Characterization of the bark storage protein gene (JcBSP) family in the perennial woody plant Jatropha curcas and the function of JcBSP1 in Arabidopsis thaliana

Plant materials and nitrogen treatment

Four-year-old adult J. curcas were grown in the experimental field of Xishuangbanna Tropical Botanical Garden (21°54′N, 101°46′E; 580 m in altitude) in Yunnan Province. Wild-type Arabidopsis thaliana ecotype Columbia (Col-0) and the transgenic lines were grown in an environmentally controlled room at 22 °C under a 16-h light/8-h dark photoperiod. Wild-type J. curcas seeds were sown in sand that had been washed several times with distilled water and grown at 30 °C under a 12-h light/12-h dark photoperiod. Then, two-month-old J. curcas seedlings were randomly divided into three groups, with 12 plants per group. The three groups were treated with different concentrations of NH₄NO₃ solution (0, 5, and 50 mM). The seedlings were watered every week with 100 mL of a NH₄NO₃ solution per cup of seedlings.

Sequence analysis

We used BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) to analyze the cDNA sequence, CDS and amino acid sequence of JcBSP gene family members in the NCBI database. The conserved domains of deduced protein sequences were analyzed by the NCBI Conserved Domain Database (NCBI-CDD, https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). The alignment of amino acid sequences was performed using DNAMAN software (version 6, Lynnon Biosoft Corporation, Canada, https://www.lynnon.com/dnaman.html).

Phylogenetic analysis

To examine the phylogenetic relationships of the BSP homologues from different species, we retrieved the deduced protein sequences from the NCBI database (https://www.ncbi.nlm.nih.gov/) and selected the sequences from species belonging to Euphorbiaceae and Populus with the highest sequence similarity to JcBSPs. Sequences of PtdPNI288 and PtdWIN4 were derived from the reference literature (Cooke & Weih, 2005; Wildhagen et al., 2010). A phylogenetic tree was built in the MEGA program (version 7.0, https://megasoftware.net/) using the neighbor-joining method with 1000 bootstrap replicates.

RNA extraction

The samples for RNA extraction included different tissues (roots, stems, shoot apices, young leaves, mature leaves, male flowers, female flowers, fruits and seeds) of four-year-old adult J. curcas, stems collected from October 2019 to August 2020, two-month-old J. curcas seedlings treated with different concentrations of NH₄NO₃ solution for 0, 2, 4, 6, and 8 weeks, and leaves of one-month-old WT and transgenic Arabidopsis. These samples were quickly frozen in liquid nitrogen and stored at −80 °C. Total RNA was extracted using the silica adsorption method (Ding et al., 2008) , and the concentration and purity of RNA were detected by spectrophotometry and agarose gel electrophoresis, respectively.

qRT-PCR analysis

Reverse transcription of total RNA was performed using the PrimeScript^® RT reagent kit with gDNA Eraser (Takara, Dalian, China). qRT-PCR was performed on the Roche 480 real-time PCR detection system using LightCycler^® 480 SYBR Green I Master Mix (Roche Diagnostics, Indianapolis, IN, USA). The qRT-PCR reactions were performed under the following conditions: 5 min at 95 °C for the initial denaturation, followed by 42 cycles of 10 s at 95 °C, 20 s at 57 °C, and 20 s at 72 °C for the PCR amplification, and 1 cycle of 30 s at 95 °C, 30 s at 65 °C and 0.06 °C/s heating up to 95 °C for the melting curve. Data was analyzed using the 2^−ΔΔCT method as described by Livak & Schmittgen (2001). All expression data obtained in the qRT-PCR assay were normalized to the expression of JcActin1 (Zhang et al., 2013) and AtActin2. The primers used for qRT-PCR are listed in Table S1.

Protein extraction

Stem samples of four-year-old adult J. curcas collected from October 2019 to August 2020 were also used for total protein extraction. The bark + phloem and xylem + pith dissected from stems were ground into powder mixtures. Fifty milligrams of powder were homogenized in 300 µl of protein extract buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100, 10% glycerol). The mixture was incubated at 4 °C for 30 min and then centrifuged at 12,000 rpm for 15 min at room temperature. The supernatant was collected and analyzed by a Bradford protein assay kit (BL524A, GUANGKE Technology Company, Kunming).

Correlation analysis

Correlation analysis between the total protein concentration and JcBSP1 expression was performed by the method of Spearman’s rank correlation using R package ggpubr (version 0.4.0, https://cran.r-project.org/).

Construction of the JcBSP1 overexpression vector and Arabidopsis transformation

The primers XD423 (CGAGCTCATGGCTATGGCGACGGTGAT) and XD424 (CGGATCCTCACTCTTCATCATAACGGA) carrying SacI and BamHI restriction sites, respectively, were used to clone the full-length JcBSP1 CDS. Then, the PCR product was cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA). To generate the 35S:JcBSP1 overexpression vector, SacI and BamHI were used to digest the plant transformation vector pOCA30 (Chen & Chen, 2002) and the pGEM-T Easy vector containing the JcBSP1 sequence, respectively, and then, the resulting fragments were ligated by using T4 DNA Ligase (Promega). The generated 35S:JcBSP1 plasmid was transferred to Agrobacterium tumefaciens EHA105. Transformation of Arabidopsis was performed using the floral dip method (Clough & Bent, 1998).

Identification of the members of the JcBSP gene family

We found six members of the JcBSP gene family in J. curcas from the NCBI database using BLAST and designated them as JcBSP1, JcBSP2, JcBSP3, JcBSP4, JcBSP5, and JcBSP6 (Table 1). All the JcBSP gene family members contain a conserved domain, purine nucleoside phosphorylase_uridine phosphorylase_1 (PNP_UDP_1), which is a signature of the phosphorylase superfamily (Fig. 1).

To investigate the evolutionary relationships among BSP homologous genes, we performed phylogenetic analysis of genes from J. curcas and other species. The phylogenetic tree showed that JcBSP1, JcBSP5, and JcBSP6 were closely related to the BSP homologues from Euphorbiaceous species, and JcBSP2, JcBSP3, and JcBSP4 were closely related to the BSP homologues from poplar (Fig. 2).

The expression patterns of JcBSPs in J. curcas

To analyze the expression patterns of JcBSP family members, we used qRT-PCR to detect the expression levels of JcBSPs in roots, stems, shoot apices, young leaves, mature leaves, male flowers, female flowers, fruits, and seeds of adult J. curcas. The results showed that JcBSP1 and JcBSP2 exhibited similar expression patterns, in which both were highly expressed in shoot apices, stems, young leaves and female flowers; JcBSP3 was mainly expressed in roots, stems, shoot apices, male flowers and fruits; the expression of JcBSP4 was concentrated in reproductive tissues, with the highest expression level in female flowers; the expression of JcBSP5 was concentrated in vegetative tissues, with the highest expression level in stems; and JcBSP6 was remarkably expressed in male flowers (Fig. 3). Based on these results, we hypothesized that JcBSPs could play important roles in the growth and development of various organs, except seeds, in which all the members were barely expressed. This finding also indicates that JcBSPs may be vegetative storage proteins rather than seed storage proteins.

Seasonal changes in total protein concentrations and JcBSP expression in the stems of J. curcas

In perennial deciduous trees, most nitrogen resources in senescing leaves are transported to perennial tissues (bark and wood), and stored as proteins during autumn and winter; the next spring, these proteins are hydrolyzed to amino acids, which are transported from perennial tissues to growing tissues (Chapin & Kedrowski, 1983; Cooke & Weih, 2005; Sauter, Cleve & Wellenkamp, 1989). Therefore, we investigated whether the total protein concentration in J. curcas stems was relevant to seasons.

From October 2019 to August 2020, we sampled the stems of adult J. curcas in two parts (bark + phloem and xylem + pith), as shown in Fig. 4A, and examined the total protein concentrations of the samples. The results showed that the total protein concentrations in the bark + phloem and xylem + pith were approximately 10.5 mg/g FW in October; then, J. curcas entered the dormant stage in December, and the total protein concentrations reached a peak in the bark + phloem (12.3 mg/g FW) and xylem + pith (16.4 mg/g FW). When J. curcas began to enter the growing season in March, the total protein concentrations decreased sharply to 7 mg/g FW in the bark + phloem, which further decreased to 5.5 mg/g FW in August. The total protein concentration in the xylem + pith showed a similar trend (Fig. 4B). The total protein concentration in the J. curcas stem exhibited a seasonal change, which accumulated in autumn and winter and decreased in spring and summer. This result indicates that the total protein in the stems is a form of nitrogen storage during the overwintering period of J. curcas, and this protein is reallocated during the growing seasons.

Based on the above results, we examined the seasonal course of JcBSP expression in the same samples mentioned above (Fig. 4A) to investigate whether the expression of JcBSPs shows the same seasonal changes as the total protein concentration. The results showed that the expression of JcBSP family members in stems exhibited different patterns over the seasonal course (Fig. 5). From autumn to winter, the expression of JcBSP1 in the two parts of the stems increased rapidly, with a higher level in xylem + pith, and then decreased sharply in spring and remained low until August. The seasonal changes in JcBSP1 expression in the two parts of the stem were entirely consistent with those of the total protein concentration. However, the seasonal expression patterns of JcBSP2, JcBSP4 and JcBSP5 in stems were inconsistent with those of the total protein concentration. In addition, only in the xylem + pith did the expression of JcBSP3 and JcBSP6 show the same seasonal changes as that of the total protein concentration, but their expression levels were much lower than that of JcBSP1. Therefore, we analyzed the correlation between seasonal changes in the total protein concentration and JcBSP1 expression. It turned out that there were significant correlations between them in the bark + phloem (r = 0.63, P < 0.01) and the xylem + pith (r = 0.67, P < 0.01) (Fig. 6). These results suggest that JcBSP1 might play an important role in seasonal nitrogen cycling.

JcBSP expression in response to nitrogen

To further verify the correlation between JcBSPs and seasonal nitrogen cycling, we investigated the response of these genes to nitrogen induction. We applied a 0, 5, and 50 mM NH₄NO₃ solution to two-month-old J. curcas seedlings. After 8 weeks of treatment, we found that the group treated with the 5 mM NH₄NO₃ solution grew better than the other two groups (Fig. 7A). This result indicated that nitrogen supply in a certain concentration range could effectively promote the growth of J. curcas.

We collected the leaves of J. curcas seedlings treated with different concentrations of NH₄NO₃ for 0, 2, 4, 6, and 8 weeks to detect changes in JcBSP family member expression. The results showed that the expression of JcBSP1 increased obviously along with the increased NH₄NO₃ concentration and application duration; JcBSP2 expression increased remarkably with the 5 mM NH₄NO₃ treatment for 2–8 weeks and with 50 mM NH₄NO₃ treatments for 8 weeks; JcBSP4 expression was induced obviously with only the 50 mM NH₄NO₃ treatment for 4 weeks; however, the expression of JcBSP3, JcBSP5 and JcBSP6 was not induced by NH₄NO₃ treatment (Fig. 7B). The results indicated that JcBSP1 and JcBSP2 expression is responsive to nitrogen induction. In particular, JcBSP1 expression was positively correlated with the nitrogen concentration and application duration. Combined with the seasonal changes in JcBSP1 expression in stems, we further concluded that JcBSP1 might be a form of nitrogen storage in the seasonal nitrogen cycling in J. curcas.

Overexpression of J. curcas JcBSP1 in Arabidopsis led to enlarged rosette leaves, flowers, and seeds

Next, we investigated the function of JcBSP1 in plant growth and development in transgenic Arabidopsis. Twenty-two independent 35S:JcBSP1 transgenic Arabidopsis lines were generated (Fig. 8A). JcBSP1 expression in seven transgenic lines showing similar phenotypes was analyzed, and most of transgenic lines yielded abundant transgene expression (Fig. S1). We investigated in further detail the phenotypes of two independent transgenic lines, L4 and L10, which exhibited high and intermediate expression levels of JcBSP1, respectively (Fig. 8B).

During plant growth and development, we found that 35S:JcBSP1 transgenic Arabidopsis produced larger rosette leaves and flowers than the wild-type (WT) plants (Figs. 8C and 8D). As shown in Figs. 8E and 8F, the rosette leaf lengths and widths were all significantly increased in transgenic lines. And larger seeds and significantly increased hundred-seed weights were found in transgenic plants (Figs. 9A and 9B). Consequently, the seed yields in transgenic lines were significantly higher than that of the WT (Fig. 9C). These results indicated that JcBSP1 was able to affect plant growth and development.