Iron–Zinc Synergy Blocks Cadmium Translocation in Rice: Minimizing Grain Contamination

Abstract

Cadmium (Cd) pollution severely constrains safe rice production and threatens food security. Leveraging Fe–Zn competitive antagonism to mitigate crop Cd accumulation is a green, sustainable remediation strategy. Based on our hypothesis, we proposed that combined Fe–Zn pretreatment in seedlings and foliar spraying during the reproductive period would reduce Cd accumulation in brown rice by inhibiting root uptake, impeding translocation, and enhancing vacuolar sequestration in flag leaves. A two-year, three-season field experiment was conducted in the Cd-contaminated double-cropping rice planting area in Hunan Province. Three treatments were applied: conventional (CK), Fe–Zn pretreatment at seedling stage (FZ), and Fe–Zn pretreatment + tillering and heading spraying (FZS). This study demonstrated that FZS reduced brown rice Cd by 25%, primarily by enhancing root retention (root Cd reduced by 17–19%) and flag leaf vacuolar sequestration (flag leaf Cd 31% higher than old leaves). FZS further decreased stem–leaf Cd by 47–54% and lowered the husk-to-grain transfer coefficient from 0.22 to 0.17. Multivariate analysis identified flag leaf interception (β = −0.25) as the dominant factor regulating grain Cd, followed by panicle accumulation (β = 0.122) and Fe–Zn dosage (β = −0.061). Integrated Fe–Zn treatment blocked soil-to-grain Cd transfer via physiological barriers and flag leaf sequestration.

1. Introduction

Cadmium (Cd) contamination in agricultural soils remains a persistent challenge, critically compromising crop safety and productivity. Rice (Oryza sativa L.), a staple crop of global importance, exhibits strong Cd accumulation capacity. This process initiates with root uptake, followed by translocation to shoots and progressive accumulation in grains through the stem–leaf–grain pathway [1]. As a primary staple crop across the Asia-Pacific, over 40% of dietary Cd exposure originates from rice consumption in Southeast Asian populations [2,3]. Current mainstream remediation strategies for Cd-contaminated farmland fall into two categories: immobilization remediation utilizing silicon/calcium/magnesium-based passivation materials to reduce Cd bioavailability [4,5,6] and phytoextraction remediation employing hyperaccumulators (e.g., Sedum plumbizincicola, Euphorbia chevalieri, and Solanum nigrum) to decrease soil Cd burden [7,8,9]. However, the former may cause secondary pollution [10], and the latter makes it difficult to ensure the safe production of crops at the same time, so both have certain limitations. Consequently, eco-compatible remediation strategies are increasingly pivotal for safeguarding safe and sustainable crop production in Cd-contaminated farmland systems.

Cd is a nonessential element for plants that hijacks essential metal ion transporters to enter plant systems [11]. In rice, Cd accumulation proceeds through three sequential phases: root uptake involving activation and absorption in the rhizosphere; xylem loading and translocation to aerial tissues; phloem-mediated remobilization to developing grains [12]. Yan’s research indicates that excessive Cd accumulation in seeds is uncoupled from both root uptake during seedling stage and root-to-shoot translocation efficiency but depends critically on phloem-mediated Cd transport to seeds [13]. While some studies have reported that phloem transport contributes >90% of the seed Cd content [14], others have proposed that the regulatory mechanism lies primarily in the root system’s Cd absorption capacity rather than aerial tissues [15]. Nutrient elements iron and zinc are critical for crop growth and development, and their regulated supply is a key measure to achieve high and high-quality rice yields, while the competitive and antagonistic effects among Fe, Zn, and Cd during migration and accumulation have been extensively studied by scholars [16,17]. Studies have shown that Fe participates in chlorophyll synthesis and electron transfer in photosystem II, thereby increasing the leaf photosynthetic rate, enhancing pollen vitality, and reducing empty and shriveled grains [18], while Zn, as a component of carbonic anhydrase, promotes CO₂ fixation, participates in auxin synthesis (regulating panicle differentiation and floret development), and increases the seed setting rate [19].

Due to shared ion transporters, Cd engages in antagonistic competition with essential metals for uptake pathways. Research confirms OsIRT1/OsIRT2 exhibit high Cd affinity, and under iron deficiency conditions, upregulated IRT1 expression enhances Cd accumulation, as evidenced by OsIrt1 overexpression studies [20,21]. Concurrently, ferrous ions (Fe²⁺) competitively inhibit Cd uptake through membrane transporters, reducing root Cd absorption [22]. Cd accumulation further upregulates calcium and iron transport channels [23], while calcium deficiency suppresses plant Cd accumulation [24]. Guan et al. demonstrated that iron/zinc pretreatment at the seedling stage effectively mitigates Cd uptake: zinc primarily inhibits root Cd absorption, whereas iron blocks root-to-shoot translocation [25]. Mechanistically, elevated iron concentrations promote iron plaque formation on rice roots, enhancing Cd fixation by 18% and significantly reducing the aerial tissue Cd content [26]. A nutrient-mediated Cd blockade, particularly the synergistic antagonism of iron and zinc, provides an effective phytomanagement strategy to mitigate Cd accumulation in rice ecosystems. However, the translocation dynamics of Cd in rice under Fe/Zn pretreatment and its consequent Cd inhibition in brown rice require comprehensive quantification. In particular, there is a lack of research on the regulatory mechanism of Fe/Zn pretreatment on the key nodes of panicle–husk–grain. Building on our prior systematic screening of iron (0–200 μM), zinc (0–1.5 μM) and their binary combinations, the optimized formulation of 100 μM Fe + 0.75 μM Zn demonstrated superior Cd reduction efficacy [25]. This study therefore implements a comprehensive two-year, three-season field validation trial in the typical double-cropping rice growing areas of the south to study the synergistic interaction mechanism of iron and zinc in the process of antagonizing Cd accumulation in brown rice. We tracked Cd migration and translocation across rice organs during key growth stages (tillering, heading, pustulation, and maturity), leveraging multivariate regression modeling to conduct an integrated risk appraisal of grain contamination from organ-specific Cd seeds of rice (Oryza stativa L. Luliangyou 996, early season and Yuzhenxiang, late season). These findings revealed an eco-sustainable approach to reduce Cd accumulation in rice via Fe/Zn nutrition regulation, providing a theoretical basis for safeguarding rice production in Cd-contaminated farmlands.

2. Materials and Methods

2.1. Soil Characterization

A two-year field experiment was conducted at the rice paddies (28°23′52″ N, 113°1′4″ E) in Beishan Town, Changsha City, during 2018 (two seasons of rice) and 2019 (late rice). The soil at the site was a typical loamy paddy soil, with a soil texture of 50.4% coarse sand, 5.6% fine sand, 26% silt, and 18% clay. The physicochemical properties and heavy metals background value of the site soil is as follows: a pH value of 5.8, 2.4% organic carbon, cation exchange capacity of 12.2 cmol/kg, 3.1 mg/kg Olsen P, 148 mg/kg available K, and 79.5 mg/kg hydrolysable N. The concentrations of total Cd, arsenic (As), and plumbum (Pb) in the topsoil (10–20 cm depth) were 0.993 mg/kg, 17.4 mg/kg, and 77.5 mg/kg; the available of Fe, Zn, manganese (Mn), and copper (Cu) in site soil were 63.7 mg/kg, 2.51 mg/kg, 5.0 mg/kg, and 2.32 mg/kg; and the available Cd, As, and Pb were 0.53 mg/kg, 0.129 mg/kg, and 0.0137 mg/kg, respectively.

The nutrient elements of the experiment site were low levels, and the available iron was lower than the average of the middle reaches of the Yangtze River [27], while the Cd in the soil was defined as moderately polluted [28]. Total heavy metal (Cd, As, and Pb) in the soil was obtained through digestion with nitric acid, hydrofluoric acid, hydrochloric acid, and hydrogen peroxide in a closed vessel microwave system [29] and quantified by inductively coupled plasma-mass spectrometry (ICP-MS). As a typical double-cropping rice producing area in Southern China, the climate of the site is subtropical monsoon with a mean annual precipitation of 1539 mm and an average daily temperature of 17.2 °C.

2.2. Chemical Reagents and Seed Germination

The rice seedling nutrient solution reagent contains NPK major elements and medium and microelement elements, such as calcium (Ca), magnesium (Mg), Fe, Zn, etc., which were guaranteed reagent (GR) grade and were purchased from Sinopharm Chemical Reagent Co., Ltd.^® (Beijing, China). Water was purified with a model 7155 Barnstead Nanopure system (Thermo Scientific, Boston, MA, USA) and subsequently filtered through a 0.45 μm membrane. After the initial screening, rice seeds were subsequently grown in a nutrient solution enrichment cultivation for a month or so, and the seedlings rich in Fe and Zn were transplanted to the field site [30]. All seeds were provided by the Hunan Academy of Agriculture Sciences. Rice seeds were germinated as previously described [25,31] and subsequently grown in a rice nutrient solution containing the following compositions (µM): (NH₄)₂SO₄, 180 µM; KNO₃, 70; KH₂PO₄, 90; MgSO₄.7H₂O, 270; Ca(NO₃)₂.4H₂O), 180; NaEDTA-Fe·3H₂O, 20; MnCl₂·4H₂O, 6.7; CuSO₄·5H₂O, 0.16; ZnSO₄·7H₂O, 0.15. The nutrient solution was adjusted to a pH of 5.5 with NaOH and HCl and changed every 5 days. The stock solution was concentrated by 10 and kept at 4 °C until use.

2.3. Pretreatment Iron and Zinc Absorption Experiments

All pretreatment iron–zinc experiments under laboratory conditions were performed as follows: budding rice seedlings transferred to a quadrate black plastic container (size: 20 × 30 × 10 cm) and grown in 1/4 strength nutrient solution with each treatment for 7 days, then for a further 26 days in 1/2 strength solution. The average temperature throughout the test period was between 28 °C (daytime) and 20 °C (night), and the relative humidity was 70 ± 5%. Metal levels in rice plants grown in the full nutrient were considered as the base and were used as controls. The plants were irrigated with distilled water as needed, and the nutrient solution was replaced every five days. Each experimental condition was carried out at least in triplicate to make sufficient seedlings in the field. Based on the results of our previous research, Fe and Zn were found to have relatively high inhibition in Cd adsorption in both shoots and roots and, thus, were selected for the field experiments. These were carried out as follows: CK (base full nutrient solution, CK), FZ (CK + Fe 100 µM and Zn 0.75 µM pretreatment at the seedling stage), and FZS (FZ + equal concentration of Fe and Zn nutrient solution was sprayed at the tillering stage and heading period). In short, the complete nutrient solution was supplemented with additional Fe and Zn at five times the standard concentration. Enrichment cultivation lasted a month or so, and then, the seedlings rich in Fe and Zn were transplanted to the field site.

2.4. Field Experiments

Field experiments were arranged in a split–split plot design with different rice treatments as the main plot. Two varieties with high Cd accumulation: Luliangyou 996, early season and Yuzhenxiang, late season were selected for study. Pre-geminated seeds were transferred into an artificial climate culture chamber at the sowing date of 2 April (early season) and 15 June (late season) in 2018 and 12 July in 2019. Thirty-day-old enrichment seedlings were transplanted on 2 May and 16 July 2018, and thirty-day-old pretreated seedlings were transplanted on 13 July 2019, with three seedlings per hill. The test plot size was 25 m² in both years, compound fertilizer was banded at a soil depth of 10 cm prior to planting three replications for each treatment, and bunds were built and covered with plastic film to minimize seepage between plots. Three treatments were set up in the experiment: seedlings cultured in conventional nutrient solution for 30 days were labeled as CK; CK + Fe 100 µM + Zn 0.7 µM in seedlings were labeled as FZ; exogenous Fe/Zn (Fe 100 µM + Zn 0.7 µM) sprayed onto a part of FZ at the rice tillering and heading stages were marked as FZS (the iron–zinc mixed reagent was applied at a dosage of 220 L per hectare, with a single spray at both the tillering and heading stages of rice growth, and the pH value of the sprayed solution should be maintained between 5.5 and 6.0), and three replications for each treatment were set.

In both years, 600 kg/ha of compound fertilizer (15–15–15, N–P–K) was banded at a soil depth of 10 cm prior to planting. Then, additional applications of urea (135 kg/ha) and potassium salts (90 kg/ha) were applied after 20 days. The fertilizer was a blend of monoammonium phosphate, urea, and muriate of potash, which provided a total of 152, 90, and 144 kg/ha of N, P₂O₅, and K₂O, respectively. Water regime of the test plot was as follows: flooding, mid-season drainage, re-flooding, and moist intermittent irrigation in a sequence. Weeds, pests, and diseases were intensively controlled by chemicals at the same time.

2.5. Sampling and Measurements

The tissues and organs of rice were sampled at typical growth stages (early season in 2018: tillering stage, heading period, pustulation period, and matured period; late seasons in 2018: tillering stage, booting stage, and matured period; late seasons in 2019: sampled at the matured period) and rinsed with deionized water more than three times, until the soil and dust on the surface of the plant were washed away. Then, the samples were kept at 105° for an hour and dried at 80° until a constant weight. The acquired samples were divided into brown rice; husk; spike branch; stem leaf (including flag leaf (FL), second leaf (SL), third leaf (TL), and old leaf (OL)); and root system tissue and organs (leaves at maturity only analyzed as flag and old parts). All plant tissues were ground into flour with plant tissue grinders for further monitoring of Cd and nutrient element contents. The yield of each treatment was recorded at harvest.

Samples of 0.3 g (two analytical replicates per plot) were digested in a microwave digestion tube by nitric acid (the nitric acid was used after being purified), and the mixture of the sample and the acid was dissolved overnight at room temperature. Then, a small amount of hydrogen peroxide was added before being put into the microwave digestion instrument (Mars 6, CEM, Charlotte, NC, USA). The heating procedure of the instrument was as follows: increased from room temperature to 120° and maintained for 5 min, then increased to 150° and maintained for 10 min, and finally, maintained at 185° for 30 min. Digestion liquid was filtered using a syringe filter (0.45 μm), then diluted to 50 mL with 18.2 MΩ cm water (Barnstead Nanopure, Thermo Scientific, Boston, MA, USA). The Cd and nutrient elements in the samples were quantified by inductively coupled plasma-mass spectrometry (ICP-MS, i-Cap Q, Thermo Fisher Scientific, USA), with rhodium (¹⁰³Rh) as the internal standard. The detection limit for the instrument was 0.02 μg/L. For quality control, GSW10049 was used as the Cd standard reference material, and the method detection limit was 0.002 mg/kg.

2.6. Statistical Analysis

A one-way analysis of variance (ANOVA), Duncan’s multiple, and Pearson’s correlation analysis were conducted to assess the significance of differences in various parameters across treatments. The experimental data were statistically analyzed using Origin Pro 2017, SPSS 20, and R 4.0.4 software. SPSS 20 was used to analyze the significance and correlation of the data, while Origin Pro 2017 and R software were employed to create the corresponding graphs, and R software was utilized for the calculation of the multiple regression model. All data are presented as means ± standard error (SE, n = 3).

3. Results

3.1. Rice Yields

Exogenous iron and zinc pretreatment combined with foliar spraying of iron and zinc significantly increased rice yield (Figure 1). Data from double-cropping rice yield in 2018 showed that, compared with the untreated control (CK, 6.53 t/ha), the FZ treatment increased rice yield by 3.1% for early rice. For late rice, the increase was 2.6%, with yield reaching 8.34 t/ha. Under the FZS treatment, rice yield was further improved: the increase for early rice season was 6.1%, and for late rice season, it was 4.6% (p < 0.05). Additionally, in 2019, there was a 5% increase in the late rice season.

3.2. Cd Concentrations in Brown Rice over Three Growing Seasons

The accumulation of Cd in rice is shown in Figure 2. Data from the 2018 double-cropping rice yield indicated that the Cd content in early rice treated with FZ was 8.9% lower than in the control (CK), dropping to 2.56 mg/kg, and in late rice, decreased by 10% under FZS treatment (1.25 mg/kg). Under FZS treatment, it was reduced: in the 2018 early rice season, it decreased by 18% compared to CK (p < 0.05), and in late rice, the Cd content decreased by 25% (p < 0.05), reaching 1.05 mg/kg. In 2019, the Cd content in late rice brown rice showed a 7.2% reduction.

3.3. Cd Accumulation and Distribution in Root–Stem–Panicle Across Growth Stages

Cd accumulation revealed growth stages: root > stem ≥ panicle > husk. Specifically in early rice (Figure 3a), FZ treatment significantly reduced Cd accumulation in roots and stems during tillering versus CK, with reductions of 19% (8.3 → 6.7 mg/kg) and 54% (2.38 → 1.09 mg/kg), respectively (p < 0.05), confirming iron–zinc pretreatment efficacy. At the heading stage, root Cd showed no significant difference between FZ/FZS and CK; however, compared to the untreated control (CK), stem and panicle Cd decreased by 15% under FZ (reaching 6.1 mg/kg and 6.19 mg/kg) and 22% under FZS (reaching 5.65 mg/kg and 5.72 mg/kg; p < 0.05). FZS treatment (combined Fe/Zn pretreatment + foliar spray) further boosted Cd reduction. During grain filling, FZ reduced Cd in stems, leaves, and panicles by 15–18% versus CK (p < 0.05), and FZS reduced panicle Cd by 15% (p < 0.05). At maturity, FZ decreased husk Cd by 27% (p < 0.05), with declining trends in the roots and panicle branches; FZS reduced Cd across all tissues (roots/stems/panicle branches/husks) by 18–32% versus CK, exceeding 27% reduction in stems and husks (p < 0.05).

Late rice Cd translocation patterns (Figure 3b) parallel those in early rice tissues. Under FZ treatment at tillering, Cd accumulation in rice roots and stems decreased by 17% and 47%, respectively, versus CK (p < 0.05). During tillering, iron–zinc pretreatment significantly inhibited Cd accumulation. At the booting stage, FZ reduced stem Cd by 18% (p < 0.05), and FZS achieved 70% and 21% reductions in roots and stems, respectively (p < 0.05), demonstrating that tillering stage top dressing with iron/zinc further suppresses Cd accumulation. Post-maturity, both FZ and FZS primarily inhibited Cd in aerial tissues: FZ decreased stem and husk Cd by 12% and 15% (p < 0.05), whereas FZS reduced them by 20% and 26% (p < 0.05).

3.4. Cd Dynamics in Rice Leaves Across Growth Stages

Cd accumulation varied across leaf positions in rice plants. As shown in Figure 4a, during tillering (early rice), FZ treatment substantially diminished the Cd levels across all leaves, with second leaf (SL) and third leaf (TL) falling by 59% (p < 0.01) and (old leaf) OL diminishing by 27%. At the heading stage, FZS treatment lowered flag leaf (FL) Cd to 0.45 mg/kg (a 21% reduction, p < 0.05) and cut OL by 12%. Throughout grain filling, FZ decreased SL and TL by 21% and 24% (p < 0.05), respectively, whereas FZS diminished them by 10% and 29% (p < 0.05), respectively. At maturity under FZ, flag leaf Cd declined by 11% (from 1.50 to 1.33 mg/kg) and OL diminished a further 21% (to 0.90 mg/kg); under FZS, flag leaf dropped by 13% (to 1.31 mg/kg), and OL fell by 33% (to 0.77 mg/kg, p < 0.05). Cd distribution in rice leaves displayed clear trends across the developmental phases. The flag leaf consistently exhibited peak Cd concentrations (mean: 1.09 mg/kg), escalating to its zenith during grain filling (1.56 mg/kg). During tillering, accumulation progressively diminished from the FL (highest) to OL (lowest). At heading, Cd in OL and TL remained marginally below the FL, while the SL recorded minimal values. Throughout grain filling, the second and third leaves sustained the lowest concentrations (0.71–0.76 mg/kg). By maturity, flag leaf Cd markedly exceeded OL by >31% (p < 0.05), confirming its dominant accumulation role.

Late rice harvest results (Figure 4b) revealed substantial Cd decrease under treatments relative to CK. During tillering, FZ curtailed Cd accumulation across all leaves by 41–69% (p < 0.05). At the booting period, FZ diminished FL and SL Cd by 15%, whereas FZS provoked further declines, depressing TL and OL by 81% and 62%, respectively (p < 0.05). At full maturity, FZ and FZS reduced leaf Cd to 0.608 mg/kg (a 34% reduction) and 0.651 mg/kg (29% lower) compared with CK (0.922 mg/kg, p < 0.05). Accumulation hierarchies revealed that, during tillering, SL peaked markedly, exceeding both flag leaf and TL. From jointing to heading, flag leaves sustained maximal concentrations, followed by SL, while OL and TL consistently exhibited minimal levels (lower than newer leaves).

3.5. Cd Transport Coefficients in Rice Organ Systems Throughout the Growth Cycle

Statistical analysis (

Table 1. The Cd transfer coefficient of each tissue and organ under different treatments throughout the growth cycle of rice. ((a): early season rice; (b): late season rice).

(a)
Growth Stages		Root/Soil	Stem/Root	OL/Stem	Panicle (Cladus)/Stem		TL/OL	SL/TL		FL/SL	Husk/Cladus		Brown Rice/Husk
Tillering stage	CK	8.4	0.28	0.078	/		1.5	3.2		1.1	/		/
	FZ	5.3	0.35	0.074	/		0.81	3.2		2.4	/		/
Heading stage	CK	17.9	0.40	0.063	1.0		0.93	0.76		1.8	/		/
	FZ	19.4	0.31	0.097	1.0		0.92	0.73		1.5	/		/
	FZS	18.4	0.30	0.071	1.0		1.2	0.82		1.2	/		/
Pustulation stage	CK	37.4	0.44	0.069	0.51		0.82	0.86		2.1	/		/
	FZ	34.5	0.40	0.074	0.49		0.68	0.89		2.5	/		/
	FZS	39.9	0.42	0.061	0.42		0.64	1.1		2.1	/		/
Maturity	CK	35.0	0.37	0.088	1.2		/	/		/	0.22		0.76
	FZ	32.0	0.47	0.060	0.99		/	/		/	0.17		0.95
	FZS	28.7	0.33	0.081	1.3		/	/		/	0.19		0.91
(b)
Growth Stages			Root/Soil	Stem/Root	OL/Stem	TL/OL	SL/TL		FL/SL			Husk/Cladus		Brown Rice/Husk
Tillering stage	CK		3.3	0.40	0.51	0.78	1.5		0.61			/		/
	FZ		2.7	0.21	0.57	0.41	2.8		0.36			/		/
Booting stage	CK		6.2	1.2	0.13	0.64	3.6		0.95			/		/
	FZ		7.1	0.95	0.16	0.58	3.3		0.96			/		/
	FZS		1.8	0.91	0.064	0.32	11.7		1.2			/		/
Maturity	CK		13.5	5.4	0.17	/	/		/			0.42		0.61
	FZ		13.2	4.8	0.12	/	/		/			0.40		0.64
	FZS		14.0	4.4	0.15	/	/		/			0.38		0.61

Note: All ratios were calculated based on the average values. “/” indicates “not applicable”.

) revealed that FZ treatment during early rice tillering significantly inhibited Cd translocation, particularly between the soil–root system and older leaves/third leaves (OL/TL). Compared to CK (TF _[Root/Soil] = 8.36; TF _[OL/TL] = 1.45), FZ reduced transport coefficients to 5.32 (a 36% decrease) and 0.81 (a 44% decline), demonstrating effective suppression of inter-organ Cd flux. Upon rice entering heading, FZ treatment inhibited Cd translocation from roots to stems and from second leaf to flag leaf, reducing TF _[Stem/Root] by 21% and TF _[FL/SL] by 17%. Similarly, FZS treatment decreased TF _[Stem/Root] by 23% and TF _[FL/SL] by 32%. At grain filling, Fe/Zn treatments inhibited Cd translocation along the root–stem–panicle continuum and from OL to TL. Specifically, FZ curtailed TF _[Stem/Root] by 8%, TF _{[Panicle/Stem]} by 3%, and TF _[TL/OL] by 17%. In FZS treatment, reductions in TF _{[Panicle/Stem]} and TF _[TL/OL] were more pronounced, declining by 17% and 21%, respectively. At maturity, Fe/Zn treatments targeted the stem–cladus–husk pathway to block Cd translocation in rice. Under FZ treatment, TF _{[Cladus/Stem]} and TF _{[Husk/Cladus]} both decreased by 20% versus CK. FZS treatment curtailed TF _[Root/Soil] by 18% and TF _{[Husk/Cladus]} by 13% compared to CK. Similarly, iron and zinc treatments effectively impeded Cd translocation across three critical pathways during late rice tillering: soil-to-root transport (Table 1b), root-to-stem translocation, and flux from OL to newer leaves. The TF _[Root/Soil] coefficient decreased by 17%, and the TF _[Stem/Root] and TF _[TL/OL] decreased by 46%. During the booting stage, under the FZ treatment, the TF _[Stem/Root] and TF _[TL/OL] decreased by 17% and 10%, respectively, and under the FZS treatment, they decreased by 21% and 49%, respectively. At maturity, Fe/Zn treatments primarily blocked Cd transport via the root–stem–husk pathway. Under FZ, TF _[Stem/Root] declined 12%, TF _[OL/Stem] 24%, and TF _[Husk/Stem] 3%; with FZS, these values decreased by 19%, 12%, and 8%, respectively. Collectively, these significant reductions in transport factors delineate key pathways for Fe/Zn-mediated suppression of Cd accumulation in rice.

3.6. Analysis of Potential Correlations in Cd Accumulation over Rice Developmental Phases

Correlation analyses revealed stage-specific patterns of Cd accumulation across rice organs. For early rice (Figure 5a and Figure S1a), at the tillering stage, root Cd showed significant positive correlations with stem Cd (R = 0.81, p < 0.01) and SL Cd (R = 0.84, p < 0.01), while TL Cd correlated with both SL (R = 0.84, p < 0.01) and OL Cd (R = 0.74, p < 0.05). At heading, panicle Cd exhibited significant correlations with root Cd (R = 0.70, p < 0.05) and stem Cd (R = 0.78, p < 0.05). During grain filling, stem Cd demonstrated significant associations with FL Cd (R = 0.77, p < 0.05), SL Cd (R = 0.82, p < 0.01), and panicle Cd (R = 0.74, p < 0.05). By maturity, grain Cd displayed strong positive correlations with OL Cd (R = 0.81, p < 0.01) and husk Cd (R = 0.87, p < 0.01), with additional significant linkage between OL and husk Cd (R = 0.75, p < 0.05). During late rice tillering, aboveground Cd correlations were highly significant (Figure 5b and Figure S1b): FL Cd strongly correlated with TL Cd (R = 0.97, p < 0.001) and OL Cd (R = 0.98, p < 0.001), while stem Cd showed strong correlations to both TL (R = 0.91, p < 0.001) and OL Cd (R = 0.91, p < 0.001), with TL and OL Cd also exhibiting robust correlation (R = 0.97, p < 0.001). At booting, TL Cd correlated significantly with root Cd (R = 0.92, p < 0.001) and OL Cd (R = 0.74, p < 0.05), and SL Cd demonstrated significant associations with stem Cd (R = 0.80, p < 0.05) and FL Cd (R = 0.77, p < 0.05). Upon maturity, grain Cd was positively correlated with husk (R = 0.84, p < 0.01) and stem Cd (R = 0.79, p < 0.05), while husk and stem Cd were significantly correlated (R = 0.94, p < 0.001).

3.7. Predicting Brown Rice Cd Contamination Risk Across Life Cycle via Multivariate Regression

A multiple regression model comprehensively analyzed parameters including Cd accumulation at heading; grain filling; maturity stages of early rice organs (stems, leaves, panicles, roots, and grains); days after sowing (times); and Fe/Zn inputs. Statistical optimization mitigated severe multicollinearity and overfitting risks, yielding a robust linear model correlating brown rice Cd accumulation at maturity with key drivers (

Table 2. Multiple regression models to estimate Cd accumulation in brown rice and panicle.

Type of Organ	Multiple Regression Models	R Square	P
Brown rice	Cd brown rice = 0.276 × OL + 0.122 × P + 0.0088 × T − 0.254 × FL − 0.061 × FZ − 0.016 × Root + 1.801	0.774	1.44 × 10−5
Panicle	Cd panicle = 0.393 × Stem − 0.073T	0.673	1.49 × 10−6

Note: OL (Cd _{old leaf}); P (Cd _panicles); T (growth dates); FL (Cd _{flag leaf}); FZ (iron and zinc inputs); Stem (Cd _stem); Root (Cd _root).

). For brown rice Cd, multiple regression models demonstrated a strong predictive capacity (R² = 0.774, p < 0.001). All models exhibited variance inflation factors (VIF) below 5.3, confirming negligible multicollinearity among these factors. Specifically, Cd concentrations in old leaves (OL), panicles, flag leaves (FL), and root systems, combined with growth dates and Fe/Zn inputs, collectively explained 77% of the variation in brown rice Cd accumulation. The regression model demonstrated significantly positive coefficients for panicles (p < 0.001) and growth dates (p < 0.01) while showing negative coefficients for flag leaf Cd and Fe/Zn inputs (p < 0.001). Notably, stem Cd was excluded from the final model, as its regression coefficient showed no significance (95% CI: −0.00398 to 0.00296). Additionally, the model excluded other foliar parameters due to multicollinearity. Given the significant correlation between panicle and brown rice Cd accumulation, we quantified the Cd accumulation in rice panicles and other organs. The analysis revealed that stem Cd accumulation and growth dates collectively explained 67% of the variation in panicle Cd accumulation (R² = 0.673, p < 0.001; Table 2), with all variance inflation factors (VIF) < 2.0. Critically, stem Cd accumulation demonstrated a significant positive direct effect on panicle Cd levels (p < 0.001) and exhibited a positive indirect effect on brown rice Cd accumulation through its impact on panicle Cd loading. Figure 6 predicted the accumulation of Cd in brown rice and rice panicles derived from the multiple regression model. As expected, the majority of observed values fell within the prediction interval, confirming adequate predictive accuracy of the linear regression model.

4. Discussion

It is well established that the synergistic effect of Fe and Zn can enhance crop dry matter accumulation, and their reasonable supplementation increases rice yield. Consistent with previous findings [16,17,18,19], the results of this study showed that FZ treatment increased rice yield by 3.1%, while FZS treatment increased yield per mu by 6.1% (Figure 1). Iron, an essential micronutrient for rice, exhibits a well-recognized inhibitory effect on Cd accumulation, representing a pivotal advance in heavy metal pollution control in recent years. Its mechanism hinges on the synergistic interplay between competitive ion uptake and enhanced root surface barrier function. As divalent metal ions, Fe²⁺ and Cd²⁺ directly compete for root cell surface transporters (e.g., OsIrt1), with competition intensity increasing as the Fe²⁺ concentration rises [32,33]. In this study, pretreatment with iron and zinc at the seedling stage increased rhizosphere Fe²⁺ availability, reducing root Cd accumulation by 19% at the tillering stage (vigorous vegetative growth; Table 1 and Figure 3), confirming ionic competition as the dominant mechanism for early Cd uptake control. Unlike the tillering stage, root total Cd accumulation showed no significant difference in the control (CK) from heading to maturity. This phenomenon was closely linked to dynamic iron plaque formation on root surfaces: under Fe-rich conditions, Fe²⁺ oxidized to form iron hydroxide-dominated plaques, which high specific surface area and surface charge enabled the strong adsorption of soil-free Cd²⁺, immobilizing it in situ and blocking transmembrane transport [34]. This barrier effect gradually replaced ionic competition in later growth stages, becoming the key defense against Cd migration to aboveground parts. Additionally, Fe–Zn co-treatment enhanced nutrient uptake by improving root morphology (e.g., increased length and surface area) [35], indirectly strengthening the Cd control.

It is worth noting that Fe and Zn play a non-negligible role in regulating Cd redistribution in the aboveground tissues of rice. After Cd is absorbed by roots and translocated to the shoots, it is transported to stems and leaves via the xylem (driven by transpiration) and subsequently redistributed across organs through stem nodes [36]. Our findings revealed that Fe and Zn treatments significantly reduced Cd translocation from old leaves to the third leaf during the tillering, jointing, and filling stages. Conversely, the Cd transfer coefficient from the second leaf to the flag leaf increased by over 50% (Table 1). Some studies indicated that old leaves serve as major Cd reservoirs in rice, sequestering Cd into vacuoles through vacuolar compartmentalization—mediated by transporters such as OsVIT1 and OsVIT2—thereby limiting its mobilization to new leaves [37,38]. Additionally, old leaves near the roots exhibit higher activities of non-protein thiols (NPTs) and glutathione S-transferases (GSTs), which chelate Cd into nontoxic complexes and reduce its mobility [39]. During the filling stage, the flag leaf modulates grain Cd accumulation through a “dual role”: (1) Transport: As the main functional leaves and the output center of photosynthetic products, the flag leaf mediates Cd–sucrose co-transport via OsSUT1 in the phloem [40,41]. (2) Retention: Concurrently, its highly active antioxidant system (CAT) scavenges reactive oxygen species (ROS) to maintain cellular integrity, while sMTP11 transporter encoded by the OsCS1 is expressed in the phloem of leaves, reducing the Cd loading efficiency in the phloem and decreasing Cd translocation from leaves to grains [42]. The dynamic balance of transport and retention identifies the flag leaf as a regulatory hub with dual “source–sink” properties, where synergistic enhancement of the antioxidant capacity and thiol synthesis may serve as the core breakthrough for reducing Cd migration to grains, reinforcing the barrier value of compartmentalized stress resistance in old and flag leaves. Our results showed that Fe and Zn treatments reduced Cd translocation from roots to stems by enhancing Cd immobilization in roots. For aboveground tissues, Fe/Zn intervention further inhibited Cd transfer efficiency from stems/rachises to husks in both early and late rice at maturity, thereby blocking Cd entry into grains. Specifically, during the tillering stage, it strengthens the root barrier. Through ion competition, it reduces Cd accumulation in roots from the environment, restricts Cd migration from old leaves to new leaves, and blocks the Cd absorption and transport pathway from soil to roots to stems and leaves. During the heading and filling stage, it mainly inhibits the migration of Cd between old and new leaves and from stems to panicle branches to husks. At the same time, it promotes the compartmentalization of Cd in the main functional leaves (flag leaves). After rice enters the reproductive growth stage, the Cd accumulation ability of the flag leaves gradually exceeds that of the old leaves. This indicates that, during the reproductive growth process, the transport of nutrients to the fruit part may also drive the upward migration of Cd. Therefore, it is particularly important to spray Fe and Zn preparations during the critical period of reproductive growth (such as the heading stage) to block the transport of Cd from the flag leaves to the grains.

A multivariate regression model reveals the core regulatory network influencing grain. Cd accumulation is jointly constituted by Cd partitioning patterns across different organs and by the Fe–Zn intervention. As the “final station” for Cd translocation to brown rice, panicle Cd accumulation directly influences the rice Cd content (contribution factor β = 0.122, p = 6.2 × 10⁻⁵), indicating that blocking intercellular Cd transport from the panicle axis to the hull is a critical pollution control node. Cd transporters (e.g., OsHMA2 and OsLCT1) in panicle vascular tissues may serve as molecular targets [42]. Fe–Zn treatment exhibits a contribution factor of −0.061 (p = 8.38 × 10⁻⁴), suggesting its potential role in inhibiting these proteins through ion competition and suppression of transporter activity. The significant negative correlation between the flag leaf Cd content and brown rice accumulation (β = −0.25, p < 0.05) challenges the traditional view that leaf Cd accumulation exacerbates brown rice pollution. The dynamic allocation of Cd in flag leaves is hypothesized to result from the synergistic action of multiple transport systems. On the one hand, OsSUT1 likely facilitates Cd loading into the phloem and its subsequent long distance translocation. On the other hand, the expression of associated proteins in iron- and zinc-enriched rice flag leaves collectively establishes an efficient cellular-level barrier, mitigating Cd efflux and reabsorption and thereby regulating Cd transport flux toward the panicles, providing a new direction for breeding low Cd varieties by selecting genotypes with enhanced the flag leaf Cd enrichment capacity. The phenomenon of stem Cd indirectly promoting brown rice accumulation by regulating the panicle loading capacity reveals the complexity of stems as “transport hubs”: while stems are the necessary channel for Cd transport from roots to panicles, the xylem–phloem Cd partitioning ratio may determine the final grain translocation efficiency. This suggests that targeted regulation of stem Cd “unloading” processes (e.g., enhancing cell wall immobilization) could offer a more precise pollution control value than simply reducing the total stem Cd. The two-step Fe–Zn regulatory strategy demonstrates spatiotemporal specificity: the tillering stage strengthens root barriers (via early ion competition, e.g., Fe²⁺ versus Cd²⁺ for IRT1 transporters) to reduce Cd uptake, while the reproductive stage enhances flag leaf sequestration and inhibits panicle translocation. In summary, this study not only quantifies the contribution weights of organ Cd accumulation and Fe–Zn intervention but also reveals the coordinated mechanism of rice responding to Cd stress through “organ division of labor” (root immobilization, leaf interception, stem regulation, and panicle termination). Future research should focus on further validating and optimizing the broad applicability of the brown rice Cd accumulation model: integrating transcriptomic and metabolomic data to elucidate the molecular networks of key Fe–Zn regulatory nodes, thereby providing theoretical support for low Cd rice breeding and agronomic regulation.

5. Conclusions

In summary, through the strategy of “pretreatment of seedlings with Fe and Zn + foliar fortification at the tillering and heading growth stages”, the migration and accumulation of Cd in rice were effectively blocked (the Cd content in brown rice of early season rice decreased by 18%, and that of late season rice decreased by 25%). This strategy strengthened the Cd blocking barrier of roots at the tillering stage, inhibited the transport of Cd from the stem to the panicle at the heading and pustulation stage, and enhanced the Cd holding capacity of flag leaves. By constructing a high-precision prediction model (R² = 0.774), it was further revealed that the Cd accumulation in brown rice was directly regulated by the Cd content in the panicle (positively), the application amount of iron and zinc, and the Cd content in flag leaves (negatively) and was indirectly affected by the Cd content in the stem (indirectly positively). Finally, this study quantified the contributions of various organs at different growth stages. By utilizing the principle of nutritional competition among Fe, Zn, and Cd, it provided an economical and green agronomic solution for the safe utilization of moderately and slightly Cd-polluted paddy fields.