P. grandiflorum seedlings were collected from the medicinal plant garden at the Shaanxi University of Chinese Medicine. The seedlings were thoroughly washed and then grown hydroponically in Hoagland’s nutrient solution. After three months, they were exposed to concentrations of 0 μM (CK) and 200 μM Cd(NO₃)₂ Three replicates of each treatment were prepared, with 12 seedlings in each. Cd treatment concentrations arebased on preexperimental study results and described in the supplementary material. The plants were cultivated in a greenhouse with 12 h of light per day, a day/night temperature of 25/20°C, and the nutrient solution was renewed every 3 d. The plants were harvested after 0, 24 h, and 10 days of Cd exposure, and were noted as the Cd 0h group, Cd 24 h group, and the Cd 10 d group, respectively. The controls were recorded as the Cd 0 h-CK group, Cd 24 h-CK group, and Cd 10 d-CK group, respectively. When harvesting the plants, care was taken to keep the roots intact. The plants were then completely immersed in 20 mM Na₂-EDTA for 20 min to eliminate Cd absorbed on the surface and subsequently washed with deionized water. Finally, the plants were divided into root, stem, and leaf tissues, all of which were then used for various research parameter assays.

Fresh samples of P. grandiflorum root, stem, and leaf tissues were frozen in liquid nitrogen and then crushed. The samples were homogenized in a pre-cooled extraction buffer (50 mM Tris-HCl, 250 mM sucrose, and 1.0 mM C₄H₁₀O₂S₂) and the homogenate was centrifuged at 3,000 rpm for 15 min, with the precipitate representing the cell wall fraction (F_CW). The supernatant was taken and centrifuged at 12,000 rpm for 45 min, with the precipitate representing the organelle fraction (F_CO) and the supernatant representing the soluble fraction (F_S) [9]. All centrifugations were conducted at 4°C. The different fractions were digested with HNO₃/H₂O₂ (3:1, v/v), and the Cd content of each fraction was determined using ICP-MS (iCAP RQ, Thermo Fisher Scientific, Bremen, Germany). ICP-MS measurement conditions and sample digestion methods are in the supplementary material.

The chemical forms of Cd were analyzed using successive extraction by designated solutions in the following order: (1) 80% ethanol, extracting inorganic Cd (F_E); (2) deionized water, extracting water-soluble organic Cd (F_W); (3) 1 M NaCl, extracting pectate and protein-integrated Cd (F_NaCl); (4) 2% acetic acid, extracting insoluble Cd-phosphate complexes (F_HAc); (5) 0.6 M HCl, extracting Cd-oxalate (F_HCl) and (6) the residual Cd (F_R) [37]. The root, stem, and leaf tissues were frozen in liquid nitrogen and crushed, homogenized in extraction solution, diluted at a ratio of 1:10 (w/v) and shaken for 22 h at 25°C. The homogenate was then centrifuged at 5,000 g for 10 min, and the supernatant was collected. The precipitate was suspended twice in the extraction solution, shaken for 2 h at 25°C, and then centrifuged. The supernatants of the three centrifugations were pooled, and the solid residue was subjected to the next extraction in the indicated order using the same steps presented above. The five extraction solutions and the residue were dried to a constant weight, digested with HNO₃/H₂O₂ (3:1, v/v), and the Cd content in each fraction was then determined using ICP-MS (iCAP RQ, Thermo Fisher Scientific, Bremen, Germany). ICP-MS measurement conditions and sample digestion methods are in the supplementary material

Quality assurance and quality control (QA/QC) of Cd in P. grandiflorum was performed using the standard reference material of garlic powder (GBW10022) from the Institute of Geophysical and Geochemical Exploration in China, which had a Cd dry weight content of 0.0611 ± 0.0001 mg kg⁻¹ that was consistent with the certified value of 0.062 ± 0.003 mg kg⁻¹ Cd. SPSS statistical software (Version 25.0, SPSS Inc., Chicago, IL, USA) was used to analyze the data, and figures were generated using the GraphPad Prism 9 software (GraphPad Software, San Diego, CA, USA).

Table 1 shows Cd concentration in leaves, stems and roots of P. grandiflorum at different sampling times after treatment initiation. TWO-WAY analysis of variance (3 types of tissues × 3 different times after initiation of the experiment) showed that that the interaction between different tissue types and sampling times had significant effects on Cd concentration (p < 0.01). In the Cd 0 h group, Cd concentrations in stems were significantly higher than those in leaves and roots. However, in the Cd 24 h group, Cd concentrations in roots were significantly higher than those in leaves and stems. In the Cd 10 d group, the concentrations of Cd in roots, stems and leaves were significantly different, with the highest concentration in roots, accounting for 60.89% of the total. The Cd concentrations in the same tissue were significantly different at different harvesting times. The Cd concentrations of the root and leaf tissues increased with increasing Cd exposure time, whereas it first decreased and then increased in the stem tissue. At the end of the experiment, the Cd levels in the root and leaf tissues of the Cd 24 h group increased by 3.2- and 11.3-fold, respectively, and Cd levels in the leaf, stem, and root tissues increased by 7.1-, 6.2-, and 50.8-fold in the Cd 10 d group, respectively.

The results of the subcellular distribution of Cd in different P. grandiflorum tissues after different Cd exposure times are shown in Fig. 1, Table S1 and Table 2. The results of the THREE-WAY analysis of variance (3 types of tissues × 3 subcellular fractions × 6 harvesting times) showed that different subcellular fractions had a significant effect on the proportion of Cd (p < 0.01), but the effect of different tissues and different sampling times on the proportion of Cd was not significant. Furthermore, the interaction effects between tissue type and subcellular distribution, sampling time and subcellular distribution, tissue type, subcellular fraction, and sampling time had a significant effect on the proportion of Cd (p < 0.01) (Table 3).

In P. grandiflorum, the proportion of Cd in the cell wall (F_CW) showed a trend of decreasing and then increasing with increasing exposure time, and the proportion of Cd in the cell wall was significantly lower than the other groups at 24 h exposure (p < 0.01), while there was no statistical difference between the other groups. The proportion of Cd in the organelles (F_CO) decreased gradually with increasing exposure time and was significantly lower than that in the control group (p < 0.01). The proportion of Cd in soluble fraction (F_S) showed a trend of increasing and then decreasing with increasing exposure time and was statistically different between the Cd 24 h group, Cd 10 d group and the control group (p < 0.05). At each harvest time, the cell wall contained the most Cd (50.96%–61.42%), followed by the organelles (26.91%–35.86%) and the soluble fraction (2.22%–22.13%), with statistical differences between the three subcellular fractions (p < 0.01) (Table 2).

In P. grandiflorum leaves, stems, and roots, Cd was predominantly distributed in the cell wall fraction, followed by the organelle and soluble fractions, and there was a statistical difference between the three subcellular fractions in the same tissue (p < 0.01) (Fig. 1). The proportion of Cd in the cell wall of the stem was significantly higher than the proportion of Cd in the cell wall of the leaves and roots (p < 0.05). There was no statistical difference in the proportion of Cd in organelles of leaves, stems and roots. The proportion of Cd in the soluble fraction of the stem was significantly lower than the proportion of Cd in the soluble fraction of the leaf and root (p < 0.05).

In addition, compared to the proportion of Cd in the soluble fraction of each tissue type in the control group of P. grandiflorum, the amounts increased significantly after 24 h and 10 d of Cd exposure (p < 0.01), with the leaf soluble fraction increasing by 2.3- and 1.2-fold, respectively, the stem soluble fraction increasing by 6.6- and 5.0-fold, respectively, and the root soluble fraction increasing 1.5- and 1.4-fold, respectively. However, compared to the control, the proportion of Cd in the organelle fraction decreased in both the Cd 24 h and Cd 10 d treatments in roots, stems and leaves of P. grandiflorum (Table 2).

The proportions and concentrations of each of the six chemical forms of Cd in the root, stem, and leaf tissues of P. grandiflorum at different exposure times is shown in Fig. 2, Table S2 and Table 4. The results of the THREE-WAY analysis of variance (3 types of tissues × 6 chemical forms × 6 harvesting times) indicated a significant effect of chemical form on the proportion of Cd (p < 0.01), while the effect of tissue type and harvesting time on the proportion of Cd was not significant. The interaction effects between tissue type and chemical form, harvesting time and chemical form, tissue type, chemical form, and harvesting time had a significant effect on the proportion of Cd (p < 0.01) (Table 5).

In P. grandiflorum, the proportion of pectate and protein-integrated Cd (F_NaCl) showed a decreasing and then increasing trend with increasing Cd exposure time. The proportion of insoluble Cd phosphate (F_HAc), on the other hand, showed an increasing and then decreasing trend. The proportion of Cd oxalate (F_HCl) increased with increasing duration of Cd exposure and was significantly higher than that of the control group (p < 0.01). The proportion of the residual Cd (F_R) decreased with increasing time of Cd exposure and was significantly lower than that of the control group (p < 0.01). At each harvest time, the residual Cd was significantly higher than the other chemical forms, accounting for 51.12%–80.69% of the total extraction of the six forms of Cd, followed by Cd oxalate (12.07%–37.16%) and insoluble Cd phosphate (6.66%–11.43%), while the other forms accounted for a negligible proportion of Cd (Table 4).

In the leaves of the P. grandiflorum, mainly the residual Cd was present, followed by Cd oxalate, insoluble Cd phosphate and pectate and protein-integrated Cd, while inorganic Cd (F_E) and water-soluble Cd (F_W) were not detected. In the stems of the P. grandiflorum, the proportions of different chemical forms of Cd were, in order, the residual Cd > Cd oxalate > insoluble Cd phosphate > pectate and protein-integrated Cd > water-soluble Cd > inorganic Cd. The proportions of water-soluble Cd, inorganic Cd and pectate and protein-integrated Cd were not statistically different, and the proportions of the residual Cd, Cd oxalate and insoluble Cd phosphate were statistically different (p < 0.01). The proportions of different chemical forms of Cd were in the same order in the roots of the P. grandiflorum as in the stems. However, only the proportions of the residual Cd and Cd oxalate were statistically different (p < 0.01), while the proportions of other chemical forms were not statistically different. Inorganic and water-soluble Cd were only detected in the stems and roots. The proportion of pectate and protein-integrated Cd was significantly higher in P. grandiflorum roots than in stems and leaves (p < 0.01). The proportions of insoluble Cd phosphate and Cd oxalate were significantly higher in the stems than in the leaves and roots of P. grandiflorum, while the proportions of residual Cd were significantly lower than in the leaves and roots (p < 0.01).

In addition, in leaf tissue, the proportions of Cd oxalate and insoluble Cd phosphate were significantly higher in the Cd 24 h and Cd 10 d groups than in the Cd 24 h-CK and Cd 10 d-CK groups, with proportions increasing by 2.4 and 2.1-fold, and 1.7 and 1.4-fold, respectively. In the root tissue, the proportion of Cd oxalate was higher in the Cd 24 h and Cd 10 d groups than in the control group, with their proportions increasing by 2.3 and 3.1-fold, respectively. The proportion of pectin- and protein-bound Cd in stems and insoluble Cd phosphate and Cd oxalate in roots increased with increasing Cd exposure time (Table 4).

The Cd contents of P. grandiflorum root tissues were higher than those in either the stem or leaf tissues, and they accounted for 60.89%–61.64% of the total Cd content of the entire plant. In addition, the Cd contents of the root tissues increased by up to 50.8-fold after the heavy metal stress treatment, and this increase was higher than that occurring in either stem or leaf tissues. These results imply that Cd preferentially accumulates in the roots of P. grandiflorum, and only small amounts are transferred to the stems and leaves. Roots are in direct contact with contaminated substrates and are the primary site for the uptake and accumulation of heavy metals in most plants [38], and a similar trend has been observed in plants such as Phragmites australis [39], K. paniculate [33], and Triarrhena sacchariflora (T. sacchariflora) [40]. Some plants accumulate large amounts Cd in the roots, thus reducing Cd toxicity in other sites. Dong et al. [41] showed that Canna indica L. tolerates Cd toxicity by sequestering Cd in root tissues; after exposure to Cd²⁺ for 45 d, the highest Cd concentration was observed in the root tissue, and this amount was 17- to 47-fold higher than that in the leaves, and 8-to 20-fold higher than that in the stem tissue. In addition, Huang et al. [8] reported that Cd complexing with root cell walls might contribute to Cd tolerance in M. alba L., as 53.27%–70.17% of accumulated Cd was found to be stored in the roots. These studies show that roots are an effective barrier to the transfer of Cd to the aboveground plant body parts [42], and reducing the impact of such heavy metals on the stems and leaves, which is a protective mechanism against heavy metal stress [43]. The results obtained here are thus in agreement with those of previous studies that the immobilization of large amounts of Cd in the roots of P. grandiflorum may be an important tolerance mechanism of this plant.

Once Cd enters a plant, it is selectively partitioned into subcellular components and bound to specific substances within the plant cells. Compartmentalization of toxic ions in different subcellular fractions is an important strategy used by plants to reduce toxicity [44]. Available studies have shown that plants generally store Cd through cell wall deposition and vacuole sequestration [29,36,45]. The experimental results obtained in this study revealed that Cd was mainly accumulated in the cell wall parts of the leaf, stem, and root tissues of P. grandiflorum (50.96%–61.42%; Fig. 1 and Table 2) and that Cd was mainly retained in the root cell walls. The cell wall is the first barrier protecting against heavy metal invasion into the cell and protects the protoplasm from heavy metal toxicity [46]. Heavy metal ions in the cell wall are usually in a bound state and are less toxic than those in the free state in the cytoplasm and organelles [47]. Similarly, Cd in Amaranthus hypochondriacus L. roots was found to be mainly distributed in the cell walls [48], and most of the Cd in K. paniculate was distributed in the cell wall fraction (45%–77%) [33]. Furthermore, in low- and high-Cd-accumulating wheat under Cd exposure, the subcellular distribution of Cd was primarily found in the cell wall [49], and the Cd content of subcellular components in the roots, stems, leaves, and fruits of three pepper varieties under different Cd treatments was reported to be the highest in cell walls [50]. The extent of heavy metal adsorption largely depends on the composition of the cell wall, in which polysaccharides (including cellulose, hemicellulose, and pectin) and proteins are the major components that provide many functional groups (such as carboxyl, hydroxyl, amino, and aldehyde groups), which allow metal cations to bind naturally to the plant [51–53]. These ligands can be involved in various reactions, including ion exchange, uptake, complexation, precipitation, and crystallization [54,55]. Therefore, immobilization of heavy metals in the cell wall can reduce the risk of heavy metal ions entering the protoplasts and can mitigate the effects of heavy metal stress on normal plant physiological activities to a certain extent, thus detoxifying plant tissues [56]. As mentioned earlier, most of the Cd was found to be distributed in the cell wall in this study, and this is likely to be one of the Cd-detoxification strategies of P. grandiflorum.

In addition to the cell wall, numerous studies have reported that vacuoles are important storage sites for heavy metals. Vacuoles can chelate and separate metal ions through the various metal ligands they contain (such as proteins, organic acids, and bases), reducing toxicity and protecting cells [52]. In this study, the proportion of Cd in soluble fractions (Fs) increased in the treated group compared to the control group by 1.4-fold to 6.6-fold (Fig. 1). The soluble fraction consists mainly of vacuoles, comprising up to 90% of the total cellular volume [57], and it is thus speculated that most of the Cd in P. grandiflorum enters the vacuoles after passing through the cell wall. This has been shown to occur in observations of Pistia stratiotes [58], and this strategy might further decrease the amount of Cd interfering with organelles. Previous reports have shown that the largest proportions of Cd in Agrocybe Aegerita [59], Phytolacca americana L. [45], and Sesuvium portulacastrum [4] were accumulated in the soluble fraction, and such results suggest that vacuoles are important Cd storage sites in these species. Similarly, in K. paniculata [33], Coptis chinensis Franch. [9], and Solanum nigrum [36], most of the Cd absorbed was found to be initially distributed in the cell wall fraction, but it later appeared to be distributed in the soluble fraction, which again suggests that vacuoles play an important role in alleviating Cd toxicity. It has been reported that there is a threshold for the barrier effect of cell walls on heavy metals and that heavy metals are transported to vacuoles when the amounts of heavy metals stored in the cell wall reaches saturation [60,61]. Therefore, it is speculated that Cd entering P. grandiflorum cells may first bind to the cell wall, and then if abundant Cd enters the cell under heavy metal stress, the vacuoles preferentially sequester the excess Cd, thus reducing its toxicity to the cell. The synergistic effect of cell wall deposition and vacuole sequestration may thus be an important mechanism for P. grandiflorum resisting Cd toxicity.

Heavy metals are absorbed by the roots and transported through the plant body to different organs and tissues in different chemical forms, especially those which render them unavailable to limit their movement within plant tissues, thereby reducing their toxicity [41]. Heavy metals generally have multiple chemical forms that differ in their ecological effectiveness and toxicity [54]. In this study, Cd in P. grandiflorum was mainly present in the extraction residues and as Cd oxalate and insoluble Cd phosphate, each accounting for 51.12%–80.69%, 12.07%–30.8%, and 6.66%–11.43% of the total extractable Cd, respectively, which indicates that Cd in P. grandiflorum is mainly present in low-mobility and low-toxicity forms. In addition, the percentage of HCl- and Hac-extracted Cd increased after Cd stress; this implies that P. grandiflorum may combine Cd into the less bioavailable Cd oxalate and insoluble Cd phosphate, which reduces the amount of free Cd entering the cytoplasm. Similar results have also been found in other plants. In the leaves and stem of M. alba L., most of the Cd absorbed was extracted by 2% HAc and 0.6 M HCl [8]. Similarly, in barley roots, the Cd forms extracted by 2% HAc and 0.6 M HCl were predominant, and they represented 36% and 33% of the total Cd content, respectively [62]. It is therefore inferred that higher proportions of Cd present in the low-mobility and low-toxicity forms in P. grandiflorum are responsible for its adaptation to Cd-enriched soils.

The results obtained in this study showed that exposure to Cd increased the proportion of Cd extracted by 0.6 M HCl in the root and leaf tissues, and a maximum 3-fold increase was seen compared to control plants. This suggests that the detoxification reaction of P. grandiflorum may involve binding Cd to oxalate to produce insoluble Cd complexes. Oxalic acid is an organic acid, and the chelation of organic acids with heavy metals is also an important heavy metal-detoxification mechanism in plants [63]. A previous study consistently found that when the organic acid concentrations increased, there was a significant statistical correlation between the organic acid levels and the heavy metal concentrations in plants (p < 0.001), and this was presumed to occur because they assist in chelating heavy metals in plants [64]. Organic acids can chelate heavy metal cations through their carboxyl groups (-COOH) to form a complex, which reduces the damage caused by heavy metals to plants [65]. Oxalate is a strong dicarboxylic acid anion and a good complexing agent for the binding of heavy metal cations [66,67]. Consistently, in the Cd-tolerant Lycopersicon esculentum variety ‘Micro-Tom’, Cd-induced oxalate was secreted in high concentrations in the root apex, where most of the Cd was also accumulated [68]. Xin et al. [40] found that Cd mainly exists in oxalate complexes, where it exhibits low activity; this limits symplastic transport and suppresses toxicity in T. sacchariflora seedlings. Similarly, Funneliformis mosseae was found to promote the conversion of inorganic and water-soluble forms of Cd to Cd-oxalate in Sphagneticola calendulacea shoots to improve their Cd tolerance [69]. It is thus considered that in P. grandiflorum, Cd²⁺ is chelated with oxalic acid to increase plant tolerance to Cd. Moreover, it has been shown that the acidic environment of the vacuole provides a favorable environment for the formation of metalorganic acid complexes [70].

Cd occurs in multiple chemical forms, and as shown in this study, different extraction solvents only extract their corresponding Cd chemical forms. The toxicity of different Cd chemicals differs; therefore, the total Cd content in plants does not accurately reflect its toxicity [9]. During production, Chinese medicinal materials are generally extracted by ethanol or water and the residue is then discarded; ethanol is a common extraction solvent used in factories and water is a solvent commonly used to decoct medicine. In P. grandiflorum, inorganic Cd forms (extracted with 80% ethanol) and water-soluble forms (extracted with d-H₂O) were not detected. It is of note that the amount of Cd transferred to the human body is not the total amount of Cd that existed in the Chinese medicinal materials, but it is the amount that was dissolved in the extraction reagents. Therefore, heavy metal limits in Chinese traditional medicines should be based on their chemical form, valence, and other such factors that are associated with their actual toxicity, so that reasonable scientific standard limits can be set for the amounts of heavy metals contained in Chinese traditional medicines. However, for vegetables, the amount and type of Cd contained in the vegetable parts consumed by people are representative of the toxicity of the heavy metal element. As P. grandiflorum is a plant used for both medicinal and food purposes, corresponding heavy metal limit standards should be set according to the plant’s use and the specific valence and chemical form of the metal involved. In this respect, heavy metal allowances should vary according to the different uses of each medicinal and vegetable plant species, particularly for medicine food homology plants.