Abstract
Globally, 94–220 million people in regions with arsenic (As)-contaminated soil or groundwater face significant health risks. Rice, the primary staple food in these areas, is the main source of As exposure for a large portion of this population. Developing low-As rice cultivars provides a sustainable strategy to reduce dietary As exposure. However, As uptake in rice shares pathways with nutrient uptake, such as phosphate (P) transporters. The lack of As-specific transporters makes it challenging to reduce As accumulation in plants by regulating As uptake genes, as such approaches risk disrupting P uptake and plant growth. Here, we functionally characterized two phosphate transporter genes, OsPht1;9/OsPht1;10, which play a key role in arsenate (AsV) uptake and translocation in rice but minimally contribute to P utilization. Under hydroponic conditions, the double mutants of OsPht1;9/1;10 exhibited a 46.2–65.7% reduction in shoot As accumulation, with the As concentrations in xylem sap being 16.5–34.8% lower than the wild type controls. In multi-year field trials at two locations, simultaneous knockout of OsPht1;9/1;10 significantly decreased grain As concentration by 19.2–47.3%, without compromising yield. This study identifies novel gene-editing targets for low-As rice development and provides a breakthrough in mitigating As contamination exposure while enhancing food safety.
Similar content being viewed by others
Introduction
Toxic metalloid arsenic (As) is classified as a Class-I carcinogen by the World Health Organization and the International Agency for Research on Cancer1. Arsenic exposure poses potential risks to human health, as it enters the food chain via contaminated diet and drinking water2. The rate of soil As exceedance is 1.1%, which is most remarkable in southern and southwestern China and in south and southeast Asia3. Rice (Oryza sativa), a staple food for nearly half of the global population, particularly in Asia, is the dominant source of As via diet intake due to its As accumulation in the grain4,5,6. Based on a machine-learning model, it is found that the As accumulation in rice grains will continue to worsen, with contamination rates projected to reach 5.2% by 20407. Understanding the mechanism of As accumulation in rice and developing low-As rice varieties are therefore critical for protecting human health8.
Besides small amounts of methylated As and thioarsenate, arsenate (AsV) and arsenite (AsIII) are the two primary As species in soils5. Unlike in soils, As in rice grains is present primarily as AsIII, AsV, and dimethylarsinic acid, with the inorganic species being more toxic to humans9,10. In flooded paddy soils, AsIII is the dominant form, contributing the most to As accumulation in rice11. AsIII is predominantly taken up by rice roots via the silicon uptake pathway12,13. A portion is effluxed back into the rhizosphere through the inward-facing cortical aquaporin OsLsi1; another fraction is complexed with phytochelatins (PCs) to form AsIII-PC, which are subsequently sequestered into root vacuoles by the transporter OsABCC113,14. The remaining AsIII is loaded into the xylem via OsLsi2 and OsNramp1, facilitating its translocation from root to shoot 12,15,16.
However, under non-flooded or alternate wet-dry conditions, rice roots mainly absorb AsV2,17. Even under flooded conditions, AsV typically accounts for 5–20% of soil As. AsIII-oxidizing bacteria and denitrification-driven oxidation can elevate AsV levels in flooded paddy soils18,19,20. Further, due to the oxygen release by rice roots, AsV concentrations are higher in the rice rhizosphere than in the bulk soil21. Therefore, understanding the mechanism of AsV uptake is important to reduce the As content in rice grains. As a phosphate (P) analog, AsV is taken up by plants via P-transporter Pht122,23. Rice has 13 Pht1 transporters, which are responsible for P uptake and translocation, with some mediating AsV uptake and translocation, including OsPht1;1/1;4/1;824. For example, OsPht1;1 expression correlates with As content in shoots, indicating that it contributes to AsV uptake and translocation25. OsPht1;4 and OsPht1;8 exhibit high affinity for AsV, and mutations in either transporter increase AsV tolerance and reduce AsV uptake by 33–57% in rice26,27,28. In addition to AsV uptake, these genes influence grain yield. For example, knockout of OsPht1;4/1;8 decreases yield by 27.9–77.1%29,30. As such, it is difficult to knock out these genes to reduce AsV uptake while maintaining grain yield.
Reducing As accumulation in rice grains is critical for enhancing food safety. However, few studies have successfully achieved this through genetic modifications of rice8. This is attributable to the fact that the same transporters mediate the uptake of both P and As. Consequently, modifying these genes to reduce As accumulation often impairs the uptake of essential nutrient P. Although some genetic modifications reduce grain As accumulation, they frequently involve the introduction of foreign genes into the rice genome. For instance, overexpression of two nodulin26-like intrinsic protein genes (OsNIP1;1/3;3) restricts AsIII translocation to rice shoots, thereby reducing grain As accumulation31. Overexpression of the arsenite antiporter gene from Pteris vittata (PvACR3;1) promotes As sequestration in vacuoles, limiting its translocation to rice grains32. Similarly, a strategy was developed to reduce As in rice grains by enhancing vacuolar sequestration via transgenic expression of rice ABC transporter gene (OsABCC1), Saccharomyces cerevisiae yeast cadmium factor gene (ScYCF1), and a bacterial γ-glutamylcysteine synthetase33. This reduction was achieved only when multiple genes were co-expressed. However, approaches based on heterologous gene introduction or overexpression remain constrained by practical limitations, as the development of transgenic crop varieties remains prohibited in many regions. It is therefore crucial to identify key genes that can reduce As levels via gene editing or that harbor natural allelic variations influencing As accumulation in rice grains.
Recently, modifying gene expression has led to decreased As accumulation in rice grains8. For example, CRISPR/Cas9 technology was used to edit OsLsi2, a silicon efflux transporter, resulting in reduced grain As without reducing grain yield. This approach offers a more targeted and efficient way to reduce As contamination in rice grains34. However, knockout of OsLsi2 significantly impairs silicon uptake and may compromise disease resistance and lodging tolerance35. Therefore, editing this gene to create lower-As accumulation rice requires careful consideration of both its potential benefits and risks. Thus, it is necessary to identify additional functional genes applicable for developing low-As rice cultivars.
In rice, however, there is significant genetic and functional redundancy among the 13 Pht1 genes because they play distinct yet overlapping roles in As/P accumulation. Genes that differentially contribute to As and P uptake are likely to be identified. Such genes could be exploited to specifically reduce As accumulation in plants. Moreover, owing to decades of fertilizer input that have rendered most arable soils P-rich36,37, partial loss-of-function of individual Pht1 transporters is likely to exert only a minor effect on overall P acquisition. To explore novel members in AsV uptake and evaluate their potential for developing low-As rice, we focused on P transporters that show a pronounced response to AsV exposure (GSA: CRA037744, Table S1, Supplementary Data 1). The transcriptome database revealed that, except for OsPht1;5 whose expression was extremely low, OsPht1;10 exhibited the strongest up-regulation among Pht1 genes under AsV exposure. In rice, OsPht1;9 and OsPht1;10 share high sequence identity and are induced by P starvation, which play a redundant role in P uptake38. Their Arabidopsis homologs, Pht1;8 and Pht1;9, are likewise involved in AsV uptake39. However, their involvement in AsV uptake and potential to limit grain As accumulation remain unclear.
In this study, we investigated the roles of OsPht1;9/1;10 in rice AsV uptake and translocation for the first time. We analyzed the expression of these genes in response to AsV exposure and determined their AsV uptake ability. Furthermore, we constructed double mutants of OsPht1;9/1;10, which were tested in hydroponic and soil experiments. Our findings demonstrate that OsPht1;9/1;10 exhibits efficient AsV transport capacity. More importantly, mutating the two genes reduces the As accumulation in rice grains without affecting grain yield. These two genes are effective editing targets for reducing As accumulation in rice grains. Taken together, our results offer valuable insights into developing low-As rice cultivars to enhance food safety, a critical advancement in addressing global As contamination issues.
Results And Discussion
Arsenic exposure induced the expression of OsPht1;10 in rice but not OsPht1;9
To investigate the underlying mechanisms of As metabolism in rice, we established a transcriptome database of rice roots exposed to AsV (GSA: CRA037744, Supplementary Data 1). Analysis revealed that while OsPht1;9 transcripts were constitutively expressed under AsV exposure, OsPht1;10 expression was upregulated by 4.2-fold compared to the control (Supplementary Data 1, 2, Table S1, Fig. 1a, b). We conducted quantitative reverse transcription PCR (qRT-PCR) to further assess the expression levels of OsPht1;9/1;10 in rice roots under AsV or P treatments. Consistent with the transcriptome data, under sufficient P conditions, AsV exposure did not affect OsPht1;9 expression, but OsPht1;10 expression was increased by 3.7-fold (Fig. 1c, d, Supplementary Data 2). However, under deficient P conditions, both OsPht1;9/1;10 transcripts were upregulated by 18–19-fold (Fig. 1c, d, Supplementary Data 2). Further, AsV exposure suppressed their expression under these conditions (Fig. 1c, d, Supplementary Data 2).
FPKM values of OsPht1;9 (a) and OsPht1;10 (b) in the transcriptome database of rice roots under 0 (As0) or 20 μΜ AsV (As20) for 3 days. Quantitative RT-PCR verification of OsPht1;9 (c) and OsPht1;10 (d) expression levels in rice roots, which grew for 3–7 days in hydroponics with 100 μM P (P100), 100 μM P/20 μM AsV (P100As20), 0 μM P (P0), 0 μM P/20 μM AsV (P0As20). Values are means ± SE of 3 independent experiments. P-values were determined by either Student’s t test (a, b) or one-way ANOVA followed by Tukey’s test (c, d). Different letters indicate significant differences between different treatments (P < 0.05).
Previous studies have shown that the expression of OsPht1;9/1;10 is lower than that of OsPht1;1/1;4/1;8, which are the primary contributors to P uptake in rice. However, under P-deficient conditions, the expression of OsPht1;9/1;10 increased significantly. These data suggest that these genes play a crucial role in enhancing P uptake by rice under P starvation38. Typically, in response to AsV exposure, plants downregulate Pht1 expression to decrease their AsV uptake as they take up both P and AsV40,41. Specifically, highly expressed Pht1 genes such as OsPht1;1/1;2/1;8 are suppressed under AsV exposure25,26, leading to decreased P uptake in rice. Under sufficient P conditions with AsV exposure, the stable or upregulated expression of OsPht1;9/1;10 suggests that these two transporters likely compensate for the reduced P uptake due to the downregulation of other Pht1. However, under deficient P conditions, their transcription is repressed by AsV due to its toxicity, prompting rice to downregulate Pht1 to minimize As uptake and translocation. Therefore, P-transporters OsPht1;9/1;10 play a crucial role in AsV uptake and translocation in rice only under P-sufficient conditions.
Yeast Mutant EY917 expressed OsPht1;9/1;10 efficiently took up AsV
Both OsPht1;9/1;10 are capable of taking up P in rice, but their ability to take up AsV remains unknown. To evaluate this, we expressed them in the yeast mutant strain EY917, which lacks all five endogenous P transporters42, with OsPht1;4 being a control owing to its known AsV transport ability28. With no P transporter, the growth of yeast mutant EY917 relied solely on P uptake by expressed OsPht1 genes. We monitored the growth of yeast transformants by measuring OD600 values and fitting growth curves (Fig. 2a–c, Supplementary Data 2). In the absence of AsV (As-), OsPht1;9/1;10 transformants exhibited lower growth compared to the OsPht1;4 transformant. The data suggest that the P uptake ability of OsPht1;9/1;10 is similar, but less efficient than that of OsPht1;4.
Growth curves of yeast cells expressing OsPht1;4 (a), OsPht1;9 (b) and OsPht1;10 (c), where the transformants were grown in liquid medium containing 5 mM P + 0 (As-) or 1.5 mM AsV (As+). The AsV tolerance (d) and As accumulation (e) of yeast cells expressing OsPht1;4/1;9/1;10. Values are means ± SE of 3 independent experiments and different letters (d, e) indicate significant differences between different transformants using one-way ANOVA with Tukey multiple comparisons (P < 0.05). DW, dry weight.
To investigate the AsV uptake ability of OsPht1;9/1;10, we assessed the As tolerance and accumulation in yeast transformants expressing these genes. Growth of all three transformants OsPht1;4/1;9/1;10 was delayed by AsV exposure, indicating their ability to take up AsV (Fig. 2a–c, Supplementary Data 2). The growth rates under As exposure relative to no As revealed that yeast expressing OsPht1;9 (85.8%) and OsPht1;10 (88.9%) exhibited similar AsV tolerance, slightly lower than that of the OsPht1;4 transformants (95.3%) (Fig. 2d, Supplementary Data 2). Furthermore, after 2-h exposure to 100 μM AsV, the OsPht1;9/1;10 transformants accumulated 52.9 and 35.3 mg kg−1 As, being 2.1- and 1.4-fold higher than that of the OsPht1;4 transformant (Fig. 2e, Supplementary Data 2). The results suggest that OsPht1;9/1;10 have a higher AsV-uptake capacity than OsPht1;4. According to Ye et al., OsPht1;4 showed a strong AsV uptake ability, besides playing a pivotal role in P uptake in rice27,28. Our findings indicate that OsPht1;9/1;10 were more efficient than OsPht1;4 in AsV uptake, suggesting they may contribute to As accumulation in rice.
Double mutant of OsPht1;9/1;10 by CRISPR-Cas9 showed enhanced As tolerance
P-transporters OsPht1;9/1;10 exhibited high AsV-uptake efficiency in yeast (Fig. 2). To further explore their roles in AsV uptake in plants, we used CRISPR-Cas9 technology to generate OsPht1;9/1;10 mutant lines as described by Miao et al., 201343. Given that OsPht1;9/1;10 share redundant functions in P uptake, we focused on editing conserved sequence regions (Fig. 3a, b), and successfully obtained double mutants.
a The gene structures of OsPht1;9/1;10. The untranslated regions, exons, introns, start codons and stop codons are indicated by grey boxes, black boxes, black bars and red triangles. The yellow lines indicate the targeted sequences for gene editing. The mutation sites (b) and peak maps (c) of three independent ospht1;9/1;10 double mutant lines. Effect of 10 μΜ AsV on root elongation of wild type (WT) and two ospht1;9/1;10 double mutant lines under 100 μΜ P (P100) (d) or 0 μΜ P (P0) (e) conditions. Values are means ± SE of 8 replicates, and P values were determined by one-way ANOVA with Tukey multiple comparisons. *P < 0.001.
The targeted gene editing sites are located 80–90 bp away from the start codons of OsPht1;9/1;10, which share a high degree of identity and similarity at 89–91% (Fig. 3a)38. The targeted sequences differ by only one base, owing to the high conservation of these regions (Table S2). This similarity facilitated the simultaneous editing of both genes in rice, with three independent double mutants (DM-1, -2 and -3). Based on the sequencing of the lesion sites, OsPht1;9/1;10 were rendered non-functional in the mutants due to specific base deletions or insertions. Specifically, in OsPht1;9, there was a deletion of base A, a deletion of base T, and an insertion of base A in DM-1, -2, and -3, respectively. In OsPht1;10, there was a deletion of the base sequence CATG and base T in DM-1 and -2/3 (Fig. 3b, c). With the high similarity between the selected target sequences, we obtained the ospht1;10 single-mutant (ospht1;10, 10M-1 and -2) lines during the generation of the double mutant lines (Fig. S1a, b). At the same time, we searched the rice mutant seed library online hosted by Biogle (http://biogle.cn/index/about) and identified an ospht1;9 single-mutant (ospht1;9, 9 M) (Fig. S2a, b). We used two ospht1;10 single-mutant lines and one ospht1;9 single-mutant line as the control to assess the functional redundancy between them in As accumulation, similar to that in P accumulation in rice.
To investigate the roles of OsPht1;9/1;10 in As accumulation in rice, we conducted a root elongation assay using wild type (WT), ospht1;9 and ospht1;10 single mutants, and ospht1;9/1;10 double mutants under AsV exposure, with or without 100 μM P. Under P sufficient conditions at 100 μM P, the impact of 10 μM AsV on root length was weak across all rice lines. Nevertheless, the double mutant lines exhibited greater root elongation than the WT, whereas the ospht1;9 and ospht1;10 mutant lines showed no significant difference from the WT under the same AsV treatment (Fig. 3d, S1c, S2c, Supplementary Data 2). However, in the absence of P, relative root elongation decreased in all lines. Notably, the double mutants showed greater tolerance to AsV, exhibiting 35–49% longer roots than WT plants (Fig. 3e, Supplementary Data 2). There was no significant difference in AsV tolerance between the WT and the ospht1;9 or ospht1;10 single mutant lines (Fig. S1c, S2c, Supplementary Data 2). Regardless of P supply, the double mutant consistently displayed stronger AsV tolerance than the wild type.
Being chemical analogs, AsV and P often compete for the same transporter during plant uptake, leading to increasing As accumulation in plants under P-deficient conditions23. This competitive interaction explains the toxic effect of AsV on rice roots without P (Fig. 3d, e, S1c, S2c, Supplementary Data 2). Previous studies identified OsPht1;4/1;8 as key genes in AsV uptake and translocation in rice. For example, ospht1;8 mutants exposed to 6 μM AsV for 3 days showed 2.1–2.9 times longer roots than the WT26. Similarly, ospht1;4 mutants demonstrated 18−51% longer primary roots than the WT after exposure to 15 μM AsV for 10 days, while OsPht1;4-overexpressing lines showed 50% shorter roots27,28. The enhanced As tolerance of ospht1;4/1;8 mutants suggests their role in AsV uptake and translocation. Consistent with these findings, our results indicate that OsPht1;9/1;10 may also play an important role in AsV uptake and translocation, as evidenced by the increased AsV tolerance observed in the double mutants.
In a previous study, no changes in the P content were found in ospht1;9 or ospht1;10 single knockdown mutants. However, the P content was significantly reduced in the double-mutants compared to WT plants38. In other words, OsPht1;9/1;10 have redundant functions in P uptake in rice, loss of one gene is compensated by the other. As such, mutation of either OsPht1;9 or OsPht1;10 alone did not enhance their AsV tolerance, suggesting that the remaining gene probably compensates for As uptake in ospht1;9 or ospht1;10 single mutants (Fig. 3d, e, S1c, S2c, Supplementary Data 2). Based on our data, OsPht1;9/1;10 exhibit functional redundancy in AsV uptake in rice.
Double Mutation of OsPht1;9/1;10 decreased AsV uptake under P-sufficient condition
To further investigate the function of OsPht1;9/1;10, double mutants and WT rice seedlings were grown hydroponically under 20 μM AsV for 3 days under either P-sufficient and P-deficient conditions. Additionally, the As concentrations in the xylem sap of double mutants and WT plants were collected and analyzed.
Under P-sufficient conditions, shoot and root As concentrations in the ospht1;9/1;10 double mutant lines decreased by 46–66% and 12–28% compared to WT (Fig. 4a, b, Supplementary Data 2). Furthermore, As concentrations in the xylem sap of the double mutants were 16–35% lower than those of WT plants (Fig. 4c, Supplementary Data 2). However, under P-deficient conditions, As contents in ospht1;9/1;10 mutant lines were comparable to those of WT plants (Fig. S3a, b, Supplementary Data 2). Additionally, there was no notable difference in xylem-sap As concentrations between the double mutants and WT plants (Fig. S3c, Supplementary Data 2). In comparison, As concentrations in both roots and shoots of the ospht1;9 mutant were comparable to those of the WT (Fig. S2d, e, Supplementary Data 2). The root As content in the ospht1;10 mutant decreased by 12–22%, whereas shoot As content remained similar to that of the WT (Fig. S1d, e, Supplementary Data 2).
As concentrations in the shoots (a) and roots (b) of WT and ospht1;9/1;10 double mutant lines after exposure to 20 μM AsV for 3 days under 100 μΜ P conditions. (c) The As content in the xylem sap of rice plants after exposure to 20 μM AsV for 2 hours under 100 μΜ P treatments. Total P concentrations in the shoots (d) and roots (e) of WT and mutant lines after exposure to 20 μM AsV for 3 days under 100 μΜ P conditions. Values are means ± SE of 4 replicates and different letters indicate significant differences between different lines using one-way ANOVA with Tukey multiple comparisons (P < 0.05). DW, dry weight.
OsPht1;9/1;10 are expressed throughout rice roots, with the expression levels higher than those in the shoots38. This similar expression pattern suggests that both genes play a role in the uptake and translocation of P and As in rice. Under AsV exposure, OsPht1;10 showed a higher expression level than OsPht1;9 (Fig. 1a, b, Supplementary Data 1, 2), and the ospht1;10 single mutant lines showed lower As contents in the roots compared to WT, indicating that OsPht1;10 plays a more significant role in root As uptake. The As translocation in rice can be reduced by simultaneously mutating OsPht1;9/1;10 genes under P-sufficient conditions, whereas neither single mutant differs from the WT, indicating the two genes have redundant functions in As translocation in rice (Figs. 4a–c, S1d, e, S2d, e, Supplementary Data 2).
Concurrently, we examined the expression of other OsPht1 genes in the roots and shoots of WT and double mutant lines exposed to 20 μM AsV. As shown in Figs. S4–S5 and Supplementary Data 2, transcript levels of OsPht1;1, OsPht1;2, OsPht1;4 and OsPht1;8, the phosphate transporters previously reported to mediate AsV uptake, did not differ between WT and the double mutant lines. Thus, the reduced As accumulation in shoots of the double mutant is not attributable to compensatory up-regulation of these genes.
Our previous study showed that mutating OsPht1;4 decreased the As accumulation in rice plants, as well as AsV concentrations in the xylem sap, under both P-sufficient and P-deficient conditions28. Wang et al. found that ospht1;8 mutants lose 33–57% of their AsV uptake ability compared to WT plants26. Consistent with our data, mutating OsPht1;8 specifically reduces root As concentration without affecting that in shoots (Fig. S6a, b, Supplementary Data 2). A recent study in wheat reported that CRISPR knockout of TaPht1;9 reduced root and shoot As levels by 24.2% and 22.9%, respectively, compared with WT controls44. In Arabidopsis, the homologs of OsPht1;9 and OsPht1;10 are Pht1;8 and Pht1;9. Under AsV stress, the double RNAi lines targeting both genes showed an even greater reduction in As content than pht1;9 single mutants39. Similar effects observed in Pht1 mutant plants suggest that OsPht1;9/1;10 contribute to AsV uptake and accumulation in rice.
Under P-deficient conditions, the expression levels of Pht1 genes are up-regulated in response to P starvation to enhance P acquisition45. Because of the functional compensation by other highly expressed OsPht1 members, the relative contribution of OsPht1;9/1;10 to AsV uptake and translocation becomes limited. Additionally, Fig. 1c, d show that, although AsV treatment does not completely silence OsPht1;9/1;10 expression under P-deficient conditions, it markedly suppresses their transcript levels, further weakening their contribution to AsV uptake. Therefore, double mutation of OsPht1;9/1;10 had little effect on As accumulation in rice under P-deficient conditions.
In contrast, the P content was decreased in OsPht1;4/1;8 mutants compared to WT, owing to their crucial roles in rice P uptake29,30. Single mutants of either OsPht1;9 or OsPht1;10 showed no effect on plant P accumulation (Fig. S1f, g, Fig. S2f, g, Supplementary Data 2). Here, the double mutant of OsPht1;9/1;10 did not alter the P content in rice, regardless of the P supply (Fig. 4d, e, Fig. S3d, e, Supplementary Data 2). With rice variety Nipponbare as a control, mutating OsPht1;9/1;10 simultaneously had a mild impact on P content and a limited effect on plant growth38. The data indicate that the contribution of OsPht1;9/1;10 to P uptake is limited compared with that of other Pht1 transporters.
OsPht1;9/1;10 mutations decrease arsenic accumulation in rice grains in soil experiments
Our results show that OsPht1;9/1;10 exhibit AsV-uptake ability and both mediate AsV uptake and translocation in rice under hydroponics (Figs. 2–4). Here, we hypothesized that the double mutation of OsPht1;9/1;10 genes may also reduce the As content in rice grains. To test this, we performed long-term soil experiments where double mutants of OsPht1;9/1;10 and WT plants were grown to maturity in As-contaminated soil containing 53 mg kg-1 As at Guangzhou in 2020. As shown in Fig. 5a, b, there was no significant difference in either phenotype or grain yield between the double mutants and WT plants. However, it is worth noting that the As content in rice plants and grains of double mutants decreased compared with WT, with plant As decreasing by 25–62% and grains by 24–39% (Fig. 5c, d, Supplementary Data 2).
Phenotypes (a), grain yield (b) and As accumulation in the plants (c) and grains (d) of WT and ospht1;9/1;10 double mutant lines cultivated in potting soil at Guangzhou site in 2020. As accumulation in the grains of rice plants cultivated in potting soil at Sanya site in 2021 (e), 2023 (f) and at Guangzhou site in 2024 (g). Data are expressed as mean ± standard error of 6 replicates and analyzed by one-way ANOVA with Tukey multiple comparisons. The different letters indicate significant differences between different lines (P < 0.05). Scale bar: 10 cm. DW, dry weight.
To validate the results, we performed two additional soil experiments at Sanya in 2021 and 2023. Similar to the results of rice grown at Guangzhou in 2020 (Fig. 5d, Supplementary Data 2), total As in rice grains of double mutants decreased compared with WT (Fig. 5e, f, Supplementary Data 2). In addition, at Guangzhou in 2024, grain-As concentrations in the double mutant decreased by 41–67% compared to WT (Fig. 5g, Supplementary Data 2). As controls, neither ospht1;9 nor ospht1;10 single mutants differed from the WT in grain-As content (Figs. S1h, S2h, Supplementary Data 2). The results in two locations over five years suggest that OsPht1;9/1;10 affect As accumulation in grains by mediating AsV uptake and translocation in rice plants.
Given the structural similarity between AsV and P, all P transporters have the ability to transport both P and AsV. Rice, a common staple crop, possesses no transporters specific to either P or AsV. Among the 13 Pht1 transporters in rice, including OsPht1;9/1;10 and OsPht1;1/1;4/1;8, which all show AsV transport ability46. However, it is challenging to regulate As accumulation in rice by mutating a single transporter. This is because these rice Pht1 transporters play different roles during P acquisition. Some Pht1 genes, such as OsPht1;1/1;4, play pivotal roles in P uptake under different P conditions, being key genes for rice to keep its P homeostasis30,47. However, some genes are expressed under P-deficient conditions, enhancing their P acquisition in rice, such as OsPht1;9/1038. Currently, due to the long-term application of P fertilizers in agricultural soils, most fields are not P-deficient48. Therefore, the double mutants of OsPht1;9/1;10 may not impair rice P accumulation or normal growth. In our soil experiments over five years, the ospht1;9/1;10 double mutants stably reduced the As content in rice grains without affecting rice yield. Our study confirms P-transporters contribute to As uptake and translocation in rice. While OsPht1;1 mutation had no effect on grain As concentrations25, and OsPht1;8 mutation even increased it (Fig. S6c, d, Supplementary Data 2). Up to now, only OsPht1;4 is known to contribute to AsV accumulation in rice grains (Table S3). Here, we identify OsPht1;9/1;10 as two new key P transporters, which can reduce the As content in rice grains.
Arsenic contamination in agricultural soils, primarily caused by intensive farming and industrial activities, is already widespread. Machine-learning projections indicate a continuous rise in As concentrations in agricultural soils and rice grains from 2000 to 2040. The proportion of As-contaminated rice grains has increased from 4.2% in 2000 to 4.8% in 2020, with projections reaching 5.2% by 20407. Since rice serves as the staple food for over half of the world’s population, particularly in Asia, reducing As accumulation in rice grains through genetic improvement is crucial for minimizing dietary As exposure8. While existing mitigation measures have shown partial efficacy, more efficient strategies are urgently required to reverse this concerning trend. Developing low-As accumulating rice varieties is therefore critical for improving global food safety and protecting public health.
In soil, especially in the rhizosphere, the presence of AsV cannot be ignored due to the oxygen release by rice roots and the presence of AsIII-oxidizing bacteria18,21. Arsenic uptake contributes to As accumulation in rice grains. Previous research shows that P transporters OsPht1 can take up AsV in rice and contribute to grain As accumulation28. After decades of fertilization, most agricultural soils are now P-sufficient. Screening for transporters that differentially contribute to AsV uptake versus P acquisition, followed by precise editing of one or two that preferentially mediate AsV over P uptake would therefore provide a critical strategy for specifically lowering As accumulation in rice. Here, we compared the effects of mutations in different Pht1 genes with AsV transport capacity on As uptake, translocation, accumulation, and yield in rice (Table S3, Supplementary Data 2). Our multi-year soil experiments on OsPht1;9/1;10 mutants demonstrate that reducing grain As accumulation can be achieved without compromising rice yield, providing a feasible strategy for developing safer rice cultivars.
Similar to the shared P transporters for As uptake, metals are often absorbed by crops through pathways intended for nutrients49. In crops, each transporter family often has multiple members with distinct functions in the accumulation of nutrients and metals. The two OsPht1 members studied here are unique in that they contribute little to P acquisition compared with other members, yet play a significant role in AsV uptake and translocation. Analyses of Zn, Fe, and Mn concentrations in rice grains, we found that loss of OsPht1;9 and OsPht1;10 did not compromise essential grain micronutrient content (Fig. S7a–c, Supplementary Data 2). By targeting specific P transporters, we maintained rice yield and nutrient utilization while reducing As accumulation in rice grains.
Genetic engineering of rice remains restricted in many countries. Therefore, researchers often need to identify natural variations in target genes from existing rice populations. This approach enables the development of improved rice cultivars without genetic modification. Here, we analyzed the OsPht1;9/1;10 gene sequences in the rice resequencing variation database (https://iric.irri.org/projects/3000-rice-genomes-project)50. From sequence analysis of OsPht1;9/10 across 3000 rice varieties, OsPht1;9 had five nucleic acid sequence site variations, one of which alters the encoded amino-acid sequence. And OsPht1;10 exhibited 62 DNA variation types, ten of which cause amino-acid substitutions (Table S4, Supplementary Data 2). These sequence variants indicate that natural rice populations harbor multiple OsPht1;10 haplotypes whose transport characteristics remain to be quantified. If future work confirms differential As uptake among them, the allelic series could be explored for conventional breeding of low-As cultivars.
Our study specifically addressed AsV uptake via phosphate transporters, which is relevant to “mid-season baking” of paddy fields, radial oxygen loss (ROL) from roots, and the elevated AsV concentrations present in irrigation water. Previous studies have shown that editing OsLsi2 and overexpressing OsNIP1;1/3;3 and OsABCC1 effectively reduce AsIII accumulation in rice grains31,33,34. In future work, combining the OsPht1;9/1;10 knockout rice lines in this study with OsLsi2-knockout and OsNIP1;1/3;3/OsABCC1-overexpressing lines to simultaneously restrict both AsV and AsIII represents an effective strategy for comprehensively reducing As accumulation in rice grains.
Taken together, our results demonstrate that P-transporters OsPht1;9/1;10 can take up and translocate AsV in rice plants. Besides OsPht1;4, known to contribute to AsV uptake in rice grains, we have added OsPht1;9/1;10 as two new members. Importantly, double mutants of OsPht1;9/1;10 significantly reduce grain As accumulation without compromising yield, demonstrating their potential for developing safer rice cultivars. These findings not only expand our understanding of As accumulation mechanisms in rice but also provide valuable genetic targets for conventional breeding to improve food safety.
Methods
Plant materials and growth conditions
Rice lines, including wild type (WT) and mutant lines of Zhonghua 11 (ZH11), were used in this study. Rice seeds were soaked in deionized water overnight, surface-sterilized with 75% ethanol (1:3, v/v) for 1 min and then with diluted NaClO (1:2, v/v) for 30 min. After thorough rinsing with sterilized water three times, the seeds were germinated on half-strength Murashige and Skoog (1/2 MS) medium for 7 days51,52. Then uniformly-grown seedlings were transferred to a nutrient solution for subsequent hydroponics and soil-culture experiments.
The hydroponic experiments were conducted in a temperature-controlled growth room with a 16-h light (30°C)/8-h dark (25°C) photoperiod and 70% relative humidity. Standard rice culture solution (Kimura B nutrient salts solution) was used at pH 5.5. The nutrient solution was replaced weekly. In addition, soil experiments were conducted over 5 years in two locations in Guangzhou and Sanya, southern China. The As-contaminated soil used in the Guangzhou experiments was collected from farmlands near Dabaoshan Mountain, Shaoguan, with As concentrations of 53.1 mg kg−1 (2020, pH 7.03) and 21.4 mg kg−1 (2024 and 2025, pH 6.18). The soil used for the Sanya experiments was obtained from nearby farmlands, with an As concentration of 28.2 mg kg−1 (2021 and 2023, pH 5.65). Rice seedlings were cultivated in pots containing 2 kg of soil from March to August in Guangzhou and from December to April in Sanya. All experiments followed standard agricultural practices.
Expression of OsPht1;9/1;10 in rice under P deficiency or AsV exposure
To study the molecular mechanisms of arsenic uptake and metabolism in rice, we performed comparative transcriptome sequencing under AsV stress. Four-week-old rice seedlings were exposed for 3 d to nutrient solution either without As or supplemented with 20 μM AsV. Roots were snap-frozen in liquid nitrogen, and total RNA was extracted using Trizol reagent kit (Invitrogen, Carlsbad, CA, USA). RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and checked using RNase-free agarose gel electrophoresis. Poly(A)-enriched eukaryotic mRNA was captured with oligo(dT)-coated magnetic beads and fragmented using high temperatures. Sequencing libraries were generated using the NEBNext Ultra RNA Library Prep Kit for Illumina (NEB, USA). Briefly, fragmented mRNA was used as a template to synthesize the first-strand cDNA with random hexamers. Second-strand cDNA was synthesized according to the manufacturer’s instructions. Then the cDNA fragments were purified, end-repaired, dA-tailed, and ligated to Illumina sequencing adapters. RNA-seq analysis was performed by Gene Denovo Biotechnology Co., Ltd (Guangzhou, China) using an Illumina HiSeqTM 4000 platform with paired-end 150-bp sequencing (PE150). Raw reads generated by sequencing were quality-controlled with fastp (version 0.18.0) to remove low-quality data, yielding clean reads53. Clean reads were then aligned to the Nipponbare rice reference genome using HISAT2 2.4 with “-rna-strandness RF” and other parameters set as default54. The mapped reads of each sample were assembled using StringTie v1.3.1 (-f 0.3) in a reference-based approach. For each transcript, an FPKM (fragment per kilobase of transcript per million mapped reads) value was calculated to quantify its expression abundance and variations, using StringTie software55. The expression levels of OsPht1;9/1;10 were retrieved from the transcriptome database as FPKM values 56.
To examine the response of OsPht1;9/1;10 expression under P deficiency or AsV stress, rice seedlings were transferred to nutrient solution containing different P and AsV concentrations (μM), including 100 P (P100), 100 P/20 AsV (P100As20), 0 P (P0), or 0 P/20 AsV (P0As20). The plants were P-starved for 7 d before being exposed to AsV for 3 d. In this study, Na2HPO4 and Na2HAsO4·7H2O (Sigma-Aldrich, USA) were used26. Total RNA from the rice roots was extracted using Plant Total RNA Isolation Kit (Vazyme Biotech, Nanjing, China), and first-strand cDNA was synthesized using HiScript II One Step RT-PCR Kit (Vazyme Biotech, Nanjing, China). Quantitative RT-PCR (qRT-PCR) was performed with SYBR Green PCR Master Mix (Vazyme Biotech, Nanjing, China) on the Step One Real-time PCR system (Applied Biosystems). The histone H3 gene was used as an internal reference gene. All specific primers used for qRT-PCR are shown in Table S2. Relative expression levels of OsPht1;9/1;10 were calculated with the 2−ΔΔCt method and normalized by the transcript level under P100 condition 57,58.
Arsenic tolerance and accumulation of yeast expressing OsPht1;9/1;10
The coding sequences of three P-transporters OsPht1;4/1;9/1;10 were cloned into the GLU1 promoter cassette of yeast expression vector pAG426GLU-ccdB (www.addgene.org/) using the Hieff Clone Universal One Step Cloning Kit (Yeasen, Shanghai, China). OsPht1;4 is known for its AsV uptake and translocation in rice, so it was therefore used as the control in the yeast experiment28. The cloning primers and homologous recombination primers are shown in Table S2. The above constructs were transformed into yeast strain EY917, which lacks all five P transporters (Δpho84, 87, 89, 90, and 91)59. Only when yeast strain EY917 expresses the target protein does it acquire P/AsV uptake activity42. Transformed yeast cells were grown on solid synthetic dropout (-Ura) medium for 4 days, with positive transformants appearing at this time.
For the growth assay, all transformants were pre-incubated to an optical density OD600 of 1 in SD-U liquid medium containing 5 mM P, the minimum P concentration required for yeast growth. The yeast cells were enriched by centrifugation at 1000 g for 5 min and then transferred to 25 mL SD-U liquid medium containing 5 mM P plus 0 (As-) or 1.5 mM AsV (As+), with an initial OD600 of 0.2. The P/AsV ratio for this experiment was selected according to Sun et al42,60,61. Under shaking conditions (200 rpm, 30°C), 1 mL medium was collected periodically to measure OD600 until the transformants were cultured to OD600 of 1. Growth curves of the transformants were fitted using the exponential function y(x) = a·ekx based on OD600 values (where k is the growth rate constant). AsV tolerance of yeast transformants expressing OsPht1;4/1;9/1;10 was presented by the percentage of k(As+)/k(As-).
For AsV uptake assay, yeast transformants expressing OsPht1;4/1;9/1;10 were pre-cultured to an OD600 of 1 in 25 mL SD-U liquid medium. After being harvested by centrifugation and resuspended in sterile water, the yeast cells were exposed to 100 µM AsV for 2 h, which is sufficient for AsV sorption by yeast cells with minimal growth limitation. Collected by centrifugation and washed with iced deionized water three times, the cells were finally dried at 60°C and digested with 50% HNO3−30% H2O2 at 105°C for As determination 62.
Construction and characterization of gene-edited rice mutants
For CRISPR/Cas9-edited mutation, the specific targets residing in the exons of OsPht1;9/1;10 were predicted using the CRISPR design tool CRISPR-GE (http://skl.scau.edu.cn/home/)43. These specific single guide RNAs (sgRNAs) were cloned into the CRISPR/Cas9 vector. After sequencing verification, the constructs were transformed into the mature embryos of rice ZH11 using Agrobacterium-mediated transformation as described by Upadhyaya et al63. Target sequences of OsPht1;9/1;10 are shown in Table S2.
For characterization of the CRISPR/Cas9-edited rice lines, PCR primers were designed based on the target region of gene sequences (Table S2). After DNA extraction of rice plants, PCR amplification of the targeted regions, and sequencing of the PCR products, gene-edited lines containing alterations or deletions were screened. We succeeded only in generating the double mutant (ospht1;9/1;10, DM-1, DM-2 and DM-3) and the ospht1;10 single-mutant (ospht1;10, 10M-1 and 10M-2) lines, but failed to generate transgenic materials with ospht1;9 single mutations. From Biogle (http://biogle.cn/index/about), we obtained an ospht1;9 single-mutant line (9 M). Homozygous lines were selected for physiological experiments 64.
Arsenic tolerance and accumulation in rice mutant lines
To determine the effects of OsPht1;9/1;10 mutations on AsV tolerance in rice roots, WT, two ospht1;9/1;10 double mutant lines, one ospht1;9 single mutant line and two ospht1;10 single mutant lines were used. After germination in the dark for 3 days, the rice seedlings were transferred to 0.5 mM CaCl2 solution containing 0 or 10 μM AsV with or without 100 μM P. After 3 days of treatment, the main root length was measured 26.
To determine As uptake and translocation, 14-day-old rice seedlings of WT, three ospht1;9/1;10 double mutant lines, one ospht1;9 mutant line, and two ospht1;10 mutant lines were transferred to a complete nutrient solution containing 20 μM AsV for 3 days. Then the rice shoots and roots were separated and dried in a 60°C oven for 3 days. After digestion, the As and P concentrations in rice were determined. In addition, xylem sap was collected to determine the As translocation from the roots to shoots according to Cao et al28. After exposure to 20 µM AsV for 2 h, all rice stems of different lines were cut at 2 cm above the root junction, and the cut surface was immediately wrapped in absorbent cotton for 2 h. The xylem sap absorbed by the cotton was collected by centrifugation for As determination. In addition, after a 7-day P-deficient pretreatment, the As and P contents in rice tissues and the As concentrations in xylem sap of WT and ospht1;9/1;10 double mutant lines were also measured.
To assess whether knocking out OsPht1;9/1;10 induces compensatory expression of other Pht1 members, we quantified the expression levels of other OsPht1 genes in the roots and shoots of WT and double mutant plants under 20 µM AsV exposure for 3 days. Because OsPht1;11 and OsPht1;13 are mycorrhiza-inducible, and OsPht1;3 and OsPht1;12 are expressed at extremely low levels that were barely detectable, we analyzed the expression of OsPht1;1, OsPht1;2, OsPht1;4, OsPht1;5, OsPht1;6, OsPht1;7, and OsPht1;8. RNA extraction and qRT-PCR were performed as described above. All specific primers used for qRT-PCR are shown in Table S2.
Soil experiments were performed at different times and locations in southern China as described earlier. After harvest at maturity, agronomic traits, including grain yield per plant, tiller number, and plant height, were measured. Rice grains and plants were dried for 3 days for P and As determination.
Determination of total As and P concentrations in yeast and rice
Dried samples of yeast, rice shoots, roots, plants, and grains were digested with 1:1 (v/v) HNO3−30% H2O2 at 105°C and then diluted with distilled water65. After filtration, the supernatant was analyzed for total P concentration by inductively coupled plasma optical emission spectrometer (ICP-OES, Avio 500) and for total As concentration by inductively coupled plasma mass spectrometry (ICP-MS, PE NexION 350D).
Statistics and Reproducibility
For quality assurance and quality control, we added indium (In) as an internal standard to blanks, calibration standards, and samples for As determination. The recovery range of In was between 90−110%. During P determination, blank and standard samples were tested every 20 samples.
Data are presented as the mean of 3−8 replicates with standard error and analyzed using SPSS software. The significant differences were determined by Student’s t test or one-way ANOVA followed by Tukey’s multiple-comparison test.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All data generated, collected, or analyzed in this study are included in the main text, Supplementary Information, and Supplementary Data. Numeric source data for all figures are provided in the Supplementary Data 2 file. The gene expression data from the transcriptomic analysis in this study are provided in Table S1 and the Supplementary Data 1 file. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2025) in the National Genomics Data Center (Nucleic Acids Res 2025), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA037744), which are publicly accessible at https://ngdc.cncb.ac.cn/gsa66.
Code availability
This study did not use any custom code.
References
IARC Monographs on the Evaluation of Carcinogenic Risk to Humans. 100C. Lyon: International Agency for Research on Cancer (2012).
Guo, Y. et al. Dynamics of cadmium and arsenic at the capillary fringe of paddy soils: A microcosm study based on high-resolution porewater analysis. Soil Environ. Health 2, 100057 (2024).
Hou, D. Y. et al. Global soil pollution by toxic metals threatens agriculture and human health. Science 388, 316–321 (2025).
Williams, P. N. et al. Greatly enhanced arsenic shoot assimilation in rice leads to elevated grain levels compared to wheat and barley. Environ. Sci. Technol. 41, 6854–6859 (2007).
Zhao, F. J., Zhu, Y. G. & Meharg, A. A. Methylated Arsenic Species in Rice: Geographical Variation, Origin, and Uptake Mechanisms. Environ. Sci. Technol. 47, 3957–3966 (2013).
Qi, C. C. et al. Global and regional patterns of soil metal(loid) mobility and associated risks. Nat. Commun. 16, 2947 (2025).
Zhang, S. et al. Escalating arsenic contamination throughout Chinese soils. Nat. Sustain. 7, 766–775 (2024).
Chauhan, R. et al. Biotechnological strategies for remediation of arsenic-contaminated soils to improve soil health and sustainable agriculture. Soil Environ. Health 2, 100061 (2024).
Sarwar, T., Khan, S., Muhammad, S. & Amin, S. Arsenic speciation, mechanisms, and factors affecting rice uptake and potential human health risk: A systematic review. Environ. Technol. Innov. 22, 101392 (2021).
Wang, J. J. et al. Thiolated arsenic species observed in rice paddy pore waters. Nat. Geosci. 13, 282–287 (2020).
Khan, M. A., Stroud, J. L., Zhu, Y. G., McGrath, S. P. & Zhao, F. J. Arsenic Bioavailability to Rice Is Elevated in Bangladeshi Paddy Soils. Environ. Sci. Technol. 44, 8515–8521 (2010).
Ma, J. F. et al. Transporters of arsenite in rice and their role in arsenic accumulation in rice grain. Proc. Natl. Acad. Sci. USA 105, 9931–9935 (2008).
Zhao, F. J. et al. The role of the rice aquaporin Lsi1 in arsenite efflux from roots. N. Phytol. 186, 392–399 (2010).
Song, W. Y. et al. A rice ABC transporter, OsABCC1, reduces arsenic accumulation in the grain. Proc. Natl. Acad. Sci. USA 111, 15699–15704 (2014).
Chen, Y. et al. The role of nodes in arsenic storage and distribution in rice. J. Exp. Bot. 66, 3717–3724 (2015).
Tiwari, M. et al. Expression in Arabidopsis and cellular localization reveal involvement of rice NRAMP, OsNRAMP1, in arsenic transport and tolerance. Plant Cell Environ. 37, 140–152 (2014).
Xu, X. Y., McGrath, S. P., Meharg, A. A. & Zhao, F. J. Growing rice aerobically markedly decreases arsenic accumulation. Environ. Sci. Technol. 42, 5574–5579 (2008).
Zhang, J. et al. Anaerobic Arsenite Oxidation by an Autotrophic Arsenite-Oxidizing Bacterium from an Arsenic-Contaminated Paddy Soil. Environ. Sci. Technol. 49, 5956–5964 (2015).
Huang, H., Jia, Y., Sun, G. X. & Zhu, Y. G. Arsenic Speciation and Volatilization from Flooded Paddy Soils Amended with Different Organic Matters. Environ. Sci. Technol. 46, 2163–2168 (2012).
Lomax, C. et al. Methylated arsenic species in plants originate from soil microorganisms. N. Phytol. 193, 665–672 (2012).
Seyfferth, A. L., Webb, S. M., Andrews, J. C. & Fendorf, S. Arsenic Localization, Speciation, and Co-Occurrence with Iron on Rice (Oryza sativa L.) Roots Having Variable Fe Coatings. Environ. Sci. Technol. 44, 8108–8113 (2010).
Meharg, A. A. & Macnair, M. R. An Altered Phosphate-Uptake System in Arsenate-Tolerant Holcus-Lanatus L. N. Phytol. 116, 29–35 (1990).
Zhao, F. J., Ma, J. F., Meharg, A. A. & McGrath, S. P. Arsenic uptake and metabolism in plants. N. Phytol. 181, 777–794 (2009).
Wu, Z. C., Ren, H. Y., McGrath, S. P., Wu, P. & Zhao, F. J. Investigating the Contribution of the Phosphate Transport Pathway to Arsenic Accumulation in Rice. Plant Physiol. 157, 498–508 (2011).
Kamiya, T., Islam, M. R., Duan, G. L., Uraguchi, S. & Fujiwara, T. Phosphate deficiency signaling pathway is a target of arsenate and phosphate transporter OsPT1 is involved in As accumulation in shoots of rice. Soil Sci. Plant Nutr. 59, 580–590 (2013).
Wang, P., Zhang, W., Mao, C., Xu, G. & Zhao, F. J. The role of OsPT8 in arsenate uptake and varietal difference in arsenate tolerance in rice. J. Exp. Bot. 67, 6051–6059 (2016).
Ye, Y. et al. OsPT4 contributes to arsenate uptake and transport in rice. Front. Plant Sci. 8, 2197 (2017).
Cao, Y. et al. Knocking Out OsPT4 gene decreases arsenate uptake by rice plants and inorganic arsenic accumulation in rice grains. Environ. Sci. Technol. 51, 12131–12138 (2017).
Jia, H. F. et al. The Phosphate Transporter Gene OsPht1;8 Is Involved in Phosphate Homeostasis in Rice. Plant Physiol. 156, 1164–1175 (2011).
Zhang, F. et al. Involvement of OsPht1;4 in phosphate acquisition and mobilization facilitates embryo development in rice. Plant J. 82, 556–569 (2015).
Sun, S. K. et al. Decreasing arsenic accumulation in rice by overexpressing and through disrupting arsenite radial transport in roots. N. Phytol. 219, 641–653 (2018).
Chen, Y. S. et al. Expressing Arsenite Antiporter PvACR3;1 in Rice (Oryza sativa L.) decreases inorganic arsenic content in rice grains. Environ. Sci. Technol. 53, 10062–10069 (2019).
Deng, F. et al. Engineering rice with lower grain arsenic. Plant Biotechnol. J. 16, 1691–1699 (2018).
Xu, X. J., Sun, S. K., Zhang, W. W., Tang, Z. & Zhao, F. J. Editing silicon transporter genes to reduce arsenic accumulation in rice. Environ. Sci. Technol. 58, 1976–1985 (2024).
Ma, J. F. et al. An efflux transporter of silicon in rice. Nature 448, 209–212 (2007).
Zou, T., Zhang, X. & Davidson, E. A. Global trends of cropland phosphorus use and sustainability challenges. Nature 611, 81–87 (2022).
Gong, H.Q. et al. A dynamic optimization of soil phosphorus status approach could reduce phosphorus fertilizer use by half in China. Nat. Commun. 16 (2025).
Wang, X. F. et al. Phosphate transporters OsPHT1;9 and OsPHT1;10 are involved in phosphate uptake in rice. Plant Cell Environ. 37, 1159–1170 (2014).
Remy, E. et al. The Pht1;9 and Pht1;8 transporters mediate inorganic phosphate acquisition by the Arabidopsis thaliana root during phosphorus starvation. N. Phytol. 195, 356–371 (2012).
Castrillo, G. et al. WRKY6 Transcription Factor Restricts Arsenate Uptake and Transposon Activation in Arabidopsis. Plant Cell 25, 2944–2957 (2013).
Navarro, C. et al. Arsenite provides a selective signal that coordinates arsenate uptake and detoxification through the regulation of PHR1 stability in. Mol. Plant 14, 1489–1507 (2021).
Sun, D. et al. Expression of New Pteris vittata Phosphate Transporter PvPht1;4 Reduces Arsenic Translocation from the Roots to Shoots in Tobacco Plants. Environ. Sci. Technol. 54, 1045–1053 (2020).
Miao, J. et al. Targeted mutagenesis in rice using CRISPR-Cas system. Cell Res. 23, 1233–1236 (2013).
Wang, P. F. et al. Wheat PHT1;9 acts as one candidate arsenate absorption transporter for phytoremediation. J. Hazard. Mater. 452, 131219 (2023).
Wang, Y., Wang, F., Lu, H., Liu, Y. & Mao, C. Phosphate Uptake and Transport in Plants: An Elaborate Regulatory System. Plant Cell Physiol. 62, 564–572 (2021).
Tang, Z. & Zhao, F. J. The roles of membrane transporters in arsenic uptake, translocation and detoxification in plants. Crit. Rev. Environ. Sci. Technol. 51, 2449–2484 (2021).
Sun, S. B. et al. A Constitutive Expressed Phosphate Transporter, OsPht1;1, Modulates Phosphate Uptake and Translocation in Phosphate-Replete Rice. Plant Physiol. 159, 1571–1581 (2012).
Bindraban, P. S., Dimkpa, C. O. & Pandey, R. Exploring phosphorus fertilizers and fertilization strategies for improved human and environmental health. Biol. Fertil. Soils 56, 299–317 (2020).
Huang, S., Yamaji, N. & Ma, J. F. Metal Transport Systems in Plants. Annu. Rev. Plant Biol. 75, 1–25 (2024).
Wang, W. et al. Genomic variation in 3,010 diverse accessions of Asian cultivated rice. Nature 557, 43–49 (2018).
Cao, Y. et al. Identification and expression analysis of OsLPR family revealed the potential roles of OsLPR3 and 5 in maintaining phosphate homeostasis in rice. BMC Plant Biol. 16, 210 (2016).
Ai, H. et al. Mutation of Enhances Tolerance to Phosphate Starvation in Rice. Int. J. Mol. Sci. 24, 2437 (2023).
Chen, S. F., Zhou, Y. Q., Chen, Y. R. & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34, 884–890 (2018).
Kim, D., Landmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12, 357–360 (2015).
Pertea, M. et al. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat. Biotechnol. 33, 290–295 (2015).
Sun, D. et al. Novel phytase PvPHY1 from the As-hyperaccumulator Pteris vittata enhances P uptake and phytate hydrolysis, and inhibits As translocation in Plant. J. Hazard. Mater. 423, 127106 (2022).
Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods 25, 402–408 (2001).
Hellemans, J., Mortier, G., De Paepe, A., Speleman, F. & Vandesompele, J. qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol. 8, R13 (2007).
Wykoff, D. D. & O’Shea, E. K. Phosphate transport and sensing in Saccharomyces cerevisiae. Genetics 159, 1491–1499 (2001).
Sun, D. et al. Novel Mycorrhiza-Specific P Transporter PvPht1;6 Contributes to As Accumulation at the Symbiotic Interface of As-Hyperaccumulator Pteris vittata. Environ. Sci. Technol. 56, 14178–14187 (2022).
Cao, Y. et al. Heterologous Expression of Pteris vittata Phosphate Transporter PvPht1;3 Enhances Arsenic Translocation to and Accumulation in Tobacco Shoots. Environ. Sci. Technol. 53, 10636–10644 (2019).
Sun, D. et al. Novel Phosphate Transporter-B PvPTB1;1/1;2 Contribute to Efficient Phosphate Uptake and Arsenic Accumulation in As-Hyperaccumulator. Environ. Sci. Technol. 58, 7346–7356 (2024).
Upadhyaya, N. M. et al. Agrobacterium-mediated transformation of Australian rice cultivars Jarrah and Amaroo using modified promoters and selectable markers. Aust. J. Plant Physiol. 27, 201–210 (2000).
Modrzejewski, D. et al. What is the available evidence for the range of applications of genome-editing as a new tool for plant trait modification and the potential occurrence of associated off-target effects: a systematic map. Environ. Evid. 8, R27 (2019).
Feng, H. et al. Expressing Phosphate Transporter PvPht2;1 Enhances P Transport to the Chloroplasts and Increases Arsenic Tolerance in Arabidopsis thaliana. Environ. Sci. Technol. 55, 2276–2284 (2021).
Zhang, S. et al. The GSA Family in 2025: A broadened sharing platform for multi-omics and multimodal data. Genomics Proteom. Bioinform. (2025).
Acknowledgements
This work was financially supported by the National Natural Science Foundation of China (Nos. 42322701, 42307009), and the Natural Science Foundation of Guangdong Province of China (No. 2024A1515011134, 2025A1515010723). We also thank Man Zhao from the School of Environmental Science and Engineering, Sun Yat-sen University, for technical support and suggestions on ICP analysis.
Author information
Authors and Affiliations
Contributions
Y.C. conceived the idea. H. -Y. F. performed the experimental studies and wrote the manuscript. C. -T. C., M. -Y. X., D. S., X. -X. S., N. G., J. -H. Q., and C. -J. L. performed the experimental studies. Y.C., Y. -T. T and R. -L. Q. supervised the work. Y.C., G. -H. X and L. Q. M revised the paper. All authors have read and approved the final manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Biology thanks Ajay Pandey and the other anonymous reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Showkat Ganie and David Favero. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Feng, H., Chen, C., Xu, M. et al. Knocking out OsPht1;9-1;10 genes decreases arsenic accumulation in rice (Oryza sativa) grains. Commun Biol 9, 518 (2026). https://doi.org/10.1038/s42003-026-09741-5
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s42003-026-09741-5
