Phosphorus Starvation- and Zinc Excess-Induced Astragalus sinicus AsZIP2 Zinc Transporter Is Suppressed by Arbuscular Mycorrhizal Symbiosis

Abstract

Zinc (Zn) is one of the most essential micronutrients for plant growth and metabolism, but Zn excess can impair many basic metabolic processes in plant cells. In agriculture, crops often experience low phosphate (Pi) and high Zn double nutrient stresses because of inordinate agro-industrial activities, while the dual benefit of arbuscular mycorrhizal (AM) fungi protects plants from experiencing both deficient and toxic nutrient stresses. Although crosstalk between Pi and Zn nutrients in plants have been extensively studied at the physiological level, the molecular basis of how Pi starvation triggers Zn over-accumulation in plants and how AM plants coordinately modulate the Pi and Zn nutrient homeostasis remains to be elucidated. Here, we report that a novel AsZIP2 gene, a Chinese milk vetch (Astragalus sinicus) member of the ZIP gene family, participates in the interaction between Pi and Zn nutrient homeostasis in plants. Phylogenetic analysis revealed that this AsZIP2 protein was closely related to the orthologous Medicago MtZIP2 and Arabidopsis AtZIP2 transporters. Gene expression analysis indicated that AsZIP2 was highly induced in roots by Pi starvation or Zn excess yet attenuated by arbuscular mycorrhization in a Pi-dependent manner. Subcellular localization and heterologous expression experiments further showed that AsZIP2 encoded a functional plasma membrane-localized transporter that mediated Zn uptake in yeast. Moreover, overexpression of AsZIP2 in A. sinicus resulted in the over-accumulation of Zn concentration in roots at low Pi or excessive Zn concentrations, whereas AsZIP2 silencing lines displayed an even more reduced Zn concentration than control lines under such conditions. Our results reveal that the AsZIP2 transporter functioned in Zn over-accumulation in roots during Pi starvation or high Zn supply but was repressed by AM symbiosis in a Pi-dependent manner. These findings also provide new insights into the AsZIP2 gene acting in the regulation of Zn homeostasis in mycorrhizal plants through Pi signal.

Keywords: AsZIP2 transporter; Astragalus sinicus; ZIP gene family; Zinc; arbuscular mycorrhizal fungi; crosstalk between Pi-Zn; phosphate.

Figure 1

Physiological analysis of phosphorus and zinc interaction in A. sinicus plants. A. sinicus plants were grown in pot cultures treated with 300 µM Pi and 50 μM Zn (+Pi+Zn), 300 µM Pi and 0.5 μM Zn (+Pi-Zn), 30 µM Pi and 50 μM Zn (-Pi+Zn) or 30 µM Pi and 0.5 μM Zn (-Pi-Zn). Pi and Zn concentrations were quantified in roots (black) and shoots (gray) of 56-d-old NM (a,b) or AM (c,d) plants. A. sinicus roots colonized by R. irregularis at 42 dpi. (a,c) Total P concentration in A. sinicus; (b,d) Zn concentration in A. sinicus. Error bars indicate the standard deviation from three biological replicates. The different letters are statistically significant differences among treatments at p < 0.05, based on Duncan’s multiple range test. DW, Dry weight.

Figure 2

Evolutionary relationship among AsZIP2 in A. sinicus and ZIP family zinc/iron transporters in different plant species. The evolutionary history was inferred using MEGA7.0 program with the Neighbour–Joining method (Kumar et al., 2016). The evolutionary distances were computed using the Poisson correction method. All ambiguous positions of the 63 amino acid sequences were removed for each sequence pair. Plant ZIP family proteins tested were classified as five subfamilies based on sequence similarity. The branch marked with a red circle represents the Zn transporter AsZIP2 of A. sinicus in this study. The accession numbers and plant species names are provided in Supplementary Table S2.

Figure 3

The expression patterns of AsZIP2 in A. sinicus. (a–d) Relative expression of AsZIP2 in response to external Pi and Zn treatments, as determined by real time qRT-PCR. 56-day-old A. sinicus plants were grown in sand cultures supplied with the indicated nutrient concentrations. AsActin from A. sinicus served as an endogenous control for normalization. Data are shown as means ± SD of three biological replicates. Averages with the different letters are significantly different at p < 0.05, based on Duncan’s multiple range test. (e–j) Histochemical localization in pAsZIP2::GUS transgenic A. sinicus hairy roots treated with different Pi and Zn concentrations. Positive GUS staining of the primary (e,g,i) and lateral (f,h,j) roots treated with 30 or 300 µM phosphate, grown under low Zn (0.5 µM) or moderately high Zn (50 µM) conditions described in each panel. (e) The primary root with a primordium; (g,i) The strong GUS staining of the central cylinder of primary roots. (h) The GUS signal is also present in the root epidermis, cortex, and primordium during Pi starvation. Scale bars, 100 µm.

Figure 4

Expression profiles of the two PHT1 family members AsPT2/3 and one ZIP family member AsZIP2 in A. sinicus in response to Pi and/or Zn availabilities. A. sinicus plants were grown in cultures treated with 300 µM Pi and 50 μM Zn (+Pi+Zn), 300 µM Pi and 0.5 μM Zn (+Pi-Zn), 30 µM Pi and 50 μM Zn (-Pi+Zn) or 30 µM Pi and 0.5 μM Zn (-Pi-Zn). (a–f) Fourteen-day-old A. sinicus seedlings were colonized by R. irregularis at 42 dpi, and roots and shoots of 56-d-old plants were collected separately with transcription levels of target genes quantified by real-time qRT-PCR. (a–d) Gene expression in NM A. sinicus plant tissues; (e,f) Gene expression in A. sinicus plants under AM conditions. The AsActin gene for A. sinicus was used as the house-keeping gene for normalization. Error bars mean standard deviation from three biological replicates. The different letters are statistically significant differences among treatments at p < 0.05, based on Duncan’s multiple range test. (g–i) GUS staining of pAsZIP2::GUS transgenic NM (g) and AM (h,i) roots of A. sinicus grown under low Pi and/or moderately high Zn conditions as indicated. A. sinicus roots colonized by R. irregularis at 42 dpi. NM, nonmycorrhizal; AM, arbuscular mycorrhizal; ac, arbuscule-containing cells; eh, extraradical hyphae; v, vesicles. Scale bars, 100 µm.

Figure 5

Arbuscular mycorrhizal colonization reduces the expression of Zn transporter gene AsZIP2 in a phosphate-dependent manner. (a–c) Mycorrhization strongly represses the AsZIP2 expression in A. sinicus roots inoculated with (AM) or without (NM) R. irregularis. (a,b) The mycorrhization in A. sinicus roots colonized by R. irregularis, grown under low Pi (30 µM) conditions at different weeks post-inoculation (wpi). (a) Mycorrhizal colonization levels were determined after WGA488 staining. F%, the total colonization frequency; M%, the percentage of mycorrhizal intensity; A%, the percentage of arbuscule abundance. (b) Fluorescence microscope images of R. irregularis within roots at different wpi. Scale bars, 100 µm. (c) The expression of AsZIP2 in NM and AM roots of A. sinicus at different weeks. NM, nonmycorrhizal; AM, arbuscular mycorrhizal. (d–g) The expression pattern of AsZIP2 in A. sinicus plants is dependent on the phosphate status during AM symbiosis. Total P concentrations were determined in the roots (d) and shoots (e) of NM and AM A. sinicus grown under the indicated Pi conditions shown. (f) The expression of AsZIP2 in AM roots and shoots of A. sinicus in response to different Pi concentrations indicated, estimated by real-time qRT-PCR. AsActin from A. sinicus served as the internal control. (g) The effect of different Pi availability on arbuscular mycorrhization in A. sinicus roots at 42 dpi. (h–k) The transcription of AsZIP2 is independent of the Zn status in A. sinicus during AM symbiosis. Zn concentrations were estimated in the roots (h) and shoots (i) of NM and AM A. sinicus during different Zn conditions indicated. Transcription of AsZIP2 in the roots (j) and shoots (k) of NM and AM A. sinicus exposed to different Zn concentrations indicated, measured by real-time qRT-PCR. Error bars represent the SD for means of three biological replicates. Different letters indicate statistically significant differences at p < 0.05, based on the Duncan’s multiple range test.

Figure 6

Complementation of Saccharomyces cerevisiae mutant zrt1zrt2 and wild-type strain BY4741. Yeast cells expressing AsZIP2 and AtZIP2 in the vector pFL61 were grown under different Zn concentrations as indicated. (a) The wild-type strain BY4741 transformed with pFL61 and Zn (zrt1zrt2) transport mutant carrying AtZIP2 were used as the positive controls, while the mutant zrt1zrt2 carrying pFL61 was applied as a negative control. The mutant zrt1zrt2 transformed with AsZIP2 was used for validation. Transformed cells were grown on SD medium without uracil (SD/-ura) and plus 1 mM EDTA and 5, 50, or 250 µM ZnCl₂. The gradient marks above each panel indicate cell dilutions from 10⁻¹ to 10⁻⁴. For tests, 5-µL drops of each dilution was spotted on solid SD/-ura media and grown for 2 days at 30 °C. (b–d) Zn concentrations in the yeast mutant zrt1zrt2 strain expressing AsZIP2, AtZIP2, or pFL61 vector and in wild-type BY4741 with pFL61, grown under 5 (b), 50 (c), or 250 µM (d) ZnCl₂ conditions. The Zn concentrations in yeast cells grown in SD/-ura media as shown in Figure 6b–d. The data are expressed as µg Zn per g cell dry weight. Error bars represent the SD for means of three independent experiments. Different letters indicate significant differences at p < 0.05, based on the Duncan’s multiple range test.

Figure 7

Subcellular localization of the AsZIP2 transporter in plant cells. (a,b) Confocal laser scanning microscopy images of N. benthamiana leaf epidermal cells transiently coexpressing either 35S::eGFP (a) or 35S::AsZIP2::eGFP (b) with the plasma membrane marker CERK1::DsRed driven by the 35S promoter. Left panel (I): GFP channel; center panel (II): mCherry channel; right panel (III): Merged. Scale bar, 50 μm.

Figure 8

Zinc (Zn) concentration in transgenic A. sinicus lines overexpressing AsZIP2 (AsZIP2-OE lines) and in control lines grown under Pi starvation or moderately high Zn supply. (a,b) Molecular phenotypes in the transgenic A. sinicus roots during Pi starvation (30 µM) or moderately high Zn (50 µM ZnCl₂) conditions. Real time qRT-PCR analysis of AsZIP2 expression in control and OE lines. AsZIP2-OE-1 to AsZIP2-OE-11 represented independent transgenic lines. AsActin served as the endogenous control. (c) The Zn concentrations in the transgenic roots of 56-d-old A. sinicus plants grown in sand cultures exposed to Pi starvation (30 µM). (d) The Zn concentrations in the transgenic roots of 56-d-old A. sinicus plants grown under moderately high Zn concentration (50 µM ZnCl₂). Error bars represent the SD for means of three technical replicates. Significant differences between the AsZIP2-OE lines and controls: ***, p< 0.001; **, p< 0.01; *, p< 0.05; Student’s t-test.

Figure 9

Zn concentration in transgenic A. sinicus lines silencing AsZIP2 (AsZIP2-RNAi lines) and in control lines grown under Pi starvation or moderately high Zn supply. (a,b) Real time qRT-PCR analysis of AsZIP2 transcript levels in the control and RNAi lines of A. sinicus grown under Pi starvation (30 µM) or moderately high Zn (50 µM ZnCl₂) conditions. AsZIP2-RNAi-5′UTR-1 to AsZIP2-RNAi-5′UTR-7 represented independent transgenic lines. AsActin served as the control. (c) The Zn concentrations in the control and AsZIP2-RNAi roots of 56-d-old A. sinicus plants grown in cultures exposed to Pi starvation (30 µM). (d) The Zn concentrations in the control and AsZIP2-RNAi roots of 56-d-old A. sinicus plants grown under the moderately high Zn concentration (50 µM ZnCl₂). Error bars represent the SD for means of three technical replicates. Significant differences between the AsZIP2-RNAi lines and controls: **, p < 0.01; *, p < 0.05; Student’s t-test.

Figure 10

Proposed working model in which the AsZIP2 zinc transporter is repressed by phosphate and arbuscular mycorrhizal symbiosis in A. sinicus under moderately high Zn conditions. (a) Under Pi starvation, low intracellular Pi availability induces AsZIP2 expression, possibly dependent on the potential AsPHR-P1BS pathway in A. sinicus, due to the presence of P1BS (GNATATNC) motif in the promoter region of AsZIP2 (see Figure S3); meanwhile, high Zn supply results in low Pi concentration in A. sinicus (see Figure 1a), and consequently promotes the transcription of AsZIP2 to facilitate the activation of Zn transporter AsZIP2 in the plasma membrane. Both the processes lead to the over-accumulation of Zn²⁺ in plant cells. (b) Under high Pi conditions, high Pi availability reduces AsZIP2 expression via the potentially inactivated AsPHR-P1BS module. On the other hand, arbuscular mycorrhization significantly increases cellular Pi concentration, which can also repress the AsZIP2 expression in a similar manner, leading to less Zn²⁺ accumulation in A. sinicus. The solid arrows present the positive influences or interactions, whereas the flat-ended lines indicate the negative influences or interactions; the question marks indicate a non-confirmed AsPHR-P1BS module that possibly controls AsZIP2 expression in A. sinicus.