C. elegans strains were maintained at 20°C on Solid Nematode Growth medium (NGM) using the Escherichia coli OP50 strain as a food source as described (Kim et al., 2010). For ICP-MS analyses, NGM medium was supplemented with other trace elements (1X Mineral Mix) as detailed in supporting data (Supplementary Table S1) and was regarded as supplemental NGM (sNGM). cdf-1(n2527), cdf-2(tm788), sur-7(ku119), ttm-1(ok3503) single mutants and cdf-1;cdf-2, cdf-1;sur-7, cdf-2;sur-7 double and cdf-1;cdf-2;sur-7 triple mutants were the generous gift of Prof. Kerry Kornfeld (Washington University in St. Louis).

Heavy metal sensitivity assays were conducted following our established method (Schwartz et al., 2010a). Briefly, two adult hermaphrodites were placed on solid NGM plates supplemented with indicated concentrations of CdCl₂, ZnCl_2, or CuCl₂. Worms were allowed to lay eggs for 5 h to obtain several dozens of eggs and adult worms were removed to obtain age-synchronized animals. The hatched worms were grown for 3–4 days until the progeny of worms under control conditions reached the adult stage. Heavy metal sensitivity was assessed by comparing the percentage of progeny that have reached the adult stage under heavy metal vs. control conditions, and the morphological changes in intestinal cells were also compared using the Nomarski microscopy. Results are presented as mean values from three independent experiments each of which had three replicates. The total number of worms tested (n) is presented above each graph bar.

The synchronized population of the ∼2,000 L1 stage worms was placed on 100 mm NGM agar plates seeded with E. coli OP50 bacteria and supplemented with or without 50 µM Cd. After 48 h, young adult hermaphrodites were collected from plates with M9 buffer and washed free from E. coli OP50 by two rounds of centrifugation (3,500 × g for 2 min). Worms, resuspended in M9 buffer were concentrated using Ultrafree-Cl Centrifugal Filter Units (MILLIPORE). Total RNA was isolated from worms with TRIZOL reagent (Invitrogen) according to the manufacturer’s recommendations. gDNA was cleared from the RNA preparations by DNAse (Roche) prior to the first strand cDNA synthesis. RT-qPCR was carried out as described earlier (Gayomba et al., 2013). Data were normalized to the expression of actin, act-1. The fold-difference (2^−ΔΔCq) or relative quantities were calculated using the CFX Manager Software, version 1.5 (BioRad).

S. cerevisiae strains used were DY1457 (α ade6 can1 his3 leu2 trp1 ura3), ZHY3 (DY1457 zrt1::LEU2 zrt2::HIS3) (Gitan et al., 2003). Cultures were grown in SD1 medium (0.67% yeast nitrogen base without amino acids) supplemented with auxotrophic requirements and indicated concentration of Cd.

The synchronized population of the L1 stage worms were grown on NGM plates to adult stage. Synchronized young adult worms were collected and transferred to the 60 mm NGM plates seeded with OP50 and supplemented with or without 50 µM CdCl₂. After 24 h, worms were collected and washed with excess S-basal (0.1M NaCl, 0.05 M KH₂PO₄, 5 mg/mL cholesterol) 3 times. Each time, worms were incubated in S-basal for 15 min before centrifugation to remove bacteria from the worm gut. Worms were finally filtered out with Ultrafree-Cl Centrifugal Filter Units (MILLIPORE) at 30 × g for 1 min to remove the remaining bacteria. Worms were then anesthetized with ice-cold 0.2% (w/v) NaN₃ for 2 min. After all the worms were completely immobilized, they were washed twice with the de-icing agent, ice-cold 1.5% (w/v) CH₃COONH₄. Immobilized animals were transferred to 200 nm thick silicon nitrite window (SiMPore, Inc.) and straightened using an eyelash. Excess liquid was removed using fine-tapered paper wick (MiTeGen, United States). Then the window with attached worms was plunge-frozen in liquid nitrogen slush and dried overnight using a Balzers High Pressure Freezer. Freeze-dried worms were stored in a cryovial at room temperature with desiccants.

Dehydrated samples were analyzed by XRF using the X-ray microprobe at 2-ID-E beamline in the Advanced Photon Source (Argonne, IL). An incident energy of 10.5 KeV with a dwell time of 0.1 s for stepsize of 0.5 µm, to excite the Cd L-edge. Full X-ray emission spectra was collected by Vortex 4-element silicon drift detector (SII, Inc.) to obtain the information of S, Cl, K, Ca, Mn, Cu, Fe, Zn, Co as well as Cd. Elemental maps were obtained and quantified using MAPS software (GLOWACKI, 2020).

Worms were grown on OP50 seeded sNGM plates containing ×1 mineral mix (Nass and Hamza, 2007) (to make sure a detectable amount of each element). The sNGM was also supplemented with the indicated concentrations of CdCl₂. Worms were harvested on fourth day of growth and washed free from E. coli OP50 by several rounds of washing and centrifugation as described above. The worm culture was thoroughly mixed and transferred into plastic reservoirs so that they could be pipetted using multichannel pipettes. Meanwhile, a 350 µL AcroPrep™ 96 filter plate with prefilter material/1.2 µm Supor^® membrane (Pall Life Sciences) was wetted with methanol (300 µL/well) and washed with DI water (400 µL/well). The C. elegans cultures were then pipetted into the filter membrane plate (100 µL/well, four replicates per mutant line, i.e., ∼1,000 worms per well), and washed and rinsed with EDTA solution (1 mM, pH 8.0) and DI water, respectively. A total of four such wash and rinse steps were performed (350 µL/well). Note that the filtration step separates the worms from the bacteria feed as the bacteria go through the membrane pores as well. Further, the filter membrane plate was dried (150 min) in an oven at 88°C. Nitric acid (45 µL/well) was added to the dried worms in the filter plate and the samples were digested in a heating block set at 88°C for about 60 min. The digested samples were drawn into a deep-well collection plate containing 0.025% Triton X-100 solution (95 µL/well using vacuum manifold). Deionized water (135 µL/well, 3 times) was then drawn through the filter membrane and also into the collection plate; thus the final solution volume per well in the plate was 500 µL (with final Triton X-100 concentration of 0.005%). Sample solutions were thoroughly mixed and analyzed using a Perkin Elmer DRC II ICP-MS with the ESI (Elemental Scientific, Inc.) SC-2 auto-sampler and the Apex Q sample introduction system. Calibration standards were prepared from single elemental stock solutions containing all the elements of interest (Na, Mg, Al, P, S, Cl, K, Ca, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, Cd, Rb, and Sr). The standards were matrix-matched (i.e., contain Triton X-100 and nitric acid). Note that the Triton X-100 was added to enable smooth self-aspiration of the PFA nebulizer of the Apex Q. The instrument software uses linear calibration to determine the concentrations of the individual elements in the digested C. elegans samples. These concentrations were used together with the dilution factor and the calculated sample weights to determine the elemental concentrations (ppm or molarity) in the original C. elegans samples.

Statistical analysis of data was performed using One-way ANOVA and Student’s t-test depending on the experimental requirements. Sample size (n) is indicated in the Figure legends. Error bars indicate ± S.E.M. Statistical significant differences are indicated with asterisks (*, p ≤ 0.05; **p ≤ 0.01).

Due to the important role of essential and non-essential elements in growth and development, ionomics has become a common practice for establishing the nutritional value of food and the health of animals (Parent et al., 2013; Ma et al., 2015; Zheng et al., 2016). The ionome of plant leaf and unicellular yeast has been found changed according to physiological state, genetic background, and media type (Eide et al., 2005; Baxter et al., 2008). Recently, the ionome of different organs from 26 mammalian species has been analyzed showing a correlation with body mass and longevity (Ma et al., 2015). However, little is known about how the ionome of a whole animal responds to changes in its surrounding environment. To study the effect of Cd exposure on the C. elegans ionome, we first determined the normal concentration of 20 elements, including essential macro- and micro-nutrients, and non-essential potentially toxic elements in Bristol wild-type (N2) worms using ICP-MS (Table 1). To allow the detection of trace metals (e.g., [Cu], [Mn], nickel [Ni], [Co], selenium [Se]) as well as potentially toxic elements (e.g., [As], [Cd], [Rb]), we used supplemented-nematode growth medium (sNGM), by adding each element at concentrations as described in Supplementary Table S1, adapted from Nass and Hamza (2007). Additionally, we also spiked E. coli OP50 culture (used to seed sNGM plates) with the same mineral mix as described above to keep the elemental concentration uniform in treatment. This concentration did not affect the growth of wild-type worms and sNGM was only used for worm cultivation during ICP-MS assays. Analysis using ICP-MS revealed that the accumulation of various elements varied, spanning several orders of magnitude, depending on the element and its function as a macro- or micro-nutrient. For example, worms accumulated 10,883.5 ± 787.9 ppm of the macronutrient phosphorus [P] but 72-fold less of the micronutrient Cu (149.7 ± 13.0 ppm) and 9,302-fold less of another micronutrient Co (1.17 ± 0.2 ppm) as compared to P (Table 1). Cd accumulation was the lowest (0.27 ± 0.01 ppm) among all elements. On the other hand, the concentration of another non-essential and toxic element, As was higher than that of several established micronutrients or beneficial elements such as Ni, Co, and Mo (Table 1). The overall accumulation pattern, from highest to lowest, was as follows (from highest to lowest accumulation): P > K > S > Ca > Mg > Cl > Na > Zn > Mn > Fe > Cu > Se > Al > As > Sr > Ni > Co > Rb > Mo > Cd (Table 1). Comparison of concentrations of these elements in C. elegans vs. their levels in the growth medium (before growing worms) revealed that Cl, Na, and Mo were accumulated to only 30%, 35%, and 46% of the media levels, respectively (Table 1). Notably, the concentration of toxic elements As and Cd were 19 and 54 fold higher than in the growth medium. The baseline concentration of As and Cd in basal media was 1 ppm and 0.05 ppm, respectively. This strongly suggests the existence of mechanisms for their uptake, accumulation and intracellular retention in worms.

To determine the effect of Cd on the C. elegans ionome, we compared the concentrations of different elements in wild-type worms grown on control sNGM media and those grown on sNGM medium supplemented with 5 or 50 μM Cd. 5 μM Cd was chosen because earlier studies showed that this concentration does not alter the growth and development of the wild-type worms but this is the highest concentration that Cd-sensitive hmt-1 mutants can tolerate (Vatamaniuk et al., 2005a; Schwartz et al., 2010a) whereas 50 μM Cd is the highest concentration tolerated by the wild-type worms with minimum signs of toxicity (Vatamaniuk et al., 2005a; Schwartz et al., 2010a). We noted that Cd achieved the biggest fold accumulation difference (253.85 fold & 1,497.12 fold respectively) in worms grown on Cd-containing media compared to worms grown on basal medium resulting in significant changes of the C. elegans ionome (Figure 1). Especially, the concentrations of essential elements such as Zn, Mn, Cu, Ca, and Co were significantly decreased, while the concentrations of potentially toxic elements such as As and Rb were significantly increased compared to control (Figure 1). This pattern was established by normalizing Cd-treatment values with control values for each element. We regarded these changes in the internal ion accumulation pattern as the ionomic signature of Cd toxicity in C. elegans. We speculate that Cd toxicity in worms arises not only from increased Cd accumulation but also from Cd-induced alterations in the ionome.

Our ICP-MS-based studies suggest that, in addition to hyper-accumulation of Cd, the decreased accumulation of essential elements and increased accumulation of toxic elements in Cd-grown worms might contribute to Cd toxicity. If this is the case, supplementing the NGM with essential elements, e.g., Zn and/or Cu could potentially reduce Cd toxicity. Zn and Cu are critical micronutrients and known to exhibit synergistic or antagonistic effects with Cd (Pence et al., 2000; Liu et al., 2008; Nzen et al., 2012; Tkalec et al., 2014; Earley et al., 2021). To test our predictions, we supplemented the medium containing Cd with either Zn or Cu and assessed Cd toxicity in worms. Cd toxicity was evaluated based on the ability of the animals to reach the adult stage and the presence of internal morphological changes, as described previously (Vatamaniuk et al., 2005a). We observed that while 100% of animals reached the adult stage when grown on the NGM without Cd, their development was significantly delayed in the presence of Cd. Specifically, only 82% ± 1.5% and 30% ± 5.4% of animals reached adulthood with 50 or 75 µM Cd treatment, respectively (Figure 2A). Also, worms treated with 75 µM Cd for 4.5 days developed necrotic lesions in intestinal cells (Figure 2D). In contrast to Cd, the supplementation of Zn did not affect the growth or cellular morphology of worms (Figures 2A, E). Control worms grown on regular NGM media show healthy intestinal cells (Figure 2C). Importantly, the addition of Zn rescued the growth and morphological defects observed in Cd-treated worms (Figures 2A, F). The ability of Zn to rescue Cd sensitivity in worms can be explained by two hypothesis: 1) Zn competes with Cd for the uptake leading to a decreased internal concentration of Cd, and 2) Zn and Cd are taken up independently, and additional Zn is needed for protecting Zn-requiring coenzymes (e.g., enzymes that use Zn as a cofactor) from Cd toxicity. To test these hypotheses, we analyzed the concentrations of Zn and Cd in worms that were grown with Cd, Zn or both added to the sNGM media. Consistent with our previous findings (Figure 1), Cd accumulation in worms was associated with a decrease in Zn concentration (Figure 2J). Also, Zn supplementation decreased Cd concentration in worms by 0.62-fold compared to those grown in the presence of Cd alone (Figure 2I), suggesting overlapping uptake pathways for Zn and Cd. Similarly, the Zn concentration in worms treated with Zn and Cd simultaneously was 0.74-fold lower than in worms treated with Zn alone (Figure 2J), however, the Zn concentration was 1.85-fold higher in worms grown with both Zn and Cd compared to control conditions (Figure 2J). Notably, Cd concentration was still relatively high, reaching up to 490 ppm in worms treated with both Cd and Zn, compared to near-zero levels in non-treated control worms. Nevertheless, the majority of worms reached the adult stage, implying that the additional Zn plays a role in Cd detoxification (Figure 2I).

To test whether the effect of Cd on Zn homeostasis is associated with the distribution of Zn uptake and/or internalization, we examined the expression of Zn transporters in Cd-treated worms. Members of two major families of transporters are involved in Zn homeostasis in animals: the Cation Diffusion Facilitator (CDF/ZnT/SLC30) family and the Zrt, Irt-like proteins (Zips/SLC39) family (Eng et al., 1998; Guerinot, 2000; Eide, 2004). The C. elegans genome encodes 14 ZIP and 14 CDF family members. Next, we tested the effect of Cd on the expression of some of these genes. Quantitative real-time PCR (qRT-PCR) analysis revealed that the expression of cdf-1, cdf-2 among CDF family, and most of Zips family [with the exception of C14H10.1 (zipt-13) and T11F9.2b], were significantly upregulated by Cd treatment (Figure 3). This experiment suggests the role of Zn-transporters in encountering Cd-toxicity.

Together, these results indicate that Zn homeostasis is important for basal Cd resistance in C. elegans through the ability of Zn to decrease internal Cd accumulation by competing for uptake, and presumably by protecting internal Zn-requiring cellular proteins against Cd toxicity.

Cd treatment also affects Cu homeostasis in Saccharomyces cerevisiae, Arabidopsis thaliana, and Drosophila melanogaster regulating the expression of genes encoding copper transporters, CTR1, COPT2, and Ctr1B, respectively (Balamurugan et al., 2009; Heo et al., 2010; Gayomba et al., 2013). Because the exposure to Cd also decreased Cu accumulation and the Cu homeostasis was shown to be important in maintaining basal tolerance to Cd in Arabidopsis thaliana (Gayomba et al., 2013), we examined if supplementation with Cu would also rescue Cd-caused toxicity in worms. Therefore, we grew worms with increasing concentrations of either Cu or Cd or both. We found that while supplementing media with lower concentrations of Cu (25 and 50 µM) did not affect the worm growth and cellular morphology, treatment with 100 µM Cu delayed the development (∼70%) and caused necrosis of intestinal cells even without Cd (Figures 2B, G). This confirms that Cu is also toxic to worms at higher concentrations (Yuan et al., 2018; Chun et al., 2017). Furthermore, when Cu and Cd were added simultaneously, Cu exacerbated Cd toxicity, causing a significant decrease in number of worms reaching adulthood and severe necrosis of intestinal cells (Figures 2B, H). Approximately 70% of worms reached the adult stage in the presence of 50 µM Cd + 25 µM Cu whereas only about 25% worms reached the adult stage in the presence of 50 µM Cd + 50 µM Cu, though 100% of worms treated with 50 µM Cu reached the adult stage in the absence of Cd (Figure 2B). Treatment of 100 µM Cu + 50 or 75 µM Cd delayed the growth significantly not allowing more than 1% worms to reach the adult stage (Figure 2B). There are 15 candidate proteins transcribed from 10 gene loci that showed higher orthology to human hCtr1 (Kim et al., 2008), among those, CHCA-1 (F58G6.9)/hCTR1 is responsible for Cu uptake into worm intestine (Yuan et al., 2018). CUA-1 is the only ortholog to human Cu exporter ATP7A/B and also functions in Cu homeostasis in worms (Chun et al., 2017). RT-qPCR analysis revealed that chca-1 and its closest homologs (F58G6.7, F58G6.8) were significantly downregulated by Cd along with K12C11.3 but two of the candidates were upregulated, i.e., F27C1.2 and F31E8.4 (Supplementary Figure S1), nevertheless, CUA-1 expression was unaltered (Supplementary Figure S1). Past studies have shown that CHCA-1 downregulation leads to a decreased Cu accumulation in worms and that CUA-1 does not respond to Cu concentration at the transcriptional level (Yuan et al., 2018; Chun et al., 2017). By contrast, studies in plants, have shown that Cd toxicity increases the expression of hCTR1 homologs, COPT1, COPT2, and COPT6, and stimulates Cu uptake (Gayomba et al., 2013). So, our findings are consistent with existing studies on the crosstalk between Cu homeostasis and Cd resistance. However, while in plants, Cu is thought to protect from Cd toxicity, Cu might exacerbate Cd toxicity in animals. We speculate that the decreased Cu accumulation in Cd-cultured worms (Figure 1) may be serving a protective function to reduce the cellular damage caused by Cd toxicity.

Earlier studies showed that in addition to cdf-1, cdf-2 and sur-7, ttm-1, another member of the CDF family was also transcriptionally highly activated by Cd (Huffman et al., 2004; Nass and Hamza, 2007). We next tested whether the function of CDF-1, CDF-2, SUR-7 and TTM-1 is required for basal Cd resistance in C. elegans. To do so, we examined Cd sensitivity of worms that lack one or more of these transporters. cdf-1, sur-7 and ttm-1 mutants were as sensitive to Cd as either of the single mutants, except for cdf-2, implying that these genes share a common pathway to confer Cd resistance. To further test the function of Zn transporters in Cd resistance, we took advantage of a heterologous yeast system. In S. cerevisiae, Zn uptake is primarily mediated by Zrt1 and Zrt2 transporters (Gitan et al., 2003). We grew WT (wild type) and ZHY3 (zrt1; zrt2) yeast cells on different concentrations of Cd and found that ZHY3 is more sensitive to increased concentrations of Cd than WT (Supplementary Figure S2). This also suggests that in the absence of Zn importers (zrt1;zrt2), cells cannot uptake enough cellular zinc to counter Cd toxicity. Taken together, we conclude that Zn homeostasis is essential for basal Cd resistance.

The ICP-MS-based studies enabled us to quantify concentrations of different elements including Cd, and the effect of Cd on the equilibrium of physiologically relevant, as well as potentially toxic metals (Figure 1). While these analyses are very informative, they do not provide information about the spatial distribution of Cd and other elements at cellular levels. It is possible that the spatial distribution of Cd and its effect on the distribution of other elements might be among the underappreciated bases for Cd-caused diseases.

To analyze the spatial distribution of Cd and other elements in worms we employed X-ray fluorescence microscopy (XRF), which has become the method of choice for visualizing the distribution of transition metals in situ (Finney et al., 2007; Tian et al., 2011). For comparative analysis, young adult worms were grown on 0 or 50 μM Cd containing NGM plates for 24 h and then subjected to XRF analyses. One of the primary tissues in C. elegans, the intestine plays a crucial role in the digestion and assimilation of food as well as the production and storage of miro- and macromolecules. C. elegans’ intestines are comprised up of 20 large epithelial cells that are arranged in bilaterally symmetric pairs to form a long tube that encloses a lumen (Figure 5A, McGhee, 2007). We found that the bulk of Cd was localized in intestinal cells in animals treated with Cd, where it exhibited a punctate distribution with peaks in the vicinity of nucleus-like structures (Figures 5B, D). Consistent with the results from the ICP-MS study (Figure 1), Cd exposure correlated with a decreased concentration of biologically essential elements such as Ca, Cu, Zn, Mn, Co (Figure 5B, compare upper and lower panels). In addition, the concentration of S, K, Cl was also reduced in Cd-grown vs. control worms (Figure 5B, compare upper and lower panels). While Cd did not alter the spatial distribution of the majority of analyzed elements, profound differences were found in the spatial distribution of Fe (Figure 5C). Whereas Fe was localized evenly at low concentrations through the entire intestine of worms grown under control conditions, it accumulated at a high concentration in discrete structures resembling the nucleus of intestinal cells (Figure 5C). Furthermore, a large portion of Fe and Cd co-accumulated as indicated by arrows in intestinal cells (Figure 5D).

Taken together, these results suggest that Cd affects the C. elegans ionome by lowering the concentration of diverse trace elements and contributes to the altered spatial distribution of Fe in intestinal cells.