Introduction

Silicon is not only beneficial for plant growth, but also improves plant resistance to biotic and abiotic stresses (Ahire et al. 2021), a corresponding transporter is required to function for the uptake of silicon in plants. Low silicon 1 (Lsi1), which is responsible for silicon uptake, is reported for the first time in rice (Ma et al. 2006). It is a member of the Nod26-like membrane intrinsic proteins (NIP III) subfamily and has two conserved NPA motifs and an ar/R selective filter (Ma and Yamaji 2015; Yamaji and Ma 2021; Konishi et al. 2023). With the deeper resolution of its structure, it was found that the pore size of the selectivity filter, which consists of G, S, G, and R, is capable of permeabilizing silicic acid, arsenic acid, and boric acid (van den Berg et al. 2021; Sharma et al. 2024). Up till now, Lsi1 has been identified in both monocots and dicots, such as rice (Ma et al. 2006; Yamaji et al. 2007), barley (Chiba et al. 2009; Yamaji et al. 2012), maize (Mitani et al. 2009), bamboo (Geng et al. 2022), regrass (Pontigo et al. 2021), wheat (Montpetit et al. 2012), pumpkin (Mitani-Ueno et al. 2011; Mitani et al. 2011), cucumber (Sun et al. 2017; Wang et al. 2015), and tomato (Sun et al. 2019). However, most of the studies only focused on the isolation and functional validation of Lsi1, and the regulatory and expression mechanism of Lsi1 gene is still unclear.

Some transgene researches have verified the function of Lsi1 by 35S promoter, such as CsLsi1, TaLsi1, PeLsi1-1, and PeLsi1-2 (Sun et al. 2017; Montpetit et al. 2012; Geng et al. 2022). So far, only the rice OsLsi1 promoter elements have been analyzed (Mitani et al. 2016). Recently, it was found that HMGI transcription factors can bind to the promoter region of OsLsi1 to up-regulate the expression of the Lsi1 gene and further enhance the rice chilling tolerance (Li et al. 2022). Therefore, the study of the upstream sequence of the Lsi1 gene is a guide to the in-depth understanding of the expression regulation mechanism of the Lsi1 gene.

Phyllostachys edulis (moso bamboo) is a perennial woody grass plant with high ecological, economic, and social values (Peng et al. 2013). Similar to rice, moso bamboo is also a high-silicon accumulation plant (Lux et al. 2003). Recently, we identified two Lsi1 genes in bamboo genome, namely PeLsi1-1 and PeLsi1-2, and confirmed that PeLsi1-1 and PeLsi1-2 can transport silicon (Geng et al. 2022). However, characteristics and functions of the Lsi1-1 and Lsi1-2 promoters remain unknown. In this study, we cloned the promoters of Lsi1-1 and Lsi1-2 and analyzed their cis-regulatory elements. Furthermore, promoter functions were identified through 5-end deletion analysis in transgene Arabidopsis plants.

Materials and methods

Plant material and growth condition

Moso bamboo seeds (collected form Guangxi Zhuang Autonomous Region, China) were first soaked in ddH2O for 1 day then soaked for 2 h with 200 mg/L gibberellin acid (Sigma, America). Seeds were planted in nutrient soil (peat soil: vermiculite: perlite = 3:1:1) using the soil culture method and kept in green house under 16 h light/8 h dark cycle and maintained at 26 ± 2 ℃.

Seeds of Arabidopsis thaliana (Columbia) were sterilized using 75% ethanol for 5 min, and then washed with 100% ethanol, followed by transfer to sterile filter paper. After the sterile filter paper was blown dry naturally, the seeds were evenly sown and cultured on 1/2 Murashige–Skoog (MS) medium for 10 days. Then, Arabidopsis seedlings were transferred to nutrient soil (Vermiculite: peat moss: perlite = 3:3:1) cultured at 23 ± 2 ℃ under 16 h light/8 h dark cycles.

Cloning and analysis of the PeLsi1-1 and PeLsi1-2 promoters

Moso bamboo (30-d-old) genomic DNA was isolated by Cetyltrimethylammonium Bromide (CTAB) method (Doyle and Doyle. 1987). The promoter sequences of PeLsi1-1 and PeLsi1-2 genes were identified from the moso bamboo database, and special primers (Table S1) were designed to amplify the promoters. The PCR program started at 95 ℃ for 2 min, then entered into 35 cycles of 95 ℃ for 30 s, 60 ℃ for 20 s and 72 ℃ for 30 s, and then 72 ℃ for 5 min. PCR products were isolated by 1% agarose gel and purified with DNA purification kit (Biomiga, Beijing), then the purified fragments were ligated with pMD-19 T vector (TaKaRa, Beijing), and transformed into E.coli. The transformed bacterial liquid was sent for sequencing. Cis-regulatory elements in cloned promoter sequences were analyzed based on the online database PlantCARE (Lescot et al. 2002) and New PLACE (Higo et al. 1999). Transcription Start Site (TSS) was predicted by TSSPlant (Shahmuradov et al. 2017).

Construction and genetic transformation of plant expression vectors

PeLsi1-1 (1910 bp) and PeLsi1-2 (1922 bp) promoter sequences were amplified by PCR with special primers carrying XbaI and NcoI restriction sites. Four 5′-end deleted fragments of PeLsi1-1 promoter (PeLsi1-1-P1 ~ P4) and five 5′-end deleted fragments of PeLsi1-2 promoter (PeLsi1-2-P1 ~ P5) were amplified by the same way. ProLsi1-1 and ProLsi1-2 promoter sequences were amplified by PCR with special primers carrying HindIII restriction site. Lsi1-1 and Lsi1-2 sequences were amplified by PCR with special primers carrying KpnI restriction site. All primers listed in Table S1 and TransStart® FastPfu DNA Polymerase (Trans, Beijing) were used in every PCR reaction. XbaI and NcoI endonuclease were used to cut down the region of CaMV35S promoter of pCAMBIA1301 and obtain linear plasmid. HindIII and KpnI endonuclease were used to cut down the region of CaMV35S promoter of pBI121 and obtain linear plasmid.

A seamless cloning kit (Trans, Beijing) was used to connect fragments and linear plasmid. All recombinant plasmids were transformed into E.coli, and verified by bacteriophage PCR. Different plasmids were extracted by Plasmid MiniPrep Kit (Trans, Beijing) and verified by further double digestion. The recombinant plasmids were named PeLsi1-1-P0/P2/P3/P4::GUS, PeLsi1-2-P0/P1/P2/P3/P4/P5::GUS, pPeLsi1-1 -Lsi1-1::EGFP and pPeLsi1-2-Lsi1-2::EGFP, respectively.

All recombinant plasmids were transformed into GV3101 Chemically competent cell (Zomanbio, Beijing) by freeze–thaw method, and then transformed into Arabidopsis thaliana by floral-dip method (Clough and Bent 1998). The Arabidopsis transformed with empty plasmids was used as control. All transgenic plants were selected on 1/2 MS medium added with 15 mg/L hygromycin or 50 mg/L kanamycin. DNAs of putative seedlings were extracted through CTAB methods and verified by PCR using primers (in TableS1). At least three independent transgenic lines were established and used for analysis. T3 generation transgenic Arabidopsis was used for subsequent experiments.

GUS histochemical analysis

GUS histochemical staining methods were used to confirm promoter driving activity using transgenic and WT Arabidopsis seedlings (2 ~ 14 days) (Jefferson et al. 1987). GUS dye solution contained 10 mM EDTA, 0.1% Triton X-100, 50 mM Phosphate buffer (pH 7.0) and 1 mM X-gluc. Transgenic and WT plants were soaked in GUS dye solution overnight at 37 ℃, and then plants were decolored with 70% ethanol until the wild-type plants were white. The results were observed under the stereomicroscope (SMZ-10, Nikon Company, Japan).

Hormone and abiotic stress treatments

Seeds of transgenic Arabidopsis were grown in 1/2 MS with 15 mg/L hygromycin. After 14 days, the different transgenic seedlings were sprayed with 0.1 mM Abscisic Acid (ABA), 0.1 mM Methyl Jasmonate (MeJA), and 1 mM Salicylic acid (SA), respectively. And the control group was sprayed with equal amounts of distilled water. The different treatments were maintained for 6 h, and all experiments were repeated three times. Each sample was harvested at the appointed time and stored at − 80℃. Similarly, seeds of transgenic Arabidopsis were grown in 1/2 MS for 12 days, then transferred to 1/2 MS solid medium containing 2 mM K2SiO3, 200 mM mannitol, 150 mM NaCl, respectively. Samples were harvested after 2 days of treatment.

Roots of 80-day-old moso bamboo seedlings were rinsed with distilled water and transferred from the soil into Hoagland nutrient for one week of hydroponic acclimatization. Then 90-day-old moso bamboo seedlings were transferred from the soil into Hoagland nutrient solution containing 100 μM ABA, 100 μM MeJA, or 1 mM SA, respectively. Roots, stems, and leaves were sampled at 0 h, 1 h, 3 h, 6 h, 12 h, and 24 h and preserved at − 80 °C, respectively. For salt stress treatment, seedlings of moso bamboo were transferred to Hoagland nutrient solution containing 150 mM NaCl, and for osmotic stress treatment, seedlings were transferred to Hoagland nutrient solution containing 200 mM mannitol. Following treatments, roots, stems, and leaves were sampled at 0 h, 1 h, 3 h, 6 h, 12 h and 24 h, respectively, and then stored at − 80 °C.

Phenotypic analysis of Arabidopsis thaliana transformed with Lsi1 gene driven by different promoters

Seeds of wild-type and T3 transgenic Arabidopsis (35S-PeLsi1-1::EGFP, 35S-PeLsi1-2::EGFP, pPeLsi1-1-Lsi1-1::EGFP and pPeLsi1-2-Lsi1-2::EGFP) were sowed on 1/2 MS medium, left at 4 °C for two days, then transferred to incubation room and incubated for 8 days.

In order to identify the Si uptake capacity of transgenic Arabidopsis, we transferred the above materials to 1/2 MS medium containing 0 mM Si or 2 mM Si, and continued to incubate them for 12 days. There were three lines of four transgenic materials for analysis, each with 30 seedlings for the experiments. We recorded plant fresh weight, root length, and phenotype of Arabidopsis (Geng et al. 2022). To measure the chlorophyll content, 300 mg leaves were fully mixed thoroughly with 3 mL of 95% ethanol. Spectrophotometer (UV752, Jinghua, Shanghai) was used to determine the absorbance values of the extracts at A665 and A649, respectively. Chlorophyll content was calculated following the formula of Arnon, Chl a (mg/g) = (13.95 × A665 − 6.88 × A649) × 3/(0.3 × 1000), Chl b (mg/g) = (24.96 × A649 − 7.32 × A665) × 3/(0.3 × 1000) (Hu et al. 2018). Plant Si content was determined by the molybdenum blue colorimetric method (Ma et al. 2001). Three biological replicates with three repeats were analyzed for each line.

To investigate the effect of Si on the vegetative growth of transgenic Arabidopsis, 7-day-old seedlings were transferred to nutrient soil and cultured for 21 days. The aboveground fresh weight of Arabidopsis, chlorophyll content, leaf water content, and water loss in detached leaf were also measured.

Gene expression analysis by real-time qPCR

RNA was extracted from whole plant using RNA kit (Biorigin, Beijing). One microgram of RNA was used for first-strand cDNA synthesis by cDNA Synthesis SuperMixs (Trans, Beijing). The Arabidopsis thaliana actin2 gene (AtActin2 gene, AT3G18780) and the Phyllostachys edulis tonoplast intrinsic protein 41 gene (TIP41 gene, PH02Gene18883.t1) were used as internal reference genes. RT-PCR was carried out using SYRB Green Qpcr SuperMix (Trans, Beijing) with CFX96 Touch Real-Time PCR system. Two steps were applied in RT-qPCR procedure, starting at 94 °C for 30 s and then entering 40 cycles of 94 °C for 5 s and 60 °C for 30 s, then entering a melting curve. The obtained data were calculated by the 2−ΔΔCt method (Schmittgen and Livak 2008).

Statistical analysis

The measured data were shown as mean ± standard deviation (SD). Data analyses and graphs were performed by Excel 2019 and GraphPad Prism 9. T test and two-way ANOVA were used, and differences were considered significant at P < 0.05.

Results

Isolation and bioinformatic analysis of PeLsi1-1 and PeLsi1-2 promoters

The promoter regions of PeLsi1-1 with the length of 1910 bp and PeLsi1-2 with a length of 1921 bp were isolated by PCR (Fig. S1). PeLsi1-1-pro was 99.79% identical to the genome sequence, only 4 bases had been variated. PeLsi1-2- pro was 98.91% identical to the genome sequence, more than 20 bases had been lost or mutated (Fig. S2). The transcription start sites of both promoters were predicted using TTSPlant, and the results showed that both promoters have three possible transcription start sites. Based on the scores, we believe that the TSS of PeLsi1-1 promoter is G, which was located -106 bp upstream of ATG. The TSS of PeLsi1-2 promoter is also G, located -226 bp upstream of the ATG (Table S2). Cis-elements for promoter regions of PeLsi1-1 and PeLsi1-2 were analyzed (Table 1). Both PeLsi1-1- pro and PeLsi1-2- pro have basic cis-acting element TATA-box, the key element for transcription imitation, and CAAT-box which controls the frequency of transcription initiation. Besides, promoters have some light response elements (e.g., GT1-motif, BoxII, G-box), hormone response elements (e.g., ABRE, CGTCA-motif, TGACG-motif), tissue-specific elements (e.g., OSE2ROOTNODULE, RYREPEATBNNAPA, POLLEN1LELAT52) and abiotic stress response elements (e.g., ARE, MBS, AT-rich sequence). Interestingly, both promoters have comparable numbers of tissue-specific elements, but PeLsi1-2-pro has more cis-acting elements in response to MeJA and specifically has elements responding to salicylic acid and gibberellin. Therefore, the above analysis indicated that hormones, abiotic and biotic stresses might regulate PeLsi1-1 and PeLsi1-2 gene expression.

Table 1 Prediction of cis-acting progenitors in PeLsi1-1 and PeLsi1-2 gene promoter by Plantcare and PLACE
Full size table

Deletion analysis of PeLsi1-1 and PeLsi1-2 promoters in transgenic Arabidopsis

To analyze the function of the promoters, deletion analysis approach was used. According to the prediction of cis-regulation elements, we constructed four 5′end deletion fragments of PeLsi1-1 promoter and five 5′end deletion fragments of PeLsi1-2 promoter (Fig. 1). All designed fragments were successfully ligated to plant expression vectors, and all recombinant vectors could be cut by double digestion to produce target bands with respective sizes (Fig. S3 and S4). By floral-dip method, all recombinant plasmids were transformed into Arabidopsis thaliana. More than three homozygous lines were obtained by screening the T1 generation seeds for hygromycin resistance, and the transgenic Arabidopsis plants were confirmed by PCR (Figs. S5, S6 and S8).

Fig. 1
figure 1

Schematic representation of the 5′ deletion promoter fragment fused to the GUS gene

Full size image

We examined the spatiotemporal expression pattern of GUS driven by promoters using T3 Arabidopsis seedlings. The results showed that both PeLsi1-1-P0 and PeLsi1-2-P0 were able to drive the expression of GUS gene. As shown in Fig. 2a, PeLsi1-1-P4 has almost no promoter activity. Similarly, we found that among the fragments of the PeLsi1-2 promoters, GUS signal staining was very weak when GUS gene was driven by PeLsi1-2-P5. These results imply that PeLsi1-1-P3 ~ P4 are the core regions of the PeLsi1-1 promoter and PeLsi1-2-P4 ~ P5 are the core regions of the PeLsi1-2 promoter. In addition, we found that the signaling of GUS proteins of 8–14-day-old transgenic Arabidopsis seedlings was relatively stable, which can be used for subsequent experimental analyses.

Fig. 2
figure 2

GUS histochemical staining and relative quantitative expression analysis of GUS gene under normal culture condition. a β-Glucuronidase (GUS) staining of transgenic Arabidopsis thaliana at various growth stages to evaluate GUS expression driven by different fragments of PeLsi1-1 and PeLsi1-2 promoters. Scale = 100 μm, 200 μm, 500 μm, 1 mm corresponds to day 2, 5, 8, 11, and 14, respectively. b GUS gene expression driven by different fragments of PeLsi1-1 in 14-day-old Arabidopsis. c GUS gene expression driven by different fragments of PeLsi1-2 in 14-day-old Arabidopsis. Genes were quantified using Actin2 as an internal reference gene and data were computed using the 2−ΔΔCt method, the error bars represent standardized deviation, mean ± SD, n = 3,*, ** and *** represent significant differences at the 0.05, 0.01 and 0.001 levels, respectively

Full size image

To verify the accuracy of GUS histochemical staining, we performed qRT-PCR analysis for GUS gene using 14-day-old transgenic seedlings (Fig. 2b, c). The expression levels of GUS gene driven by PeLsi1-1-P4 and PeLsi1-2-P5 were the lowest among the promoter fragments. The results of real-time PCR were consistent with the GUS histochemical staining. We also found the highest expression levels of GUS gene driven by PeLsi1-1-P3 and PeLsi1-2-P4, which were 4.37 and 3.19 times higher than those of PeLsi1-1-P0 and PeLsi1-2-P0, respectively. It was noteworthy that the expression level of the GUS gene driven by PeLsi1-2-P4 was higher than that driven by the 35S promoter, suggesting that PeLsi1-2-P4 was a highly efficient promoter.

Functional analysis of promoters in response to hormone and abiotic stresses

After ABA treatment, the expression levels of GUS gene driven by PeLsi1-1-P1/P2/P3/P4 and PeLsi1-2-P1 were significantly elevated (Fig. 3a, d). The expression of GUS genes driven by other fragments did not change significantly. The results indicated that the critical regions of PeLsi1-1 and PeLsi1-2 promoters in response to ABA hormone are PeLsi1-1-P3 ~ P4 and PeLsi1-2-P0 ~ P1, respectively.

Fig. 3
figure 3

GUS gene expression driven by different segments of PeLsi1-1 and PeLsi1-2 promoters in Arabidopsis thaliana under 100 μM ABA, 100 μM MeJA and 1 mM SA treatments for 6 h. a ~ c: GUS gene expression driven by different segments of PeLsi1-1 promoter under ABA, MeJA, and SA treatments. d–f GUS gene expression driven by different segments of PeLsi1-2 promoter under ABA, MeJA, and SA treatments. Genes were quantified using Actin2 as an internal reference gene and data were computed using the 2−ΔΔCt method, the error bars represent standardized deviation, mean ± SD, n = 3,*, ** and *** represent significant differences at the 0.05, 0.01 and 0.001 levels, respectively

Full size image

After treatment with MeJA, expression of GUS gene driven by PeLsi1-1-P0/P1/P2/P3/P4 and PeLsi1-2-P0/P1 was at a higher level, whereas the other fragments of the promoters did not work (Fig. 3b, e). These results indicated that PeLsi1-1-P3 ~ P4 and PeLsi1-2-P1 ~ P2 are the key regions of PeLsi1-1 and PeLsi1-2 genes regulated by MeJA.

SA was able to significantly increase the expression level of GUS genes, which was driven by PeLsi1-1-P2/P3 and PeLsi1-2-P1/P2/P3. While SA significantly decrease the expression level of GUS, which was driven by PeLsi1-1-P0/P1 and PeLsi1-2-P0 (Fig. 3c, f). Following the interesting results, we analyzed the SA-related elements in the promoter region and found a SA negative regulatory element WBOXATNPR1 in the regions of PeLsi1-1-P0 ~ P2 and PeLsi1-2-P0 ~ P1; and after deleting this region, the other SA positive regulatory elements then exerted a positive regulatory effect after being subjected to SA. These results indicated that PeLsi1-1-P1 ~ P2 and PeLsi1-2-P0 ~ P1 are the key regions for negative SA regulation, while PeLsi1-1-P3 ~ P4 and PeLsi1-2-P2 ~ P3 are the key regions for positive SA regulation.

We similarly verified the effects of salt stress and osmotic stress on the function of fragments of the promoters of PeLsi1-1 and PeLsi1-2 (Fig. 4). The expression levels of GUS gene driven by PeLsi1-1-P0/P1/P2/P3 and PeLsi1-2-P0/P1 were significantly increased after salt treatment, indicating that PeLsi1-1-P3 ~ P4 and PeLsi1-2-P1 ~ P2 were the key regions in response to salt stress. After osmotic stress, the expression levels of GUS gene driven by PeLsi1-1-P0/P1/P2/P3, PeLsi1-2-P0, and PeLsi1-2-P4 were more markedly elevated compared to the control. The results implied that PeLsi1-1-P3 ~ P4 is the critical region for PeLsi1-1 to respond to osmotic stress, and there are two critical regions for PeLsi1-2 to respond to osmotic stress, PeLsi1-2-P0 ~ P1 and PeLsi1-2-P4 ~ P5.

Fig. 4
figure 4

GUS gene expression driven by different segments of PeLsi1-1 and PeLsi1-2 promoters in Arabidopsis under 150 mM NaCl, 200 mM mannitol and 2 mM K2SiO3 treatments for two days. a–b GUS gene expression driven by different segments of PeLsi1-1 promoter under NaCl, mannitol, and K2SiO3 treatments. c–d GUS gene expression driven by different segments of PeLsi1-2 promoter under NaCl, mannitol, and K2SiO3 treatments

Full size image

Identification of the core region of the PeLsi1-1 and PeLsi1-2 promoters in response to Si

PeLsi1-1 and PeLsi1-2 were down- regulated by 2 mM Si (Geng et al., 2022), and we also used 2 mM K2SiO3 to treat Arabidopsis to verify whether exogenous silicon can regulate the activity of the PeLsi1-1 and PeLsi1-2 promoters. We found that the expression levels of GUS genes in Arabidopsis transformed with PeLsi1-1-P0/P1/P2/P3 were down-regulated by Si, whereas the expression levels of GUS gene driven by PeLsi1-1-P4 were not significantly affected by Si treatment (Fig. 4c). Therefore, the PeLsi1-1-P3 ~ P4 is the key region regulated by exogenous Si.

Similarly, the expression levels of GUS gene in PeLsi1-2-P1/P2/P3/P4 Arabidopsis transformed with PeLsi1-2-P1/P2/P3/P4 were down-regulated by Si. However, the expression level of GUS gene was not affected by Si in Arabidopsis transformed with PeLsi1-2-P5 (Fig. 4f), indicating that the PeLsi1-2-P4 ~ P5 region is the critical region in response to Si.

Expression profile of PeLsi1-1 and PeLsi1-2 gene in response to abiotic stresses

To better understand the expression of PeLsi1-1 and PeLsi1-2 genes under different abiotic stresses, we first analyzed the expression of PeLsi1-1 and PeLsi1-2 genes in different tissues of moso bamboo seedlings. The results showed that PeLsi1-1 and PeLsi1-2 genes were mainly expressed in roots, while the expression levels in stems and leaves were extremely low (Fig. 5a, b). So in the subsequent experiments, we focused on analyzing the expression of PeLsi1-1 and PeLsi1-2 genes in roots under hormone, salt stress, and osmotic stress. The results showed that ABA was able to down-regulate the expression of PeLsi1-1 gene. SA and salt stress were able to down-regulate the expression of PeLsi1-2 gene, while the expression of PeLsi1-1 and PeLsi1-2 showed dynamic changes under treatment of other hormones or stresses (Fig. 6). Therefore, we hypothesized that PeLsi1-1 and PeLsi1-2 genes are able to respond to multiple abiotic stresses, which is in accordance with the presence of cis-acting elements in the promoter region.

Fig. 5
figure 5

Expression patterns of PeLsi1-1 and PeLsi1-2 genes in different tissues and responding to abiotic stress of moso bamboo seedlings. a–b Expression patterns of PeLsi1-1 and PeLsi1-2 in different tissues. c–d Expression of root PeLsi1-1 gene after treatment with 150 mM NaCl and 200 mM mannitol, respectively. e–f Expression of root PeLsi1-2 gene after treatment with 150 mM NaCl and 200 mM mannitol, respectively. Genes were quantified using TIP41 as an internal reference gene and data were computed using the 2.−ΔΔCt method. The error bars represent standardized deviation, mean ± SD, n = 3, and different alphabets indicate significant differences at P < 0.05

Full size image
Fig. 6
figure 6

Expressions of PeLsi1-1 and PeLsi1-2 genes in roots of moso bamboo seedlings under different hormone treatments. a–c The expression of PeLsi1-1 gene after treatment with 100 μM ABA, 100 μM MeJA and 1 mM SA, respectively. e–f The expression of PeLsi1-2 gene after treatment with 100 μM ABA, 100 μM MeJA and 1 mM SA, respectively. Genes were quantified using TIP41 as an internal reference gene and data were computed using the 2−ΔΔCt method. The error bars represent standardized deviation, mean ± SD, n = 3, different alphabets indicate significant differences at P < 0.05

Full size image

Si accumulation in Arabidopsis transformed with Lsi1 driven by different promoters

We identified 10 strains of each T1 generation, and performed genomic PCR and RT-PCR using the T2 generation of these 10 strains. Since there were no significant phenotype differences among the 10 strains, we selected three of them for subsequent experiments based on their expression levels (Fig S9).

When grown in Si-free medium, Arabidopsis plants transformed with 35S-PeLsi1-1, 35S-PeLsi1-2, pPeLsi1-1-Lsi1-1, and pPeLsi1-2-Lsi1-2 were not phenotypically different from WT. There were also no significant differences in root length, chlorophyll content, and plant fresh weight. When grown in medium supplemented with 2 mM K2SiO3, Arabidopsis plants transformed with 35S-PeLsi1-1 and 35S-PeLsi1-2 showed a dwarf phenotype, and their root length, chlorophyll content, and plant fresh weight were significantly lower than WT (Fig. 7). In contrast, the phenotype of Arabidopsis transformed with pPeLsi1-1-Lsi1-1 and pPeLsi1-2-Lsi1-2 did not differ from the WT, and root length, chlorophyll content, and plant fresh weight were also not significantly different (Fig. 7). Interestingly, the transgenic Arabidopsis increased its silicon accumulation capacity by about threefold under 2 mM K2SiO3 treatment.

Fig. 7
figure 7

Growth and silicon accumulation in wild and transgenic Arabidopsis. a Phenotypes of T3 homozygous transgenic Arabidopsis treated with 0 or 2 mM K2SiO3. Scale bars = 2 cm. b Si content; c root length; d whole-plant fresh weight; e chlorophyll content. Values are presented as the mean ± SD of three independent biological replicates, different letters indicate significant differences (P < 0.05)

Full size image

Discussion

Lsi1 genes and their promoters are induced by various factors

Lsi1 functions as an influx silicon acid transporter in silicon-accumulating plants, which is a member of the NIP III subfamily and was originally identified in rice (Ma et al. 2006). Recently, we identified the silicon influx transporter genes PeLsi1-1 and PeLsi1-2 in moso bamboo (Geng et al. 2022). Nevertheless, the regulation mechanism of Lsi1 gene expression is not clearly defined. In this study, about 1.9 kb of PeLsi1-1 and PeLsi1-2 promoters were cloned from moso bamboo. Bioinformatics analysis revealed that both promoters have basic promoter elements, such as TATA box and CAAT box, indicating that the promoters have transcriptional potential (Porto et al. 2014). GUS staining and RT-qPCR results indicated that PeLsi1-1-P3 and PeLsi1-2-P4 fragments had the highest activity among all promoter fragments. Both fragments contain at least six CAAT box elements, which are enhancer elements that can increase the ability of promoters to drive gene expression (Andersson and Sandelin 2020).

In rice, OsLsi1 expression was down-regulated by ABA and dehydration stress, and there exists ABRE elements in the promoter region of Lsi1 (Yamaji and Ma. 2007). We found that a certain number of cis-acting elements in the promoter region of the Lsi1 gene are present in different species (Fig S7), and the presence of these elements is related to the regulation of the Lsi1 (NIPIII) gene by a variety of factors (Zhu et al. 2019; Zhang et al. 2022). Through 5′ deletion analysis of the promoters, we identified PeLsi1-1-P3 ~ P4 (− 441 bp ~ − 236 bp) as the critical region in response to ABA, MeJA, osmotic stress and salt stress, PeLsi1-2-P1 ~ P2 (− 914 bp ~ − 791 bp) as the critical region in response to ABA, MeJA and salt stress, and PeLsi1-2-P0 ~ P1 (− 914 bp ~ − 791 bp) and PeLsi1-2- P4 ~ P5 (− 533 bp ~ − 199 bp) were critical regions in response to osmotic stress (Figs. 3 and 4). We combined the predicted results of the cis-acting elements of the PeLsi1-1 and PeLsi1-2 promoter regions and found that ABRE is the key element in response to ABA (Nakashima et al. 2009), CGTCA or TGACG motif are key regions in response to MeJA (Chen and Chen. 2000), elements, such as GT1GMSCAM4, DRE2COREZMRAB17, or DRECRTCOREAT, exist as key regions in response to salt stress (Park et al. 2004; Dubouzet et al. 2003), MBS, MYCATRD22, LTRECOREATCOR15 and other elements are present in the critical region for osmotic stress response (Abe et al. 1997; Kim et al. 2002). SA was able to reduce the expression level of GUS gene driven by PeLsi1-1-P0/P1 and PeLsi1-2-P0, but after deletion of these segments, GUS gene driven by other segments were up-regulated by SA, implying that PeLsi1-1-P1 ~ P2 (− 1921 bp ~ − 914 bp) and PeLsi1-2-P0 ~ P1 (− 672 bp ~ − 533 bp) were the key regions negatively regulated by SA (Fig. 3c, f). By analyzing the missing sequences, we found that WBOXATNPR1 in PeLsi1-1-P1 ~ P2 and PeLsi1-2-P0 ~ P1 might be a negative regulatory element responding to SA (Chen and Chen 2002). Consistently, the expressions of PeLsi1-1 and PeLsi1-2 in roots of moso bamboo seedlings were also regulated by ABA, MeJA,SA, NaCl, and mannitol (Figs. 5 and 6), implying that different elements in the promoters mediate the expression of PeLsi1-1 and PeLsi1-2 responding to different stimuli factors, respectively.

Regulation of Lsi1 expression involved in silicon accumulation

The Lsi1 gene is stably expressed under normal culture conditions, but significant changes in Lsi1 gene expression occur after artificially applying higher concentrations of silicon (Wang et al. 2015; Deshmukh et al. 2013; Vulavala et al. 2016). Especially, the graminoids rice and ryegrass showed a significant decrease of Lsi1 gene expression after short-term treatment of high Si concentration (Pontigo et al. 2021; Xu et al. 2024). Our previous research also found that the expression levels of PeLsi1-1 and PeLsi1-2 genes were significantly decreased after short-term Si treatment and the motility of PeLsi1-1 in the cytoplasmic membrane was reduced as detected by super-resolved microscopy, whereas the expressions of the two genes were significantly increased after long-term Si treatment (Geng et al. 2022). The region between − 327 and − 292 in the OsLsi1 promoter is involved in Si-mediated regulation of rice OsLsi1 expression (Mitani et al. 2016). In this study, we also found that short-term high silicon concentrations suppressed the expression of GUS gene driven by PeLsi1-1 or PeLsi1-2 promoters and identified PeLsi1-1-P3 ~ P4 (− 441 bp ~ − 236 bp) and PeLsi1-2-P4 ~ P5 (− 533 bp ~ − 199 bp) as key regions in response to exogenous Si (Fig. 4c, f).

When wheat TaLsi1 was overexpressed in Arabidopsis, a species with a very low innate Si uptake capacity, resulted in a fourfold increase in Si accumulation in transgene plants compared to WT. However, this Si absorption caused deleterious symptoms (Montpetit et al. 2012). In contrast, in this study, compared with Arabidopsis transformed with 35S-Lsi1-1 and 35-Lsi1-2, plants transformed with pPeLsi1-1-Lsi1-1 and pPeLsi1-2-Lsi1-2 accumulated a higher concentration of Si and did not show deleterious phenotype (Fig. 7). To understand the above complex physiological and biochemical phenomena, we carried out a working model (Fig. 8). In plants with Si active transporters, when the roots were exposed to higher Si concentration, conformation and movement of Lsi1 undergo a rapid transformation which Lsi1 may serve as a Si receptor to transmit Si signal to nucleus leading to the down-regulation of Lsi1 gene expression (Fig. 8a). With the prolongation of high silicon exposure, silicon is rapidly transported by roots and distributed in stems and leaves (Ma et al. 2011; Ma and Yamaji. 2015; Mitani and Ma. 2021). To meet the requirement of Si accumulation (e.g., phytosilicon) in some plants, an unknown long-distance signaling from stems and leaves transmits to roots to up-regulate the expression of silicon transporter genes, accelerating the absorption, transport and accumulation of silicon in the aboveground parts (Fig. 8b). The cis-regulatory elements present in the native promoter might regulate the expression of Lsi1 by receiving both cell receptor-dependent or long-distance feedback signals, allowing the plant to accumulate appropriate amount of silicon for proper growth. The present study, thus, for the first time helps demonstrate that the native moso bamboo PeLsi1 promoters are promising genetic elements in transgenic approaches.

Fig. 8
figure 8

A proposed working model for Lsi1 gene expression regulation. a Lsi1 gene expression regulated by short-term high silicon concentration. b Lsi1 gene expression regulated by long-term high silicon concentration. The red line represents inhibition and the blue line represents facilitation (colour figure online)

Full size image

Conclusion

In summary, PeLsi1-1 and PeLsi1-2 promoters have the key regions related to ABA, MeJA, SA, salt stress, and osmotic stress. Accordingly, the expression of PeLsi1-1 and PeLsi1-2 can be regulated by multiple hormones and abiotic stresses. Moreover, exogenous silicon can negatively regulate the PeLsi1-1 and PeLsi1-2 promoters. The promoters of PeLsi1-1 and PeLsi1-2 were more effective than the 35S promoter in driving the expression of PeLsi1-1 or PeLsi1-2 in Arabidopsis, leading to increased Si uptake. A proposed model was built to explain the mechanism of plant Si uptake and accumulation.