Arsenic transfer along the soil-sclerotium-stroma chain in Chinese cordyceps and the related health risk assessment

Sample collection and preparation

We selected three sites from the endemic areas in Qinghai-Tibet Plateau for this study. Site A was located at 29°36′N, 94°36′E; Site B was located at 29°35′N, 94°36′E; and Site C, was located at 35°14′N, 91°48′E. We took fifteen soil samples from the 10–20 cm topsoil and twenty Chinese cordyceps samples about 0.3 g each from each sampling site in mid-July 2017. The samples were kept in an icebox and were transported to the laboratory.

In the laboratory, these samples were freeze dried. The Chinese cordyceps samples were divided into two subsamples: SC and ST. ST was light and thin compared with SC, and each of the ten ST subsamples were combined to form a batch sample. Each of the five SC subsamples were combined to form a batch sample. The twenty Chinese cordyceps samples collected from each sampling spot were divided into four batches of sclerotium samples and two batches of stroma samples, which were named SC_A∕B∕C and ST_A∕B∕C according to sampling sites. Soil samples were ground into powders with a grain size of less than 150 mesh. Every five powdered soil samples were combined into one batch and named A∕B∕C according to sampling sites.

Sample digestion

0.1 g of each pre-dried sample was digested with concentrated nitric acid (16 mol/L) using the high-temperature and microwave-assisted methods to determine the total As concentration in SC and ST. The digestion methods followed that of our earlier study (Guo et al., 2018b). To determine the As speciation, 1 g of the sample powder was digested with 20 mL 0.15 mol/L HNO₃ at 90 °C using a water bath for 12 h (Guo et al., 2018b). The sample was cooled to room temperature and all of the digested product was centrifuged for 15 min at the speed of 7104 g. The collected supernatant was filtered through a sieve with a mesh aperture of 0.22 µm and kept in cold storage until analysis.

Approximately 0.1 g of each powdered sample was blended with a mixed solution of hydrochloric acid at 12mol/L and concentrated nitric acid at 16mol/L with the volume ratio of 3:1 to determine the total As concentration in the soil. Digestion was performed according to the standard method, HJ 803-2016 (MEPP, 2016). Different As species were extracted according to the method used by Thomas, Finnie & Williams (1997). Briefly, 10 mL 1 mol/L phosphoric acid (H₃PO₄) was added into 0.2 g pre-dried sample, processed, and cooled in a microwave. The extract was filtered and diluted with distilled water and prepared for analysis.

Arsenic determination of sample

The total As was measured by ICP-MS (Agilent 7800, Santa Clara, CA, USA). The separation of As species (iAs^III, iAs^V, MMA, DMA and AsB) were conducted by HPLC (Agilent 1260, Santa Clara, CA, USA) and the separated As species were determined by ICP-MS. Based on our previous study (Guo et al., 2018b), the iAs^III could not be separated from the other arsenic species. To determine the level of iAs^III, 1 mL H₂O₂ was added into the extraction to fully oxidize the iAs^III to iAs^V and the arsenic species were analyzed before (Figs. 2B to 2C, 2F) and after (Figs. 2D to 2E, 2G) H₂O₂ was added. The iAs^III was calculated by subtracting the level of iAs before addition to H₂O₂ from the level of iAs^V after addition to H₂O₂.

Each test was performed in triplicate. The concentrations of total As and As species were quantified using calibration curves, which were made with standard samples (National Institute of Metrology, Beijing, China).

The precision of our results was tested by a blank reagent and the Chinese national standard for the green Chinese onion: GBW10049 (GSB-27) and the yellow croaker: GBW08573. Linear responses ranged between 0.5 and 500 µg/L for the total As determination and between 0.2 and 300 µg/L for the As species determination; the correlation coefficients were greater than 0.9997 (Table S1). The relative standard deviation (RSD) was less than 10% (Table S2) and the recovery of these certified reference materials was within the acceptable range (Table S3).

Health risk assessment of As in soil

Arsenic has been categorized as a chemical carcinogen by USEPA (2011), as well as non-carcinogens for human. Arsenic can migrate into plants and enter the human body through oral ingestion as part of the food chain. The inhalation of soil and dust, and dermal contact are also exposure pathways for humans. Therefore, to comprehensively assess arsenic exposure all three exposure pathways are taken into consideration.

According to the model of human health evaluation by the United States Environment Protection Agency (USEPA) (USEPA, 1989), the average daily doses (ADD, mg/(kg d)) through the three exposure pathways (ingestion: ADD_ing; inhalation: ADD_inh; dermal contact: ADD_dermal) were calculated separately as follows: (1)${\displaystyle {ADD}_{ing}=C\times \frac{IngR\times EF\times ED}{BW\times AT}\times 10^{-6}}$ (2)${\displaystyle {ADD}_{inh}=C\times \frac{InhR\times EF\times ED}{PEF\times BW\times AT}}$ (3)${\displaystyle {ADD}_{dermal}=C\times \frac{SA\times AF\times ABF\times EF\times ED}{PEF\times BW\times AT}\times 10^{-6}}$

where C (mg/kg) is the concentration of total As in soil, EF (days/year) is the exposure frequency, ED (years) is the exposure duration, BW (kg) is the body weight, AT (days) is the average time, PEF (m³/kg) is the particular emission factor, SA (cm²) is the surface area of exposed skin, AF (mg/cm²) is the skin adherence factor, ABF is the absorption factor, IngR (mg/d), and InhR (m³/days) is the ingestion rate and inhalation rate, respectively. The parameters for children and adults are shown in Table S4 and refer to the Chinese assessment guidelines for an environmental site (MEPP, 2014) and US exposure factors handbook (USEPA, 2011).

The hazard quotient (HQ) was calculated separately as follows: (4)${\displaystyle HQ=\frac{ADD}{Rf\mathrm{D}}}$ (5)${\displaystyle HI=\sum {HQ}_i}$

where RfD is the non-carcinogenic reference dose for As (mg/(kg d)); the values through ingestion, inhalation, and dermal contact are: 3.0 ×10⁻⁴, 1.5 ×10⁻⁵, and 3.0 ×10⁻⁴, respectively (MEPP, 2014; USEPA, 2013; Liu et al., 2008). HI is the total exposure hazard index. If HQ or HI <1, there is no concern for non-carcinogenic effects, whereas potential non-carcinogenic risks may occur in cases where HQ or HI >1.

Carcinogenic risk (CR) was calculated as follows: (6)${\displaystyle CR={ADD}_{ing∕inh∕dermal}\times SF}$ (7)${\displaystyle {CR}_T=\sum CR}$

where SF is the slope factor of As and the values through ingestion, inhalation, and dermal contact is: 1.5, 4.3 ×10⁻³, and 1.5, respectively (USEPA, 2011; MEPP, 2014; USEPA, 2013; Liu et al., 2008). CR_T is the sum of CR for the three pathways. The probability of cancer risk for humans over a lifetime is characterized by CR with an acceptable range from 1.0 ×10⁻⁶ to 1.0 ×10⁻⁴. If CR <1.0 ×10⁻⁶, which suggests no significant effect on human beings; CR >1.0 ×10⁻⁴ is likely to be harmful to humans.

Statistical analysis

Data were analysed using Microsoft Excel 2013 (Microsoft, Redmond, WA, USA) and SPSS 13.0 (IBM, Chicago, IL, USA). The levels of As were calculated as the means ± standard deviations (SD). Wilcoxon and Kruskal-Wallis tests were used to check the significance in the concentrations of total As and As species among different samples. The significant difference was considered to be p < 0.05.

Total arsenic concentration

The concentrations of total As in the soil samples are presented in Table 1. The highest level of total As was shown in Site A (16.31 mg/kg) and the lowest level was shown in Site B (13.03 mg/kg). The mean concentrations of total As in Sites A, B and C were 1.5, 1.2, and 1.4 times higher than the background soil values in China, respectively (Wei et al., 1991).

The concentrations of total As in SC and ST are reported in Table 1. The mean concentration of total As in SC from the study area was between 4.64 and 5.68 mg/kg. By comparing the reference value of total As in functional foods (NHFPC, 2014), it was observed that total As in SC was about five times higher. The mean level of total As in ST ranged from 0.82 to 1.13 mg/kg, which was close to the reference value (NHFPC, 2014).

The concentration of total As decreased as follows: soil >SC > ST (p < 0.01, Wilcoxon and Kruskal–Wallis tests, Table S5).

Arsenic species

The concentrations of different As species in soil samples are shown in Table 1 (the chromatograms are shown in Figs. 2F to 2G). The results showed that inorganic As was abundant in Sites A, B, and C, and iAs^V was significantly higher than iAs^III (Table S6). The concentration of organic As was significantly lower than inorganic As (Table S6), and small amounts of AsB were detected in organic As.

The concentrations of different As species in the SC and ST samples were presented in Table 1 (the chromatograms are shown in Fig. 2). Under the H₂O₂ treatment, most of the As species in the large peak area were not oxidized to iAs^V (Figs. 2D to 2E), which proved that the major overlapped peak was not the toxic iAs^III but various unknown organic As species (oAsU). However, it was not possible to evaluate their definite compounds and structures due to the lack of appropriate standards.

Inorganic As was in the minority in the SC samples, in which iAs^III was significantly higher than iAs^V (Table S6). The concentration of organic As was significantly higher when compared with inorganic As. Among these organic As, oAsU was abundant, while DMA and MMA were almost negligible. In the ST samples, inorganic As was also in the minority, while iAs was significantly higher than iAs^III (Table S6). The concentration of organic As was significantly higher than inorganic As (Table S6). Among the detected organic As species, oAsU was the predominant species and AsB and DMA were detected in minor amounts in some samples.

Hazard assessment of the soil

The calculated average daily doses (ADD) for non-carcinogens and carcinogens are summarized in Tables 2 and 3. The ADD decreased through different exposure pathways in the following order: ADD_ing >ADD_inh >ADD_derm, indicating that ingestion is the major exposure pathway. Children are more vulnerable to toxicity than adults because of the higher ADD. The results of human health risk assessment of As in soil suggested that the potential non-carcinogenic risk was negligible since the HI was less than 1 (Table 4). The cancer risks (CR) for human health were at an acceptable level (total CR<1 × 10⁻⁴) (Table 4).

Arsenic transfer chain during Chinese cordyceps formation

As a special organism growing in the Tibetan Plateau (Li et al., 2008), Chinese cordyceps is considered to be a consumer and a de-composer in the food-chain. Our study revealed the total As abundance in the soil of the habitat of the Chinese cordyceps was higher than the average overall abundance. Furthermore, inorganic As accounted for the majority of total As in soil. The original organisms in this soil ecosphere have developed detoxifying strategies to survive and adapt to the toxic circumstances. Although the As transfer chain and corresponding metabolism from the soil to Chinese cordyceps have not been investigated, previous research on plants (Zhao et al., 2009; Lomax et al., 2012), animals (Healy et al., 1998), and fungi (Gonzálvez et al., 2009; Soeroes et al., 2005; Chang et al., 2019) may explain the complicated delivery and transformation of As as follows: first, through the passive absorption from plants’ roots, the original As in the soil is transported and isolated into the plant vacuoles to avoid its toxic effects. During this process, the inorganic As keeps its original speciation because plant cells cannot regulate the methylation of As due to lack of methyltransferase (Zhao et al., 2009; Lomax et al., 2012). The host Thitarodes larvae, which take the plants’ tender roots for two to three years as their preferred food (Fig. 1B), first reduce the ingested iAs^V to iAs^III by their reductase and then methylate iAs^III to low toxic MMA or DMA via methylationase. Subsequently, MMA and DMA are detoxified into other nontoxic As compounds. Notably, fungus also contain methylationase for the methylation (Tang et al., 2016; Zhang et al., 2017). We found that both the As concentration and speciation were significantly different between the soil environment and SC. The larva-fungi union may have highly efficient detoxifying mechanisms through which the inorganic As ingested by Thitarodes larvae had been turned into organic As.

It was not possible to accurately evaluate the effect of Chinese cordyceps on As transformation based on changes in the SC since SC was the complex of host larvae and mycelium. Thus, we focused on the concentration and distribution of As across the ST, which grew only from the Chinese cordyceps without any interference from the host tissue. Here we found that the level of total As from SC to ST has been reduced greatly. The level of iAs^III was significantly higher than that of iAs^V in the SC, but it was the opposite in the ST. The results provided strong evidence that although the host larvae ingested large amounts of toxic iAs^III from the soil due to iAs^III solubility (Andrahennadi & Pickering, 2008), Chinese cordyceps can turn substantial parts into low toxic iAs^V to prevent toxicity to offspring (ascospores in ST).

The cultivation of wild Chinese cordyceps, which occurred from Thitarods in the habitat’s natural soil has not been successful because the occurrence mechanism has been unknown. Artificial laboratory cultivation was based on the cultivated Thitarods fed with prepared feed containing a low As background, and its life span of six months was much shorter than wild Chinese cordyceps (two to three years). Our previous study (Guo et al., 2018a) compared the total As and As species in wild Chinese cordyceps and cultivated Chinese cordyceps. The cultivated Chinese cordyceps were bred under artificial circumstance with trace As in place of the high concentrations of As, which occur naturally on the Tibetan Plateau. Our results showed that As concentration in the cultivated Chinese cordyceps was much lower than that in wild Chinese cordyceps. This finding provided important evidence that the species and As level were affected by the comprehensive function of soils, host larvae, and Chinese cordyceps fungus for wild Chinese cordyceps. It may be inferred, based on the previous study and the results of this experiment, that unlike Laccaria amethystea (Larsen, Hansen & Gössler, 1998) and Collybia butyracea (Kuehnelt, Goessler & Irgolic, 1997) which can accumulate As, Chinese cordyceps can reduce As.

Arsenic concentration in soil and health risk assessment

We found that the total As concentration in soil samples measured by ICP-MS was much higher than the sum of the five As species measured by HPLC-ICP-MS. The difference between the two was unextracted arsenic ores (Liu et al., 2018). Therefore, inorganic arsenic was the predominant form found in the soil and so we took the concentration of total As to assess the potential risk posed by soil arsenic. A previous study reported that the soil’s As level in Lhasa was higher than that in our tested sites (Cheng et al., 2014) and the elevated As concentration may be related to transportation pollutants in addition to the local background values.

Arsenic can exist in almost all environmental media, especially in the soil. It can accumulate in plants and eventually sneak into the body through the food chain (Wei et al., 2016; Tsuda et al., 1992). Animal husbandry and the dairy industry have long occupied the important position in the local economy where this study was conducted. Arsenic can pose significant health risks through the soil-plants-food-human pathway. However, there was no serious threat to human health based on our results, although As geological background value was higher than that in China. It is worth noting that children were generally more susceptible than adults, which is consistent with many other studies (Chen et al., 2019; Li et al., 2018). However, due to the toxicity variations of As species, further studies should focus on the potential risk caused by toxic As species rather than the total As.