NIEHS) Superfund Research Program Grant P42 ES04940.

Footnotes

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Bao YP, Bolan NS, Lai JH, Wang YS, Jin XH, Kirkham MB, et al. , 2021. Interactions between organic matter and Fe (hydr)oxides and their influences on immobilization and remobilization of metal(loid)s: A review. Crit. Rev. Environ. Sci. Technol. 52, 4016–4037. 10.1080/10643389.2021.1974766. [DOI] [Google Scholar]
  2. Bradham KD, Scheckel KG, Nelson CM, Seales PE, Lee GE, Hughes MF, et al. , 2011. Relative Bioavailability and Bioaccessibility and Speciation of Arsenic in Contaminated Soils. Environ. Health Perspect. 119(11), 1629–1634. 10.1289/ehp.1003352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cai XL, Wang PF, Li ZJ, Li Y, Yin NY, Du HL, et al. , 2020. Mobilization and transformation of arsenic from ternary complex OM-Fe(III)-As(V) in the presence of As(V)-reducing bacteria. J. Hazard. Mater. 381, 120975. 10.1016/j.jhazmat.2019.120975. [DOI] [PubMed] [Google Scholar]
  4. Calatayud M, Xiong C, Laing DG, Raber G, Francesconi K, Wiele TVD, 2018. Salivary and Gut Microbiomes Play a Significant Role in in Vitro Oral Bioaccessibility, Biotransformation, and Intestinal Absorption of Arsenic from Food. Environ. Sci. Technol. 52, 24, 14422–14435. 10.1021/acs.est.8b04457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Carlin DJ, Naujokas MF, Bradham KD, Cowden J, Heacock M, Henry HF, et al. , 2016. Arsenic and environmental health: state of the science and future research opportunities. Environ. Health Perspect. 124, 890–899. 10.1289/ehp.1510209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen HR, Zhang DR, Li Q, Nie ZY, Pakostova E, 2022. Release and fate of As mobilized via bio-oxidation of arsenopyrite in acid mine drainage: Importance of As/Fe/S speciation and As(III) immobilization. Water Res. 223, 118957. 10.1016/j.watres.2022.118957. [DOI] [PubMed] [Google Scholar]
  7. Denys S, Caboche J, Tack K, Rychen G, Wragg J, Cave M, et al. , 2012. In vivo validation of the unified BARGE method to assess the bioaccessibility of arsenic, antimony, cadmium, and lead in soils. Environ. Sci. Technol. 46, 6252–6260. 10.1021/es3006942. [DOI] [PubMed] [Google Scholar]
  8. Du HH, Nie N, Rao W, Lu L, Lei M, Tie BQ, 2021. Ferrihydrite-organo composites are a suitable analog for predicting Cd(II)-As(V) coexistence behaviors at the soil solid-liquid interfaces. Environ. Pollut. 290, 118–040. 10.1016/j.envpol.2021.118040. [DOI] [PubMed] [Google Scholar]
  9. Du HH, Peacock CL, Chen WL, Huang QY, 2018. Binding of Cd by ferrihydrite organo-mineral composites: Implications for Cd mobility and fate in natural and contaminated environments. Chemosphere. 207, 404–412. 10.1016/j.chemosphere.2018.05.092. [DOI] [PubMed] [Google Scholar]
  10. Ehlert K, Mikutta C, Jin Y, Kretzschmar R, 2018. Mineralogical Controls on the Bioaccessibility of Arsenic in Fe(III)-As(V) Coprecipitates. Environ. Sci. Technol. 52 (2), 616–627. 10.1021/acs.est.7b05176. [DOI] [PubMed] [Google Scholar]
  11. Gao JF, Cupolillo A, Nappini S, Bondino F, Edla R, Fabio V, et al. , 2019. Surface Reconstruction, Oxidation Mechanism, and Stability of Cd3As2. Adv. Funct. Mater. 29, 26, 1900965. 10.1002/adfm.201900965. [DOI] [Google Scholar]
  12. Hong J, Liu LH, Luo Y, Tan WF, Qiu GH, Fan, Liu., 2018. Photochemical oxidation and dissolution of arsenopyrite in acidic solutions. Geochim. Cosmochim. Acta. 239, 173–185. 10.1016/j.gca.2018.07.034. [DOI] [Google Scholar]
  13. Jiang W, Lü J, Luo L, Yang K, Lin YF, Hu FB, et al. , 2013. Arsenate and cadmium co-adsorption and co-precipitation on goethite. J. Hazard. Mater. 262, 55–63. 10.1016/j.jhazmat.2013.08.030. [DOI] [PubMed] [Google Scholar]
  14. Jiang Z, Nie K, Arinzechi C, Li JX, Liao Q, Si MY, et al. , 2023. Cooperative effect of slow-release ferrous and phosphate for simultaneous stabilization of As, Cd and Pb in soil. J. Hazard. Mater. 452, 131232. 10.1016/j.jhazmat.2023.131232. [DOI] [PubMed] [Google Scholar]
  15. Juhasz AL, Smith E, Clay, Nelson C, Thomas DJ, Bradham K, 2014. Variability associated with As in vivo-in vitro correlations when using different bioaccessibility methodologies. Environ. Sci. Technol. 48(19), 11646–11653. 10.1021/es502751z. [DOI] [PubMed] [Google Scholar]
  16. Li BY, Wei DN, Zhou YM, Huang YY, Tie BQ, Lei M, 2021. Mechanisms of arsenate and cadmium co-immobilized on ferrihydrite inferred from ternary surface configuration. Chem. Eng. J. 424, 1385–8947. 10.1016/j.cej.2021.130410. [DOI] [Google Scholar]
  17. Li HB, Chen XQ, Wang JY, Li MY, Zhao D, Luo XS, et al. , 2019. Antagonistic Interactions between Arsenic, Lead, and Cadmium in the Mouse Gastrointestinal Tract and Their Influences on Metal Relative Bioavailability in Contaminated Soils. Environ. Sci. Technol. 53, 14264–14272. 10.1021/acs.est.9b03656. [DOI] [PubMed] [Google Scholar]
  18. Li HB, Cui XY, Li K, Li J, Juhasz AL, Ma LQ, 2014. Assessment of in vitro lead bioaccessibility in house dust and its relationship to in vivo lead relative bioavailability. Environ Sci Technol. 48(15), 8548–8555. 10.1021/es501899j. [DOI] [PubMed] [Google Scholar]
  19. Li HB, Li MY, Zhao D, Li J, Li SW, Juhasz AL, et al. , 2019. Oral bioavailability of As, Pb, and Cd in contaminated soils, dust, and foods based on animal bioassays: a review. Environ. Sci. Technol. 53(18), 10545–10559. 10.1021/acs.est.9b03567. [DOI] [PubMed] [Google Scholar]
  20. Li HB, Li MY, Zhao D, Li J, Li SW, Xiang P, et al. , 2019. Arsenic, lead, and cadmium bioaccessibility in contaminated soils: Measurements and validations. Crit. Rev. Environ. Sci. Technol. 50, 1303–1338. 10.1080/10643389.2019.1656512. [DOI] [Google Scholar]
  21. Li HB, Xue RS, Chen XQ, Lin XY, Shi XX, Du HY, et al. , 2022. Ca Minerals and Oral Bioavailability of Pb, Cd, and As from Indoor Dust in Mice: Mechanisms and Health Implications. Environmental Health Perspectives. 130 (12). 10.1289/EHP11730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li J, Chen S, Li HB, Liu X, Cheng JM, Ma LQ, 2021. Arsenic bioaccessibility in rice grains via modified physiologically-based extraction test (MPBET): Correlation with mineral elements and comparison with As relative bioavailability. Environ. Res. 198, 111198. 10.1016/j.envres.2021.111198. [DOI] [PubMed] [Google Scholar]
  23. Li SW, Liu X, Sun HJ, Li MY, Zhao D, Luo J, et al. 2017. Effect of phosphate amendment on relative bioavailability and bioaccessibility of lead and arsenic in contaminated soils. J Hazard Mater. 339, 256–263. 10.1016/j.jhazmat.2017.06.040. [DOI] [PubMed] [Google Scholar]
  24. Liu RQ, Kong SQ, Shao YS, Cai DW, Bai B, Wei XG, et al. , 2023. Mechanisms and health implications of toxicity increment from arsenate-containing iron minerals through in vitro gastrointestinal digestion. Geoderma. 116377. 10.1016/j.geoderma.2023.116377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ollson CJ, Smith E, Juhasz AL, 2018. Can in vitro assays account for interactions between inorganic co-contaminants observed during in vivo relative bioavailability assessment? Environ. Pollut. 233, 348–355. 10.1016/j.envpol.2017.10.089. [DOI] [PubMed] [Google Scholar]
  26. Qian Q, Liang J, Ren ZF, Sim JK, Xu XY, Rinklebe J, et al. , 2023. Digestive fluid components affect speciation and bioaccessibility and the subsequent exposure risk of soil chromium from stomach to intestinal phase in in-vitro gastrointestinal digestion. J. Hazard. Mater. 463, 132882. 10.1016/j.jhazmat.2023.132882. [DOI] [PubMed] [Google Scholar]
  27. Rai PK, Lee SS, Zhang M, Tsang YF, Kim KH, 2019. Heavy metals in food crops: Health risks, fate, mechanism, and management. Environ. Int. 125, 365–385. 10.1016/j.envint.2019.01.067. [DOI] [PubMed] [Google Scholar]
  28. Schwertmann U, Cornell RM, 2000. Ferrihydrite. Iron Oxides in the Laboratory. Wiley-VCH Verlag GmbH, Weinheim, Germany. 103–112. [Google Scholar]
  29. Shi MQ, Min XB, Ke Y, Lin Z, Yang ZH, Wang S, et al. , 2021. Recent progress in understanding the mechanism of heavy metals retention by iron (oxyhydr)oxides. Sci. Total Environ. 752, 141930. 10.1016/j.scitotenv.2020.141930. [DOI] [PubMed] [Google Scholar]
  30. Tian L, Shi ZQ, Lu Y, Dohnalkova AC, Lin Z, Dang Z, 2017. Kinetics of Cation and Oxyanion Adsorption and Desorption on Ferrihydrite: Roles of Ferrihydrite Binding Sites and a Unified Model. Environ. Sci. Technol. 51(18), 10605–10614. 10.1021/acs.est.7b03249. [DOI] [PubMed] [Google Scholar]
  31. Tiberg C, Gustafsson JP, 2016. Phosphate effects on cadmium(II) sorption to ferrihydrite. J. Colloid Interface Sci. 471, 103–111.14. 10.1016/j.jcis.2016.03.016. [DOI] [PubMed] [Google Scholar]
  32. Wang HT, Liang ZZ, Ding J, Xue XM, Li G, Fu SL, et al. , 2021. Arsenic bioaccumulation in the soil fauna alters its gut microbiome and microbial arsenic biotransformation capacity. J. Hazard. Mater. 417, 126018. 10.1016/j.jhazmat.2021.126018. [DOI] [PubMed] [Google Scholar]
  33. Xiao Q, Liang JD, Luo HJ, Li HM, Yang J, Huang S, 2020. Investigations of conformational structures and activities of trypsin and pepsin affected by food colourant allura red. J. Mol. Liq. 319, 114359. 10.1016/j.molliq.2020.114359. [DOI] [Google Scholar]
  34. Xie KT, Xie NG, Liao ZY, Luo XS, Peng WJ, Yuan Y, 2023. Bioaccessibility of arsenic, lead, and cadmium in contaminated mining/smelting soils: Assessment, modeling, and application for soil environment criteria derivation. J. Hazard. Mater. 443, 130321. 10.1016/j.jhazmat.2022.130321. [DOI] [PubMed] [Google Scholar]
  35. Yan YP, Wan B, Mansor M, Wang XM, Zhang Q, Kappler A, et al. , 2022. Co-sorption of metal ions and inorganic anions/organic ligands on environmental minerals: A review. Sci. Total Environ. 803, 149918. 10.1016/j.scitotenv.2021.149918. [DOI] [PubMed] [Google Scholar]
  36. Yin NY, Li YP, Cai XL, Du HL, Wang PF, Han ZL, et al. , 2021. The role of soil arsenic fractionation in the bioaccessibility, transformation, and fate of arsenic in the presence of human gut microbiota. J. Hazard. Mater. 401, 123366. 10.1016/j.jhazmat.2020.123366. [DOI] [PubMed] [Google Scholar]
  37. Yin NY, Zhang ZN, Cai XL, Du HL, Sun GX, Cui YS, 2015. In Vitro Method To Assess Soil Arsenic Metabolism by Human Gut Microbiota: Arsenic Speciation and Distribution. Environ. Sci. Technol. 49, 17, 10675–10681. 10.1021/acs.est.5b03046. [DOI] [PubMed] [Google Scholar]
  38. Yin YP, Wang YR, Ding CF, Zhou ZG, Tang X, He LQ, et al. , 2024. Impact of iron and sulfur cycling on the bioavailability of cadmium and arsenic in co-contaminated paddy soil. J. Hazard. Mater. 465, 133408. [DOI] [PubMed] [Google Scholar]
  39. Yu HY, Wu B, Zhang XX, Liu S, Yu J, Cheng SP, et al. , 2016. Arsenic metabolism and toxicity influenced by ferric iron in simulated gastrointestinal tract and the roles of gut microbiota. Environ. Sci. Tech, 50 (13), 7189–7197. 7197. 10.1021/acs.est.6b01533. [DOI] [PubMed] [Google Scholar]
  40. Zhang Y, Cao J, Wang X, Liu H, Shao Y, Chu CL, et al. , 2022. The effect of enzymes on the in vitro degradation behavior of Mg alloy wires in simulated gastric fluid and intestinal fluid. Bioact. Mater. 7, 217–226. 10.1016/j.bioactmat.2021.05.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zheng SN, Wang Q, Yu HY, Huang XZ, Li FB, 2020. Interactive effects of multiple heavy metal(loid)s on their bioavailability in cocontaminated paddy soils in a large region. Sci. Total Environ. 708, 135126. 10.1016/j.scitotenv.2019.135126. [DOI] [PubMed] [Google Scholar]
  42. Zheng XM, Zhang ZY, Chen JC, Liang HT, Chen X, Qin Y, et al. , 2022. Comparative evaluation of in vivo relative bioavailability and in vitro bioaccessibility of arsenic in leafy vegetables and its implication in human exposure assessment. J. Hazard. Mater. 423, 126909. 10.1016/j.jhazmat.2021.126909. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1966674-supplement-1.docx (3.2MB, docx)

RESOURCES