Microplastic–Cadmium Interaction in Paddy Soils: An Overlooked Risk Exacerbating Cadmium Contamination in Rice and Microbial Dysbiosis

TL;DR

Microplastics and cadmium together harm rice growth and increase cadmium uptake, while altering soil microbes in flooded rice paddies.

Contribution

Reveals how microplastics enhance cadmium bioavailability and toxicity in rice paddies through soil property changes and microbial shifts.

Findings

Combined microplastic and cadmium stress reduced rice growth and increased cadmium accumulation in plant tissues.

Microplastics altered soil properties like pH and organic matter, indirectly suppressing rice yield.

Microplastics intensified cadmium-induced oxidative stress and shifted microbial communities toward cadmium-tolerant taxa.

Abstract

The co-occurrence of microplastics (MPs) and cadmium (Cd) in agricultural soils poses ecological risks, yet their interactions in flooded rice paddies remain unclear. Therefore, this study investigated the individual and combined effects of polyethylene MPs (mPE) and Cd on rice (Oryza sativa L.) growth, Cd accumulation, and soil microbial communities. Combined stress (5 mg/kg Cd + 1% mPE) significantly reduced rice growth (4.1–13.8% in plant height) and increased Cd accumulation in roots, stems, and seeds, driven by MP-enhanced Cd bioavailability. MPs altered soil pH, organic matter (OM), and moisture content (MC), indirectly suppressing yield. Microbial analysis revealed decreased bacterial alpha diversity (0.86–8.36%), favoring Cd-tolerant taxa (e.g., Solirubrobacteraceae), while fungal responses were weaker under flooding. Structural equation modeling indicated that Cd exerted direct toxicity through tissue accumulation, whereas MPs acted indirectly by modifying soil properties and inducing oxidative stress. Under co-exposure, MPs intensified Cd-induced oxidative stress, enhancing both direct and indirect toxicity pathways. Mantel tests identified DTPA-extractable Cd (r = 0.70) and OM (r = 0.55) as key drivers of Cd uptake. These findings highlight the complex interplay of MPs and Cd in rice paddies, with implications for managing co-contaminated agroecosystems.

Figures

Full text

1. Introduction

Microplastics (MPs, <5 mm) have emerged as a pervasive environmental contaminant, with agricultural soils worldwide containing 10–13,000 particles/kg [1]. These particles infiltrate terrestrial ecosystems through pathways such as plastic mulch fragmentation, organic fertilizer application, and atmospheric deposition, ultimately threatening agroecosystem stability and food safety via plant uptake and trophic transfer [2,3].

Of particular concern are their interactions with co-occurring contaminants like heavy metals, which may exacerbate ecological risks through synergistic effects [4]. Recent evidence indicates that rice paddies located near industrial clusters and urban centers are particularly vulnerable, as industrial effluent and atmospheric deposition contribute significant loads of both cadmium (Cd) and plastic debris [5]. Furthermore, studies in paddy soils confirm that microplastics and cadmium can co-exist under flooded conditions, where the presence of microplastics affects Cd availability and soil biogeochemical processes [6]. Studies confirm that sewage sludge application can increase soil microplastic loads by several hundred percent, persisting for decades, while simultaneously introducing heavy metals like Cd into the soil matrix [7].

Once infiltrated into the soil, MPs alter key physicochemical properties, such as porosity and aggregation, which indirectly modulate biogeochemical processes and metal bioavailability [8,9]. MPs exhibit dual roles in heavy metal dynamics: they may either enhance metal mobility through adsorption–desorption cycles or reduce bioavailability via surface complexation. For example, polyvinyl chloride MPs (mPVC) increased chromium (Cr, VΙ) accumulation in sweet potato tubers by 28%, amplifying phytotoxicity [10]. MPs can increase the absorption of heavy metals including lead (Pb), Cadmium (Cd), copper (Cu), and zinc (Zn), by lettuce through promoting the abundance of metal-activation bacteria in rhizosphere soil [11]. While polylactic acid MPs (mPLA) reduced cadmium uptake in Brassica chinensis L. by 19% through competitive adsorption [12]. Such contradictory findings highlight the context-dependent nature of MP–metal interactions, necessitating crop- and environment-specific investigations.

Rice (Oryza sativa L.) is one of the world’s three principal cereal crops and is cultivated on more than 1.4 × 10^9^ hectares of land globally [13]. Cd is a primary pollutant of farmland and rice in southern China, attracting considerable attention due to its severe phytotoxicity and pronounced bioaccumulation potential [14]. Emerging evidence highlights the widespread co-contamination of soils by Cd and MPs. For instance, MPs can alter soil physicochemical properties and enhance Cd bioavailability through adsorption–desorption dynamics [15]. Additionally, MPs may act as carriers for heavy metals, facilitating their vertical migration or modifying microbial-mediated metal transformations [16]. These interactions can disrupt soil microbial community structure and function, further influencing nutrient cycling and plant health [15,17,18].

However, current research predominantly focuses on dryland crops or simplified hydroponic systems, while the unique flooded conditions of paddy fields remain understudied. Unlike dryland soils, the anaerobic environment and dynamic redox shifts in rice paddies may modulate MPs’ aging processes and their synergistic effects with Cd [19,20]. Specifically, the role of soil microorganisms in mediating Cd-MP interactions under flooding conditions, a critical driver of Cd speciation, has yet to be systematically explored. Previous studies lack integration of microbial metabolic activity with metal bioavailability assessments, limiting mechanistic understanding of the soil-rice system.

Therefore, we hypothesized that MPs can accelerate Cd accumulation in rice tissues and reduce the microbial diversity in Cd-contaminated soil. This study aims to (1) evaluate the potential impact of Cd and MPs on rice growth and Cd accumulation in rice; (2) examine the impacts of MPs and Cd on the physicochemical properties and enzyme activities in paddy soil; (3) analyze the influence of Cd and MPs on soil microbial communities in paddy soil. This study can provide insights into the ecological risks and mechanisms of MPs in Cd-contaminated paddy fields.

2. Materials and Methods

2.1. MPs

Polyethylene (PE), the main constituent of agricultural mulch films, was selected as the target MP. Polyethylene MPs (mPE, additive-free, size: approximately 148 μm) were purchased from Shunjie Technology Co., Ltd. (Dongguan, China). Their particle size, surface morphology and functional groups were characterized by Laser Diffraction Particle Size Analyzer (MAZ3000, Mastersizer, Malvern, UK), Field Emission Scanning Electron Microscopy (FESEM, Verios G4 UC, Thermo Fisher Scientific, Waltham, MA, USA) and Fourier Transform Infrared Spectroscopy (FTIR, TENSOR27, Bruker, Ettlingen, Germany), respectively (Figure S1a–c; Table S1).

2.2. Pot Experiment

Soil was collected from Shilin County, Yunnan Province, China (24.64° N, 103.25° E), air-dried at 15–20 °C, and sieved through 2 mm. The soil contained 0.42 mg/kg Cd, below the Chinese screening threshold of 0.6 mg/kg [21]. Key soil properties were shown in Table S2. Rice seeds (Dianyou 42) were obtained from the Hybrid Rice Research Center, Yunnan Agricultural University. After surface sterilization with 5% NaClO for 10 min, seeds were germinated under controlled conditions (28 °C/20 °C, 12 h light/dark, 32,000 lux, 60–70% RH). Germination continued for approximately 20 days until the two-tiller stage.

The selection of Cd and mPE concentrations was based on environmental relevance and the requirement for a mechanistic gradient. Two Cd levels (1.5 and 5 mg/kg) were designed to represent a typical contamination gradient: 1.5 mg/kg reflects moderate-to-high pollution levels frequently reported in agricultural soils near industrial or mining areas in southern China [4], while 5 mg/kg simulates a “worst-case” scenario in severely degraded land. For mPE, concentrations of 0.5% and 1% (w/w) were utilized to represent contamination “hotspots” (e.g., intensive plastic mulching or sewage sludge application) and to project future high-accumulation scenarios [9,22]. Accordingly, the pot experiment was designed with three Cd levels and three mPE levels, resulting in 9 treatments with three replicates each: CK, 0.5% mPE, 1% mPE, 1.5 mg/kg Cd, 1.5 mg/kg Cd-0.5% mPE, 1.5 mg/kg Cd-1% mPE, 5 mg/kg Cd, 5 mg/kg Cd-0.5% mPE, and 5 mg/kg Cd-1% mPE.

Each pot (27 cm top diameter × 25 cm height) contained 6 kg soil thoroughly mixed with Cd and/or MPs. After a 30-day equilibration, three uniform seedlings were transplanted into each pot. Basal fertilizer (N:P:K = 15:7:12) was applied at 600 kg/ha. To simulate paddy field conditions, soil moisture was maintained under continuous flooded conditions throughout the rice growing period. Pots were maintained in a greenhouse (20–35 °C, 60–70% RH, natural photoperiod) and repositioned weekly. Rice plants (mature period) were immediately harvested, and soil samples were collected 180 days after potted cultivation. Plant traits (height, root length, biomass) were recorded, and tissues (root, stem, leaf, grain) were rinsed and stored for analysis. Soil was subsampled for physicochemical properties, enzyme activities, and microbial diversity. Detailed methods are provided in the Supplementary Materials Text S1.

2.3. Sample Analysis

Enzyme activities in rice leaves, including peroxidase (POD), superoxide dismutase (SOD), catalase (CAT), and malondialdehyde (MDA), as well as soil enzyme activities such as urease (S-UE), invertase, dehydrogenase, β-glucosidase, and alkaline phosphatase, were measured using commercial kits from Nanjing Jiancheng Bioengineering Institute, Nanjing, China. Soil pH was measured with a pH meter (Puchun PHS-3C, Shanghai, China) [23]. Organic matter (OM) was determined using the external heating method [23]. Moisture content (MC) was measured gravimetrically [24]. Available Cd, total Cd, and Cd fractions (residual, Fe-Mn oxide-bound, organic, exchangeable + carbonate-bound, water-soluble) were measured by Inductively Coupled Plasma Mass Spectrometry (ICP-MS, ICAP RQ, Thermo Fisher, USA) [25,26,27]. Further details are provided in the Supplementary Materials Text S2.

2.4. Bioconcentration Factor and Translocation Factor

2.4.1. Bioconcentration Factor

Bioconcentration factor (BCF) is a critical parameter for assessing heavy metal accumulation in crops, reflecting the capacity of different plant parts to accumulate heavy metals. When the BCF > 1, it signifies a strong ability of the crop to accumulate heavy metals. The formula for calculating BCF is as follows:

$$[eqn]$$

where $[eqn]$ represents the BCF of Cd in the rice part $[eqn]$ , and $[eqn]$ is the Cd concentration in the rice part $[eqn]$ , while $[eqn]$ is the Cd concentration in the corresponding root zone soil.

2.4.2. Translocation Factor

Translocation Factor (TF) was employed to quantify the efficiency of Cd movement from the soil into the plant and its subsequent migration between different tissues [28]. To evaluate the internal migration of Cd among the four primary plant parts, TF was calculated for the following pathways:

$$[eqn]$$ $$[eqn]$$ $$[eqn]$$

where $[eqn]$ , $[eqn]$ , $[eqn]$ , and $[eqn]$ represent the Cd concentrations (mg/kg, dry weight) in the respective rice tissues.

2.5. Microbial Diversity

The introduction of high pollutant concentrations inevitably alters soil microbial community composition. Thus, the concentrations of the following pollutants were measured: control group (CK), 0.5% mPE, 1.5 mg/kg Cd, and 1.5 mg/kg Cd-0.5% mPE (n = 3 per treatment, 12 samples total). Sequencing and bioinformatics were performed at Majorbio Biopharmaceutical Technology Co., Ltd. (Shanghai, China) using the Majorbio Cloud platform (https://cloud.majorbio.com). The microbiological diversity analysis encompassed operational taxonomic unit (OTU) identification, Alpha/Beta diversity analysis, ANOSIM/Adonis tests, inter-group significance tests, and Linear Discriminant Analysis Effect Size (LEfSe) multilevel species discriminant analysis. Detailed methods are provided in the Supplementary Materials Text S3.

2.6. Statistical Analysis

Results were presented as mean ± standard deviation, with significant differences determined by the LSD at p < 0.05 via SPSS 26.0. Structural equation modeling (SEM) was conducted using R 4.5.1, with the partial least squares (PLS) approach applied for data analysis. The Mantel test correlation heatmap was generated via the Chiplot platform (https://www.chiplot.online). Figures were created using Origin 2025 and Edraw 14.5.5.

3. Results and Discussion

3.1. Growth of Rice Plants

Rice growth was affected by mPE and Cd content, with increasing pollutant levels leading to enhanced inhibition (Figure 1a). The inhibitory effects of 1 and 5 mg/kg Cd on rice were more pronounced than those of 0.5% and 1% mPE. Nevertheless, compared with CK, neither single mPE nor Cd treatments significantly inhibited rice growth (p > 0.05). These results were consistent with previous studies, where 2% polyethylene terephthalate MPs (mPET) or 5 mg/kg Cd did not significantly inhibit rice biomass [4]. By contrast, 2% polypropylene MPs (mPP) markedly reduced maize biomass [18]. This discrepancy may be attributed to differences in MP types and particle sizes. For instance, this study and Liu, Cui, Li, Xu, Sun, Xu and Wang [4] used 165 μm mPE and 51 μm mPET, whereas Zhao, Xu, Wang, Li, Zhao, Cao, Zhang, Zhang, Wang, Chen and Zou [18] investigated 50–100 nm polystyrene (mPS). Smaller-sized MPs (<200 nm) generally have more pronounced inhibitory effects on plant growth, as they can be internalized by roots, causing mechanical injury and oxidative stress [29,30]. Larger-sized MPs may inhibit root elongation growth by blocking ion channels [31]. Furthermore, variations in adsorption specificity, degradability, and degradation products contribute to differing toxicities [32]. MPs can also modify soil nutrient dynamics, reducing available phosphorus and potassium, and thereby constraining crop growth [33].

Under combined exposure, growth inhibition was significantly amplified. High-level co-contamination (5 mg/kg Cd-1% mPE) reduced plant height, root length, productive ears (NPE), kernels per spike (KPS), thousand kernel weight (TKW), and yield compared to CK (p < 0.05) (Figure 1a). This suggested that MPs intensified Cd phytotoxicity, likely by facilitating Cd migration into rice plants [34]. Similar synergistic inhibition of maize growth was reported under 1% mPE-15 mg/kg Cd, with plant height reduced by 4.1% to 13.8% [17]. Conversely, some studies reported that 2% mPS can mitigate the growth-inhibiting impact of 10 mg/kg Cd on maize and Brassica chinensis L., likely mPS functioning as physical barriers that reduced water uptake and oxidative stress [29].

Rice oxidative stress responses supported the growth results. SOD, POD, CAT, and MDA levels increased under Cd and mPE stress (Figure 1b), indicating elevated oxidative stress and activated antioxidant defenses, consistent with previous reports of MP-induced redox imbalance [20]. Detailed results and discussion are presented in the Supplementary Materials Text S4.

3.2. Cd Content and Transport in Rice Tissues

Co-exposure to mPE and Cd induced significant tissue-specific Cd accumulation in rice (Figure 1c). Roots exhibited the highest Cd levels (1.72–3.25 mg/kg), consistent with roots as primary entry points. Specifically, under single mPE stress, only the stem Cd content increased 23.0–36.4% compared to the CK (p < 0.05). Single Cd treatments raised Cd across all tissues, with significant increases at 5 mg/kg Cd compared to CK (p < 0.05). Under combined mPE and Cd treatments, Cd accumulation in roots, stems, and seeds was significantly higher than in single Cd treatments (p < 0.05), particularly in 5 mg/kg Cd-1% mPE, demonstrating that MPs facilitate Cd uptake. These findings align with reports of MPs enhancing Cd accumulation in plants [18].

MPs could enhance Cd availability, which may be the primary driving factor for the uptake by plants [35]. MP surfaces carry various substances like heavy metals, dissolved organic matter (DOM), and surface-active agents, potentially increasing Cd solubility or bioavailability in soil [36]. Additionally, MPs can also alter soil pore structure and permeability, accelerating Cd transport to roots, and shift microbial communities to modify nutrient cycling and Cd speciation [37,38]. However, a previous study indicated that MPs reduce Cd bioavailability by blocking reactive sites [39]. This result contrasts with this study, and the discrepancy may be attributed to variations in experimental conditions, such as simulating gastrointestinal environments versus crop cultivation.

Cd translocation was also altered. In the combined treatment groups, mPE facilitated Cd transport from roots to stems, but inhibited transfer from stems to leaves and seeds (Figure 1d). Similar patterns were noted in maize [18], likely because MPs enhance root Cd absorption from soil, leading to higher root Cd levels. Under single mPE treatments, Cd enrichment factors in roots were not significantly different from CK. In contrast, BCF under single Cd and composite (i.e., Cd-mPE) treatments were lower than CK due to higher absolute Cd concentrations in contaminated soils (Figure 1e). Notably, BCF was higher in combined treatments than in single Cd treatments, with 1.5 mg/kg Cd-1% mPE showing significantly higher enrichment than 1.5 mg/kg Cd alone (p < 0.05). This phenomenon can be attributed to MPs promoting Cd sequestration in roots, potentially by stimulating metabolic pathways associated with amino acid synthesis, aromatic compound degradation, as well as pantothenic acid and coenzyme A [40]. Overall, these results confirm that Cd predominantly drives tissue accumulation, while MPs exacerbate Cd uptake and redistribution within rice.

3.3. Soil Properties and Enzymatic Activities

Co-exposure to mPE and Cd significantly altered soil physicochemical properties. Soil pH increased under 1% mPE and 1.5 mg/kg Cd-0.5% mPE treatments compared to CK (p < 0.05). Conversely, soil MC declined (Figure 2a). This observed decline in MC coupled with the increase in pH suggests a physical-to-chemical cascade triggered by mPE. Mechanistically, the incorporation of hydrophobic mPE particles can disrupt the continuity of soil capillaries and increase macro-porosity [41]. This structural shift not only reduces the soil’s water-holding capacity but also improves localized aeration within the typically anaerobic flooded paddy soil. Enhanced aeration can stimulate the activity of aerobic decomposers, potentially accelerating the mineralization of native soil organic matter (SOM) or the transformation of root exudates [8,18], which aligns with the observed upward trend in SOM (Figure 3a). Furthermore, the increase in pH may be linked to altered nitrogen dynamics; improved micro-aeration can promote nitrification/denitrification processes that involve the consumption or release of H^+^ ions, thereby shifting the rhizosphere pH balance [4,42,43]. These changes collectively indicated that MPs indirectly suppress rice growth via soil property modification.

The incorporation of mPE markedly modulated Cd bioavailability and speciation (Figure 2b). Interestingly, single mPE treatment had negligible effects on DTPA-extractable Cd due to a “dilution effect” from its low sorption capacity compared to soil minerals [44]. Specifically, DTPA-extractable Cd and total Cd in the combined treatment were significantly lower compared to the single Cd treatment (p < 0.05). This reduction in the soil Cd pool, coupled with the observed increase in tissue accumulation (Figure 1c), suggested that mPE facilitates the transfer of Cd from soil to rice. The discrepancies between our findings and previous dryland studies likely stem from the distinct redox chemistry of paddy ecosystems and mPE surface properties [45,46]. In this study, the relatively short exposure duration and anaerobic flooded conditions inhibited the extensive photo-oxidative aging of mPE. Consequently, the mPE surfaces remained predominantly hydrophobic with fewer oxygen-containing functional groups compared to aged MPs in dryland soils. This limited the direct adsorption of Cd^2+^, allowing mPE to act instead as a physical disturber of the soil matrix. Moreover, our soil’s specific texture (silty clay) might have amplified the “dilution effect” of mPE on soil minerals, leading to a more pronounced redistribution of Cd into exchangeable fractions than what has been observed in sandy or loamy soils.

At the phylum level, the dominant bacterial taxa with relatively high abundance were Actinobacteriota (22.31–30.32%), Chloroflexi (18.66–22.69%), and Proteobacteria (11.98–18.82%) At the phylum level, dominant bacterial taxa included Actinobacteriota (22.3–30.3%), Chloroflexi (18.7–22.7%), and Proteobacteria (12.0–18.8%) (Figure 3b), which was consistent with previous studies [8]. Dominant fungal phyla were Ascomycota (67.5–76.4%), Unclassified fungi (13.8–22.7%), and Mortierellomycota (2.8–7.4%) (Figure S2b). Overall, bacterial taxa were more abundant than fungal taxa. Specific abundance data of bacteria and fungi were provided in Tables S3 and S4, respectively.

Mechanistically, analysis of Cd speciation further illustrated these processes (Figure 2b). Although water-soluble Cd decreased with MP addition, likely due to direct adsorption onto mPE surfaces [47], mPE contributed to a higher proportion of exchangeable-carbonate Cd in soil in this study, reflecting MPs’ charged functional groups compete with other metal ions for adsorption or ion exchange, potentially leading to the transformation of Cd from less active forms to more active exchangeable forms [48]. This mPE-induced transformation toward active speciation, regulated by the concomitant shifts in pH and SOM, directly explains the substantial elevation in rice root Cd accumulation [49]. A previous study indicated that higher soil pH reduces DTPA-extractable Cd by enhancing negative charges on soil surfaces, decreasing Cd^2+^ affinity [50]. Moreover, OM affects DTPA-extractable Cd by complexing with Cd, adjusting soil pH and microbial activity, enhancing cation exchange, and improving soil structure, thereby reducing Cd mobility, solubility, and bioavailability [51,52]. Thus, mPE acts as a “transient carrier” or “activator” rather than a permanent sink for Cd in the rhizosphere.

Soil enzymatic activities were suppressed by MPs to varying degrees, though most differences were not statistically significant (Figure 2c). CAT activity remained relatively stable, potentially reflecting a balanced state of oxidative stress regulation in the soil microbial community despite the physical presence of mPE. Detailed results and discussion are presented in the Supplementary Materials Text S5.

3.4. Soil Microorganism Community Structure and Diversity

3.4.1. Microorganism Community OTU and Structure

To assess the impact of mPE and Cd on soil microbial diversity, four treatments were examined: CK, 0.5% mPE (mPE), 1.5 mg/kg Cd (Cd), and 1.5 mg/kg Cd-0.5% mPE (Cd-mPE). In bacteria analysis, only 3 shared OTUs (accounting for 3.03%) were detected across treatments, with CK showing the highest number of unique OTUs (781) compared with mPE (700), Cd (692), and Cd-mPE (674) (Figure 3a). This indicates that both MPs and Cd reduced bacterial richness. Nevertheless, the fungal community in the soil exhibited an opposite trend. The fungal communities exhibited 41 shared OTUs (3.06%), and CK contained the fewest unique OTUs (Figure S2a). This difference may be that fungi are generally more sensitive to alterations in soil compared to bacteria, leading to more significant responses to MPs and Cd [53]. Cd toxicity and physical obstruction by MPs likely contributed to fungal decline [54], while slower growth rates and narrower ecological tolerance render fungi more vulnerable [55]. Conversely, bacteria can resist pollutants and adapt to environmental changes via gene mutations or by acquiring exogenous genes [56], explaining their resilience (e.g., OTUs increasing) under heavy metals and MP stress.

3.4.2. Alpha Diversity and Beta Diversity Index

Bacterial alpha diversity indices (Sobs, Shannon, ACE, Chao) decreased by 0.86–8.36% under Cd-MPs co-exposure, while the Simpson index increased by 20.2% (Figure 3c). Reduced Shannon and Chao indices indicate niche contraction, whereas higher Simpson reflects dominance of Cd-tolerant taxa [57]. Interestingly, ACE and Chao indices in Cd-mPE exceeded those in Cd-only soils, suggesting that MPs partially mitigated Cd suppression of microbial diversity, possibly by supplying stable hydrocarbons as carbon sources [8]. Although higher microbial diversity usually benefits soil health, the suppression of rice growth likely arose from Cd stress amplified by mPE-induced Cd mobilization. Specific microbial taxa enriched in mPE-treated soil (e.g., Cd-tolerant or metal-reducing bacteria) might compete with rice for nutrients or indirectly exacerbate Cd bioavailability. Thus, the ability of mPE to reduce the Cd levels in the soil may be an important factor promoting the increase in soil microbial diversity. MP type also influenced outcomes; for example, poly (butylene succinate) MPs (mPBS) and mPLA can release succinic acid or lactic acid, altering microbial composition [58]. By contrast, fungal alpha diversity showed no significant changes (Figure S2c), consistent with flooding-related fungal inhibition [18]. These results highlight that MP effects on microbial communities are context-dependent and shaped by soil, crop, and polymer type.

Community composition analyses (PCA, PCoA, NMDS) revealed distinct clustering of MP-containing groups (mPE, Cd-mPE) compared with CK and Cd treatments (Figure 3d–f). An NMDS stress of 0.1008 confirmed significant effects on bacterial communities. ANOSIM and Adonis analyses demonstrated that the inter-group variances of bacterial communities exceeded the intra-group variances, indicating a relatively substantial influence of mPE and Cd on soil bacterial communities (Figure S2g, Tables S5 and S6). Nevertheless, fungal beta diversity changes were less significant, with relatively compact clusters (Figure S2d–f). Their structural complexity and slower turnover may delay responses [56,59]. Nevertheless, ANOSIM/Adonis confirmed that fungal communities were also affected, though to a lesser degree (Figure S2g).

3.5. Soil Microorganism Differential Analysis

3.5.1. Inter-Group Microbial Differential Analysis

At the phylum level, Zixibacteria and Abditibacteriota exhibited significant differences in relative abundance across treatments (p < 0.05) (Figure 4a). The abundance of Zixibacteria significantly increased in MP-containing treatment groups (i.e., mPE, Cd-mPE) compared to the treatment groups without MPs (i.e., CK, Cd), while the abundance of Abditibacteriota exhibited slight enrichment, likely due to their ability to exploit MP degradation products as carbon sources [60,61]. These findings suggested that these microorganisms were key players in response to environmental changes.

Crucially, our analysis revealed a significant enrichment of Solirubrobacteraceae specifically in the Cd-mPE treatment group (Figure 4b). The enrichment of this family suggests that mPE-modified micro-environments provide specific niches for taxa capable of utilizing plastic-associated carbon or responding to improved aeration [62]. This microbial shift likely facilitates a “priming effect,” where the presence of mPE stimulates the metabolic turnover of indigenous soil organic carbon, as reflected in the dynamic changes in SOM observed in our combined treatments (Figure 2a). Mechanistically, Solirubrobacteraceae comprises known Cd-tolerant and stress-adapted strains that are frequently recruited in metal-polluted soils. Integrating this biological shift with the soil physical data, we infer that mPE-induced modifications to soil microstructure (e.g., pore architecture and aggregate stability) fostered a microenvironment conducive to their proliferation. Members of this family are capable of secreting extracellular polymeric substances (EPS), which can chelate, sequester, or redox-transform Cd ions [63,64]. This microbial activity directly modulates Cd speciation and bioavailability at the rhizosphere interface, ultimately explaining the enhanced Cd partitioning and its subsequent uptake by rice roots.

Regarding fungal communities, significant shifts were observed in Ascomycota and Mortierellomycota (Figure S3a,b). Detailed results and discussion were provided in the Supplementary Materials Text S6.

At the family level, ten taxa differed significantly, with Solirubrobacteraceae, Illumatobacteraceae, and Sporichthyaceae showing the most pronounced increases (Figure 4b). This variation may be attributed to mPE-induced modifications in soil physical architecture, such as changes in pore structure, aeration, and water-holding capacity, which create favorable niches for these adaptable bacteria [63,65]. For instance, Illumatobacteraceae are known to thrive in contaminated environments through specialized defense mechanisms like antibiotic synthesis [66,67].

3.5.2. LEfSe Multi-Stage Discriminant Analysis of Species Differences

LEfSe analysis further identified specific microbial signatures across treatments (Figure 4c). In CK soils, Zixibacteria were significantly enriched, but their decline in polluted treatments reflects the oxidative stress imposed by the co-occurrence of mPE and Cd [68]. Conversely, Cd contamination alone promoted metal-resistant groups such as Desulfobacca, which are known for detoxification mechanisms.

In Cd-mPE treatments, the further enrichment of Solirubrobacteraceae and Sporichthyaceae (confirmed by LDA scores, Figure S4) suggests that mPE may either exert a buffering effect on Cd toxicity or reorganize Cd bioavailability through adsorption–desorption dynamics [4,69]. Interestingly, the enrichment of sulfate-reducing bacteria (e.g., Desulfobacterales) in Cd-heavy treatments likely contributed to Cd immobilization through sulfide precipitation [70]. However, the shift toward Longimicrobiaceae and Aminicenantales in mPE-containing groups indicates enhanced metabolic activities that could facilitate Cd migration in the soil matrix. [71]. Collectively, these findings suggest that mPE and Cd drive a reorganization of the microbial network, promoting “pollution-adapted” taxa that play dual roles in Cd stabilization and plant stress response.

Furthermore, fungal LEfSe analysis showed significant shifts in Mortierellomycota and Ascomycota (Figure S3c,d). These alterations imply MPs and Cd jointly modify fungal community structure, potentially affecting nutrient cycling and plant–microbe interactions. Detailed results and discussion were provided in the Supplementary Materials Text S6.

3.6. SEM and Mantel Test Analysis

The three SEM quantified the pathways and intensities of Cd, MPs, and their combined treatments within the soil–rice system, with all models achieving satisfactory fit indices (GOF > 0.70; Figure 5a–c). In the Cd-only model (GOF = 0.8302), Cd loading was the primary driver of the system. It exerted a dominant influence on soil Cd availability (path coefficient = 0.9979 ***) and directly promoted Cd accumulation in rice tissues (path coefficient = 2.171 *). The resulting growth inhibition was primarily attributed to direct physiological damage from tissue Cd rather than enzyme-mediated oxidative stress, as the indirect pathway via soil enzymes was relatively weak. This suggests a linear toxicity pathway: soil Cd loading → tissue Cd uptake → direct growth impairment, consistent with observations in other cereals where Cd directly impairs photosynthesis and nutrient uptake [17].

In the MP-only model (GOF = 0.7001; Figure 5b), MPs acted through indirect and direct mechanisms. MPs strongly modified soil properties (pH, MC, OM) (path coefficient = 0.9276 ***), which strongly suppressed rice growth (path coefficient = 1.630 *). MPs also directly disturbed the rice enzymatic system, leading to further growth inhibition. Although MPs slightly enhanced Cd accumulation, this effect was weaker and less consistent. These results confirm that MP toxicity is largely mediated by the “mPE → soil alteration/enzyme disruption → growth inhibition” axis, echoing findings that microplastics disrupt soil aggregation and antioxidant balance [72,73].

In the combined Cd+MPs model (GOF = 0.7115; Figure 5c), both pollutants acted as core variables with interactive effects. Cd maintained its dominant role, driving soil Cd (path coefficient = 0.9883 ***) and tissue Cd accumulation (path coefficient = 1.111 *). MPs continued to affect soil properties (path coefficient = 0.8745 *), though less strongly than in the MP-only model, suggesting Cd moderated this pathway. Importantly, MPs significantly enhanced oxidative stress (path coefficient = −0.3845 *), making enzyme disturbance a key determinant of growth inhibition. Compared with Cd alone, MPs amplified indirect toxicity via oxidative stress, shifting the balance from purely direct Cd effects to mixed direct–indirect pathways. Crucially, this SEM further elucidates the mechanisms behind the observed synergistic risks. The model reveals that mPE-induced shifts in soil physicochemical properties mediate the enhanced translocation of Cd from the soil to rice roots and stems. This quantitatively accounts for the counterintuitive observation that even as total soil Cd decreased, tissue Cd concentrations significantly increased (Figure 1c). We propose that mPE amplifies Cd ecological risk through two synergistic mechanisms: (i) geochemical mobilization, where mPE-induced changes facilitate the transformation of Cd into more bioavailable exchangeable and carbonate-bound fractions; and (ii) physiological disruption, where mPE intensifies oxidative stress, making enzymatic disturbance a key determinant of growth inhibition.

Mantel test analysis exhibited Cd accumulation in crops strongly correlated with soil DTPA-extractable Cd (r = 0.70, p < 0.01) (Figure 5d), indicating that its bioavailability of Cd in soil directly determined crop uptake (r = 0.28, p < 0.05). Detailed results and discussion were provided in the Supplementary Materials Text S7. Unlike most studies focusing on physical and chemical properties, this study integrates soil enzyme activity and microbial diversity to propose a more comprehensive mechanism for Cd migration and crop growth regulation.

Overall, Cd exerted direct toxicity through accumulation in tissues, whereas MPs mainly acted indirectly by altering soil properties and enzyme activity. Under co-exposure, the interaction is not simply additive: MPs intensify Cd-induced oxidative stress, shifting the balance toward both direct and indirect toxicity pathways. Moreover, mPE and Cd act synergistically in paddy soils via a “Physical–Biochemical–Geochemical” cascade (Figure S5). Initially, mPE physically increases macroporosity and aeration, which triggers biochemical shifts by accelerating organic matter mineralization and raising soil pH. These alterations foster Cd-tolerant microbes (e.g., Solirubrobacteraceae), which secrete EPS and mobilize Cd from stable pools into bioavailable forms. Ultimately, this geochemical mobilization, coupled with mPE-induced oxidative stress, drives elevated Cd accumulation in rice tissues, confirming mPE’s role as a potent risk amplifier in the soil–rice system.

4. Conclusions

This study demonstrated that co-exposure to mPE and Cd intensifies phytotoxicity in rice by enhancing Cd bioavailability through altered soil physicochemical properties and microbial-mediated transformations. Elevated Cd accumulation in rice tissues was accompanied by bacterial community shifts toward metal-resistant taxa (e.g., Solirubrobacteraceae, Desulfobacterales), while fungal responses were weaker. Soil enzyme activities (e.g., CAT, S-UE) contributed to both Cd detoxification and nutrient cycling, with microbial diversity supporting Cd immobilization. DTPA-extractable Cd and OM were identified as key regulators of Cd mobility. Cd exerted direct toxic effects, whereas MPs acted mainly through indirect pathways by modifying soil properties and inducing oxidative stress. Under co-contamination, MPs amplified Cd-induced oxidative stress, increasing the relevance of indirect toxicity pathways. These results highlight the need to integrate MPs into heavy metal risk assessments, particularly in paddy ecosystems.

5. Future Research Directions

Building upon the insights gained from this study, future research should move toward more granular mechanistic elucidation and real-world risk management. Future research should prioritize three granular directions: (1) Molecular mechanisms: Employ advanced spectroscopy to identify Cd-MP binding configurations and investigate how long-term environmental weathering alters MP surface chemistry and Cd desorption. (2) Rhizosphere-mediated regulation: Use multi-omics and root exudate profiling to link microbial shifts (e.g., Cd-tolerant taxa) and plant gene expression to Cd redistribution. (3) Field validation and risk assessment: Conduct multi-season trials to verify laboratory findings and integrate dietary exposure modeling for grain Cd accumulation.

Collectively, these priorities will strengthen predictive capacity and management strategies for MP–heavy metal co-contamination in agroecosystems.