The soil was collected from the 0–20 cm layer of a fallow vegetable field located at 23°54′ N and 113°27′ E in Qingyuan City, Guangdong Province, China. After plant residues and stones were removed, the soil was air-dried and sieved to <2 mm. The soil pH, organic C, total N, and C: N ratio were 6.06, 16.6 g·kg⁻¹, 1.30 g·kg⁻¹, and 12.7, respectively, and Pb content was 20 mg·kg⁻¹.

Earthworms of A. aspergillum and E. fetida were purchased from a biofertilizer company in Qingyuan, Guangdong Province, China. Before the experiment, the earthworms were cultured in uncontaminated soil with a moisture content of 40%–60% of field water capacity at 25°C, and organic matter was added for better growth.

Seeds of B. campestris were provided by the Guangdong Academy of Agricultural Sciences.

A total of 96 pots (18.5 cm × 12 cm × 16 cm) were used in this experiment. Each pot was filled with 1 kg soil, which had been spiked with PbCl₂ to set up 4 Pb contamination levels of 0, 100, 500, and 1,000 mg∙kg⁻¹ (OECD, 2000) and had been aged for 10 months. Six treatments were further set up in four replicates for each Pb level: 1) no plant or earthworm (S), 2) B. campestris (SP), 3) A. aspergillum (SA), 4) E. fetida (SE), 5) B. campestris + A. aspergillum (SPA), and 6) B. campestris + E. fetida (SPE). In each pot of the SP, SPA, and SPE treatments, three seedlings of B. campestris were transplanted, and to each pot of SA, SE, SPA, and SPE, 23 ± 1 g of healthy clitellate adult earthworms were introduced. The transplanted seedlings were healthy and uniform in size. Similarly, the introduced earthworms were uniform in weight, 5.73 ± 0.4 g individual⁻¹ for A. aspergillum and 0.4 ± 0.1 g individual⁻¹ for E. fetida.

During the experiment, the temperature was maintained at 25°C, and the light intensity was 400–800 lx. The earthworms were prevented from escaping according to Zhou et al. (2016) and Chen et al. (2017). Soil moisture in each pot was maintained at field capacity. The day before harvesting, leaf chlorophyll concentration was measured on the newest fully expanded leaf using a Konica Minolta SPAD-502 chlorophyll meter (Alordzinu et al., 2021). After 30 days, plant height, shoot diameter, and the number of branches were recorded. The plants were harvested, washed, and oven-dried. The earthworms in each pot were collected, counted, and weighed before being cultured for 7 days at 25°C to empty the guts. The soil in each pot was air-dried and passed through 2-mm and 0.149-mm sieves for chemical analysis.

Soil pH and Eh were measured at 1:2.5 soil: water ratio. Soil total N was quantified by Kjeldahl digestion, organic C was determined by the dichromate digestion method, and CEC was estimated by the ammonium acetate (1 mol L⁻¹, pH 7.0) method (Sparks et al., 1996). Dissolved organic carbon (DOC) was quantified according to Dai et al. (2004). Available N was measured by the alkaline diffusion method (Cornfield, 1960).

Soil total Pb was determined by flame atomic absorption spectroscopy (FAAS) after soil samples were digested with HCl–HF–HNO₃–HClO₄ in an electrically heated block digester (Amacher, 1996). Soil bioavailable Pb was extracted with pH 7.3 diethylenetriaminepentaacetic acid (DTPA) extractant consisting of 0.005M DTPA, 0.1M triethanolamine (TEA), and 0.01M calcium chloride (CaCl₂) at a soil: solution ratio of 1:2 for 2 h, centrifuged at 4,000 g for 20 min, then filtered. Finally, Pb was quantified by FAAS (Lindsay and Norvell, 1978).

For Pb concentration in earthworms, 0.20 g of dried and crushed (0.2 mm) earthworm sample was homogenized with 8 ml concentrated HNO₃ and 2 ml concentrated HClO₄ for 12 h and digested at 250°C for 2 h. After cooling, the solution was diluted to a final volume of 50 ml using deionized water, and Pb concentration was quantified by FAAS. Using a microwave digestion system, 0.20 g of plant sample was wet-digested with aqua regia solution (4 ml HNO₃ and 1 ml HCl), and Pb concentration was quantified using FAAS (Moral et al., 1996).

Leaf area was calculated according to the Montgomery equation (Shi et al., 2019; He et al., 2020): LA   ( c m 2 ) = MP ( LL × LW ) Where LA is leaf area (cm²), LL is leaf length (cm), LW is leaf width (cm), and MP is Montgomery coefficient, which is 0.68.

The transfer factor (TF) of Pb was calculated as (Gupta et al., 2008): TF =   Pb   content   in   the   aboveground   parts Pb   content   in   the   belowground   parts The survival rate (SR) of earthworms (%) was calculated as (Wu et al., 2020a): SR ( % ) = Number   of   earthworms   at   the   end   of   the   exp ⁡ eirment Number   of   earthworms   at   the   start   of   the   exp ⁡ eriment × 100  

Data were processed using the SPSS statistical software and analyzed using one-way ANOVA, followed by Duncan’s multiple range test and t-test. Data are presented as mean ± standard deviation, and the significance level was set at p < 0.05. Principal component analysis (PCA) and co-inertia analysis were performed using the Ade-4 package in R to explore the influences of earthworms on soil properties, Pb bioavailability in soil, and Pb accumulation in the plant (Thioulouse et al., 1997).

In all treatments, A. aspergillum tolerated Pb contamination better than E. fetida (Table 1). At the 0 and 100 mg kg⁻¹ Pb contamination levels, the biomass of A. aspergillum was decreased by 16% and 5%, respectively, and at the 500 and 1,000 mg kg⁻¹ Pb levels, it was increased by 11% and 2%, respectively, in SPA at the end of the experiment compared with at the beginning of the experiment (p > 0.05). At the 0 and 1,000 mg kg⁻¹ Pb contamination levels, the biomass of A. aspergillum was reduced by 12% and 13%, respectively, and at the 100 and 500 mg kg⁻¹ Pb levels, it was increased by 2% and 18% (p < 0.05), respectively, in SA at the end of the experiment compared with at the start of the experiment. The survival rate of A. aspergillum was high in all treatments, ranging from 81% to 100% (Table 1).

The biomass of E. fetida was markedly decreased at the end of the experiment compared with at the beginning of the experiment. At the 100, 500, and 1,000 mg kg⁻¹ Pb contamination levels, the biomass of E. fetida was significantly decreased by 54% (p < 0.05), 47% (p < 0.01), and 38% (p < 0.01), respectively, in SPE at the end of the experiment compared with at the start of the experiment. It decreased significantly by 22% (p < 0.05), 21% (p < 0.05), 22% (p < 0.05), and 24% (p < 0.05) in SE at the 0, 100, 500, and 1,000 mg kg⁻¹ Pb levels, respectively. Generally, E. fetida survived better in the SE treatments, with the survival rates ranging from 77% to 79% across the Pb contamination levels. It displayed lower survival rates in the SPE treatments at the 100, 500, and 1,000 mg kg⁻¹ Pb levels, which were 46%, 53%, and 62%, respectively (Table 1). At the higher Pb contamination levels, the earthworms accumulated more Pb in their bodies (p < 0.05, Figure 1). At the 500 mg kg⁻¹ Pb level, E. fetida accumulated significantly more Pb in SE than A. aspergillum in SPA (p < 0.05).

The co-inertia analysis showed that the first axis explained 34.2% of the total variation, and the second axis explained 18.6% of the total variation (Figure 5). Soil bioavailable Pb and CEC were significantly positively correlated with plant Pb accumulation, biomass, leaf area, height, and shoot diameter (Figures 5A,B), which clearly distinguished the treatments between high and low Pb contamination levels (Figure 5C). Compared with the treatments with low Pb contamination levels, those with high Pb contamination levels exhibited higher soil CEC, available Pb content, plant biomass, leaf area, and plant Pb accumulation. In addition, soil Eh was positively correlated with leaf chlorophyll concentration, while soil pH was negatively correlated with leaf chlorophyll concentration. The treatments with earthworm inoculation (SPA and SPE) were separated from those without earthworm inoculation (SP), with the former associated with higher projected values of soil Eh and leaf chlorophyll concentration but a lower projected value of soil pH.

Previous studies concern the effects of metals on earthworms in terms of loss of weight, mortality, and metal accumulation in tissue (Xiao et al., 2020a; Zhang et al., 2022). until yet, few studies have focused on the differential survival and Pb accumulation between the anecic species A. aspergillum and the epigeic species E. fetida in Pb-contaminated soils.

At all the tested Pb contamination levels in our study, the earthworm species A. aspergillum exhibited a higher survival rate than E. fetida. This earthworm species have been commonly found to survive well in stressful environments. According to Wu et al. (2020b), A. aspergillum survived against soil acidification and aluminum toxicity with a survival rate of 85.8% and an extremely low weight loss rate of 13.4%. In contrast, the biomass and survival rate of E. fetida were low at the tested Pb levels, especially the high Pb levels. Similar findings were reported by Nahmani et al. (2007a) and Sizmur et al. (2011) that the population, biomass, sexual development, and cocoon production of E. fetida were dramatically reduced when the concentrations of heavy metals (e.g., Zn, Cd, Cu, and Pb) in soil exceeded acceptable levels. However, E. fetida had a lower survival rate in treatments that contained plants than those without plants. Our findings could be explained by the possibility that soil Pb and certain compounds in plant root exudates interact to generate toxic molecules that impair E. fetida. For better survival of E. fetida in highly polluted soils, organic matter is often added as a nutrient and energy source (Nahmani et al., 2007b; Zhang et al., 2016; Wu et al., 2020a; Raiesi et al., 2020).

In this study, both earthworm species accumulated Pb in their bodies and accumulated more Pb in the more heavily polluted soils. Studies show that earthworm species differ in feeding preference, behavioral characteristics, and metal bioaccumulation (Hickman and Reid, 2008; Butt and Lowe, 2011; Zhang et al., 2022). Compared with A. aspergillum, E. fetida accumulated more Pb at the 500 and 1,000 mg kg⁻¹ Pb levels. The reason may be because Pb is readily bound to organic matter, and E. fetida requires a large amount of organic matter for food. Therefore, more Pb was accumulated by E. fetida with the ingestion of organic matter. Other researchers reported similar results that E. Veneta, an epigeic species, accumulated significantly more Pb than Lumbricus Terrestris, an anecic species, and Allolobophora chlorotica in 112 days (Sizmur et al., 2011). Earthworm bioaccumulation of Pb can be a useful indicator of soil bioavailable Pb for Pb risk assessment and remediation strategy development (Xiao et al., 2020a; Oorts et al., 2021).

Pb deactivates critical enzymes, inhibits enzyme activities, detrimentally affects various physiological and biochemical processes, impairs photosynthesis, causes cell death, and leads to leaf chlorosis and stunted growth of plants (Zulfiqar et al., 2019). Earthworm introduction has improved plant growth in Pb-contaminated soils (Wang et al., 2006; Yu et al., 2005a). In a Pb/Zn mine tailings-polluted soil with Pb at 1,202 mg·kg⁻¹, the shoot biomass of Leucaena leucocephala was improved from 30.0 to 34 g after the addition of Pheretima guillelmi (Ma et al., 2006). Similarly, E. fetida was reported to considerably promote the growth of maize (Zea mays) and barley (Hordeum vulgare) in soil contaminated with Cu, Cd, Pb, and Zn (Ruiz et al., 2009). With the introduction of Pontoscolex corethrurus, an endogeic and geophagous earthworm species, to a Pb-contaminated soil (1,000 mg·kg⁻¹ Pb), the shoot biomass of Lantana camara increased from 21.52 to 31.77 g, and its root biomass increased from 19.11 to 24.21 g (Jusselme et al., 2013; Jusselme et al., 2012). In this study, the introduction of earthworms significantly increased the biomass, leaf chlorophyll concentration, and leaf area of B. campestris at the high Pb contamination levels (Table 3; Figure 4). Chlorophyll, a primary pigment of green leafy vegetables, was most affected by earthworm inoculation (Figure 4). The significant relationships between leaf chlorophyll concentration and the soil properties of pH, Eh, and DOC (Figure 5) indicate that the earthworms influenced leaf chlorophyll concentration by increasing soil Eh and DOC content and decreasing soil pH. Soil Eh is an indicator of soil aeration status. Earthworm activity in soil improves soil aeration (higher Eh), which improves plant growth and quality (higher chlorophyll concentration) (Noh and Jeong, 2021). The digestion process of earthworms speeds up the decomposition of soil organic matter (higher DOC) (Zhang et al., 2016), which provides more nutrients for plant growth and results in better plant growth with a higher chlorophyll concentration. Soil microorganisms strongly influence the forming and decomposition of soil organic matter (Sokol et al., 2022). Further studies should be performed to elucidate the underlying mechanism of how earthworms affect plant growth at the molecular level (Jacquiod et al., 2020).

The Pb uptake of B. campestris is affected by biotic and abiotic environmental factors (Kaur et al., 2018; Xu et al., 2018; Farahat et al., 2021).

In this study, Pb contamination was the deciding factor in Pb accumulation in the plants (Figure 4). When B. campestris was exposed to high levels of Pb, more Pb was accumulated in the roots, and less Pb was translocated to the aboveground organs (Table 4; Figure 4), consistent with the finding of Zhang Z. et al. (2020). The earthworms affected Pb bioavailability and plant accumulation as well by altering soil properties, increasing DOC, and reducing pH (Figure 5). Some studies indicated the association between the increased bioavailable Pb and the influences of earthworms on soil pH, organic carbon, and DOC (Lemtiri et al., 2016; Nannoni et al., 2014; Wen et al., 2004). The possible explanation is that earthworms may release soluble organic acids, which are responsible for soil pH decrease (Sizmur et al., 2011). Earthworm activities stimulate metal transformation. It was found that available Cd and Zn concentrations in A. morrisi casts were higher than those in the bulk soil, and E. fetida increased the organically bound Pb (Zhang C. et al., 2020). In this study, the inoculation of earthworms significantly affected soil CEC and available Pb content, resulting in higher Pb accumulation in B. campestris and higher biomass and leaf area (Figure 5). Wu et al. (2020b) also found more exchangeable base cations in earthworm casts than in the no-ingested soil. High soil CEC implies high concentrations of such exchangeable base cations as K, Ca, and Mg, which contribute to the increase of plant biomass (Cui et al., 2021). In addition, these cations compete with Pb for binding sites on soil organic matter and oxides, resulting in Pb release into the soil solution and absorption by the plants (Costa et al., 2017; Zhou et al., 2020).

E. fetida increased Pb accumulation in the aboveground and belowground parts of B. campestris by 12.9%–257% and 0.53%–118%, respectively (Table 4). This is mainly due to the increase in plant biomass (Table 3). Kaur et al. (2018) also showed that E. fetida addition mitigated the harmful effects of metals on plant growth, improved photosynthetic efficiency, and boosted metal uptake. Moreover, numerous studies have demonstrated that geophagous earthworms significantly increase Pb uptake of plants by improving Pb availability and plant growth (Ruiz et al., 2011; Ardestani et al., 2019; Zhang C. et al., 2020). In contrast, the introduction of A. aspergillum mainly increased the Pb concentration instead of its accumulation in plants (Table 4). This is due to the higher DTPA-extractable Pb concentration and lower plant biomass in SPA compared with SPE (Figure 2; Table 3). The higher DTPA-extractable Pb concentration in the soil led to higher Pb uptake by the plants, which resulted in the higher Pb concentration in the plants with lower plant biomass. The mechanisms of how different earthworm species influence metal accumulation in plants demand further studies.