Cca-miR398 increases copper sulfate stress sensitivity via the regulation of CSD mRNA transcription levels in transgenic Arabidopsis thaliana

Copper sulfate stress assays in hickory

Hickory seeds were germinated indoors under artificial climate conditions (25 °C, 100 µmol photons m⁻²s⁻¹, 60% relative humidity, 16/8 day/night cycles). The 30-day-old hickory roots were subjected to the following stress treatment conditions: 0, 100 and 200 mM copper sulfate. 10 days following treatment, the lengths of 30 hickory roots from each treatment group were surveyed and photographed. Roots were then reaped 10 days after treatment, and the samples were stored immediately at −80 °C for RNA isolation. The expression patterns of cca-miR398 and CSDs (Table S1) were then examined by qRT-PCR.

Cca-miR398 constructs and the generation of transgenic Arabidopsis

The precursor of cca-miR398 was PCR-amplified from the hickory roots. The cca-miR398-overexpression vector was made based on the vector pMD19-simple of T-Vector pMD™ 19 kit (TaKaRa, China). A fragment of 400 bp mRNA flanking the miRNA sequence (Table S2), including the fold-back motif, was inserted into the vector. The precursor of cca-miR398 was ligated into pCAMBIA13011, which was pre-digested by XbaI and KpnI. The resulting construct (containing pCAMBIA13011 and cca-miR398 precursor) was ligated with T4 DNA ligase, transferred into an agrobacterium tumefaciens EHA105 strain, and then used to infect an Arabidopsis ecotype “Columbia” (Col-0) via the floral dipping method (Mara, Grigorova & Liu, 2010). Transgenic lines were subsequently selected with hygromycin B and confirmed with PCR amplification.

Copper sulfate stress assays in transgenic and WT Arabidopsis

Primary Arabidopsis transformants were selected on Murashige-Skoog (MS) medium (Murashige & Skoog, 1962) supplemented with 50 mg L⁻¹ hygromycin B. For seed germination root length determination, the seeds were sterilized and placed on three varieties of MS agar medium supplemented with different concentrations of copper sulfate (40, 80, and 160 mM). The seeds were then germinated under artificial climate room conditions (25 °C, 100 µmol photons m⁻²s⁻¹, 60% relative humidity, 16/8 day/night cycles). The germination rate was counted, and pictures were taken on the seventh day. Five-day-old seedlings were separately transferred to medium supplemented with copper sulfate (40, 80, and 160 mM). After eight days of growth, the root lengths were surveyed and photographed.

For copper sulfate tolerance assessment, transgenic seeds and WT seeds were sown in pots for four weeks. The seeds were grown under artificial climate room conditions (25 °C, 100 µmol photons m⁻²s⁻¹, 60% relative humidity, 16/8 day/night cycles). During the fourth week, plants were irrigated with various concentrations copper sulfate (0, 500, 1,000, 1,500, and 2,000 mM). After 15 days of treatment, related physiological indicators were detected and leaves were respectively harvested from each treatment group before being stored at −80 °C until RNA isolation. Expression patterns of candidate genes (Table S1) were then examined by qRT-PCR after RNA isolation and first-strand cDNA synthesis.

Expression analysis by qRT-PCR

RNA samples extracted from Arabidopsis leaves and hickory roots were subjected to qRT-PCR. The transcription of miRNA of the Arabidopsis and hickory was completed using a Mir-XTM miRNA First-Stand Synthesis Kit (TaKaRa, China). The reactions were performed with a CFX96 Real-time PCR Detection System (BioRad, U.S.A) using a 96-well format. The qRT-PCR process consisted of one cycle of 30 s at 95 °C followed by 40 cycles of 95 °C for 5 s and 60 °C for 34 s. All reactions included nine replicates (three biological triplicates and three technical triplicates). Data were analyzed with CFX96 system software (BioRad, U.S.A), and the relative gene expression levels in various tissues were calculated based on the 2^−ΔΔCT method (Livak & Schmittgen, 2001). The actin and 5.8S rRNA were used as a reference gene for CSD genes and miRNA, respectively. The qRT-PCR primers are listed in Table S3.

Measurements of physiological indicators

Physiological indicators refer to the activities of SOD, peroxidase (POD), and catalase (CAT) in Arabidopsis leaves. These activities were determined using respective SOD, POD, and CAT assay kits (A001-2, Nanjing, China). An IMAGING-PAM chlorophyll fluorometer (Heinz Walz GmbH, Effeltrich, Germany) was used to record the fluorescence intensity readout (Shao et al., 2008). A portion at each leaf center with a 1-cm diameter was selected to provide the measurements. The means of chlorophyll fluorescence parameters, including fluorescence images of the maximal Y(II) quantum yield (Fv/Fm), an index of the photochemical efficiency of effective PSII quantum yield system (PSII), the electron transport rate (ETR), non-photochemical quenching (NPQ), and the coefficient of photochemical quenching (qP), were calculated. Fv/Fm fluorescence images were derived simultaneously from the IMAGING-PAM software.

Statistical analysis

Statistical analyses were performed by using SPSS version 13.0. Differences were analyzed based on a one-way ANOVA model followed by Tukey’s range test.

Morphological effects in hickory roots under copper sulfate stress

After 10 days of copper sulfate stress treatment, hickory root development was observed to be significantly different among the three concentrations. Root growth was reduced seriously by increasing the concentration of copper sulfate (Fig. 1A), and more copper sulfate was absorbed into the hickory root in the 200 mM group than in the 100 mM group (Fig. 1B). Relative expression of cca-miR398a significantly decreased (P < 0.01) with increased concentrations of copper sulfate (Fig. 1C).

Sequence comparison and gene expression of CSDs in hickory roots

CSD genes of Arabidopsis, poplar, grape, rice, sorghum, and corn were downloaded from Phytozome v 10.3 (http://phytozome.jgi.doe.gov/pz/portal.html). A phylogenetic comparison of their sequences with hickory CSDs is shown in Fig. 1D. Hickory CSDs were identified using the database with respect to the corresponding class (CSD1, CSD2, and CSD3), and were distributed into three subfamilies, as shown in Fig. 1D. The cluster analyses indicated that they were derived from a common ancestor, although CSD2 and CSD3 appeared to be more closely related, both evolutionarily and structurally. Thus, three CSD-like genes (Unigene048628, Unigene032279, and Unigene032278) were predicted to be hickory orthologues of the CSD family in Arabidopsis (Fig. 1D). The qRT-PCR analysis was used to investigate whether CSD genes were involved in the regulation of abiotic stress in hickory roots. Transcript levels of CSDs in the roots significantly increased (P < 0.01) under the copper sulfate treatments and the expression patterns of CSDs were similar to each other (Fig. 1E).

Sequence comparison of cca-miR398 with Arabidopsis miR398 family members

To analyze the functions of cca-miR398, a sequence comparison of miR398 family members from hickory and Arabidopsis was performed. As shown in Fig. 1F, the mature sequence of cca-miR398 had only one nucleotide variation (at position 15) compared to miR398b and c sequences of Arabidopsis. Variation with two nucleotides was observed (at positions 15 and 21) between cca-miR398 and miR398a (Fig. 1F).

In an Arabidopsis model plant, miR398 has been shown to target two CSD genes, CSD1 and CSD2 (Li et al., 2017; Mermod et al., 2019). In Arabidopsis CSD2 mRNA, a complementarity site of miR398 is in the coding sequence (Li et al., 2017; Mermod et al., 2019), which is identical in hickory CSD2 (Fig. 1F). In Arabidopsis CSD1 mRNA, the complementarity site of miR398 is in the 5′-UTR and contains an intron interruption (Li et al., 2017; Mermod et al., 2019).

The sequence of 5′-UTRs in CSD1 and CSD3 are still not available in current hickory database, making a detailed comparison of the 5′-UTR sequences impossible. However, based on the high level of sequence similarity in coding regions, hickory CSD1 and CSD3 may also contain the miR398 complementarity site in their 5′-UTRs.

Influences of copper sulfate on phenotypes of WT and cca-miR398 overexpressed transgenic Arabidopsis

The expression of cca-miR398 was up-regulated (P < 0.01) in transgenic Arabidopsis (Fig. 2A). Seed germination of transgenic Arabidopsis was more sensitive than WT Arabidopsis to copper sulfate. For seedlings overexpressing cca-miR398, the greening of the cotyledons was greatly reduced when the seeds were germinated on medium containing different supplementary copper sulfate (Figs. 2B–2E). In the 40 mM copper sulfate medium, 84% of the germinated WT seedlings turned green while only 63% of the transgenic seeds turned green after seven days (Fig. 2J). In addition, the transgenic Arabidopsis had significantly shorter (P < 0.01) roots than WT plants, regardless of the presence or absence of copper sulfate (Figs. 2F–2I and 2K).

Effects of copper sulfate on SOD, POD and CAT activity in the leaves of WT and transgenic Arabidopsis

In general, copper sulfate treatments boosted accumulations of SOD, POD, and CAT more in WT Arabidopsis than in transgenic Arabidopsis. There were almost no differences in the activities between the transgenic and WT lines under normal conditions. As copper sulfate stress increased, the activities of SOD, POD, and CAT increased until 1,000 mM and then decreased (Fig. 3). Under 1,000 mM treatment, SOD, POD, and CAT activities in the transgenic Arabidopsis increased from 1,385, 930, and 89 U g⁻¹FW to 1,730, 3,727, and 142 U g⁻¹FW, respectively, while SOD, POD, and CAT in the WT plants almost doubled their activities (Fig. 3). There were activity increases for SOD, CAT, and POD at normal conditions to 1,000 mM and a decrease at 1,000 to 2,000 mM stress conditions (figs. 3B and 3C). The transgenic lines showed significantly lower (P < 0.01) SOD, POD, and CAT activities than the WT lines after copper sulfate treatments, which may reflect cca-miR398 targeting the three enzymes and thereby reducing antioxidant enzyme activity under stress conditions (Fig. 3).

Effects of copper sulfate on chlorophyll fluorescence parameters in the leaves of WT and transgenic Arabidopsis

The results showed that there were no differences in chlorophyll fluorescence parameters between the transgenic lines and WT under normal conditions. However, the related parameters were significantly lower in the transgenic lines than the WT lines (P < 0.01) after copper sulfate treatments. Under copper sulfate treatments, Fv/Fm decreased significantly (P < 0.01) in the leaves of transgenic plants when compared to WT leaves (Fig. 4A). Within the range of copper sulfate treatments, qP decreased rapidly between the transgenic lines and WT, reflecting the declining proportion of open PSII reaction centers (Fig. 4B). At a concentration of 1,500 mM copper sulfate or higher, qP began to stabilize, indicating changes in the regulative capability of PSII reaction centers. The capacities of qP in the transgenic Arabidopsis were more sensitive to copper sulfate stress.

Plants in copper sulfate treatments showed that the changing trends of PSII activity were compatible with that of qP (figs. 4B and 4C). Simultaneously, NPQ increased to the highest point under treatments of 500 and 1,000 mM copper sulfate, indicating that WT plants regulated excessive energy dissipation more efficiently than transgenic Arabidopsis (Fig. 4D). The change in ETR curves was similar to that of PSII and the overexpression of cca-miR398 resulted in less ETR susceptibility of ETR to copper sulfate stress (Fig. 4E). In treatments lower than 1,500 mM copper sulfate, ETR in the leaves of transgenic Arabidopsis was almost zero in contrast to the 45.45 ± 1.75µmol m⁻² s⁻¹ in WT leaves.

Images of Fv/Fm in the transgenic Arabidopsis leaves varied with increasing copper sulfate concentrations, which differed from those of WT leaves (Fig. 4F). The imaging color of Fv/Fm in the WT leaves remained partially blue (Fv/Fm = 0.8) with treatments of up to 2,000 mM copper sulfate, although there were small necrotic areas (Fv/Fm = 0). The slow change in the Fv/Fm image color in WT leaves suggested that there was high PSII activity when treated with less than 2,000 mM copper sulfate. The Fv/Fm image colors in the transgenic leaves showed greater degrees of necrosis. The necrosis in the transgenic Arabidopsis leaves accounted for about 90% area of the whole leaf, suggesting that the quantum efficiency of light energy transfer in PSII was almost destroyed in the plants exposed to less than 2,000 mM copper sulfate.

Copper sulfate stress affecting the expression patterns of cca-miR398, its targets, and copper-responsive miRNAs in WT and transgenic Arabidopsis

qRT-PCR was performed to analyze the transcript abundance of candidate genes in Arabidopsis under copper sulfate stress. The qRT-PCR results showed that the expression of these genes was significantly lower (P < 0.01) in the transgenic lines than in the WT lines after copper sulfate treatments (Figs. 5A–5C). CSD expression levels increased rapidly with an increase of copper stress in WT plants yet decreased in transgenic plants (Figs. 5A–5C). The results also indicated the involvement of cca-miR398 and CSDs in the abiotic stress response of hickory. CSD3 belongs to CSD protein family (Fig. 1), and we could not validate if it was a target of cca-miR398 by 5′-RACE, but we were still able to examine its expression profile based on qRT-PCR. The CSD3 expression level remained similar to CSD1 and CSD2 (Figs. 5A–5C), suggesting that miR398 was involved in the regulation of CSD3 at the mRNA level.

As expected, cca-miR398 levels increased in all transgenic plants compared with WT Arabidopsis (Fig. 2A), and the expression of cca-miR398 decreased as copper sulfate concentrations increased. We also examined the transcript levels of several copper-responsive microRNAs, including miR397 and miR408, in both WT and transgenic Arabidopsis. The results showed that expressions of miR397 and miR408 were down-regulated under low copper sulfate treatments (Figs. 5E and 5F). The transcript abundance of miR397 was significantly higher (P < 0.01) in the transgenic lines compared with the WT lines under different copper sulfate treatments (Fig. 5E); however, the abundance of miR408 was significantly lower (P < 0.01; Fig. 5F).