CRISPR/Cas9-mediated targeted mutagenesis of GmSPL9 genes alters plant architecture in soybean

Abstract

Background: The plant architecture has significant effects on grain yield of various crops, including soybean (Glycine max), but the knowledge on optimization of plant architecture in order to increase yield potential is still limited. Recently, CRISPR/Cas9 system has revolutionized genome editing, and has been widely utilized to edit the genomes of a diverse range of crop plants.

Results: In the present study, we employed the CRISPR/Cas9 system to mutate four genes encoding SQUAMOSA PROMOTER BINDING PROTEIN-LIKE (SPL) transcription factors of the SPL9 family in soybean. These four GmSPL9 genes are negatively regulated by GmmiR156b, a target for the improvement of soybean plant architecture and yields. The soybean Williams 82 was transformed with the binary CRISPR/Cas9 plasmid, assembled with four sgRNA expression cassettes driven by the Arabidopsis thaliana U3 or U6 promoter, targeting different sites of these four SPL9 genes via Agrobacterium tumefaciens-mediated transformation. A 1-bp deletion was detected in one target site of the GmSPL9a and one target site of the GmSPL9b, respectively, by DNA sequencing analysis of two T0-generation plants. T2-generation spl9a and spl9b homozygous single mutants exhibited no obvious phenotype changes; but the T2 double homozygous mutant spl9a/spl9b possessed shorter plastochron length. In T4 generation, higher-order mutant plants carrying various combinations of mutations showed increased node number on the main stem and branch number, consequently increased total node number per plants at different levels. In addition, the expression levels of the examined GmSPL9 genes were higher in the spl9b-1 single mutant than wild-type plants, which might suggest a feedback regulation on the expression of the investigated GmSPL9 genes in soybean.

Conclusions: Our results showed that CRISPR/Cas9-mediated targeted mutagenesis of four GmSPL9 genes in different combinations altered plant architecture in soybean. The findings demonstrated that GmSPL9a, GmSPL9b, GmSPL9c and GmSPL9 function as redundant transcription factors in regulating plant architecture in soybean.

Keywords: CRISPR/Cas9; Plant architecture; Plastochron length; SPL; Soybean.

Fig. 1

Schematic figure of target sites in four GmSPL9 genes. (a) Gene structure of GmSPL9a with two target sites GmSPL9a-SP1 and GmSPL9a-SP2. (b) Gene structure of GmSPL9b with two target sites GmSPL9b-SP1 and GmSPL9b-SP2. (c) Gene structure of GmSPL9c with one target site GmSPL9c-SP3. (d) Gene structure of GmSPL9d with two target sites GmSPL9d-SP3 and GmSPL9d-SP4. Nucleotides in red represent the protospacer adjacent motif (PAM). Nucleotides underlined indicate the target sites. Gray stripe, untranslated regions; black stripe, exon; black line, intron

Fig. 2

Schematic figure of the binary vector designed for mutagenesis of the four GmSPL9 genes using the CRISPR/Cas9 technology. The pYLCRISPR/Cas9P_35S-BS was derived from the pYLCRISPR/Cas9P_35S-B [38]. The target adaptor SP1, targeting two sites (GmSPL9a-SP1 and GmSPL9b-SP1), directed by the Arabidopsis thaliana U3d promoter; the target adaptor SP2, targeting two sites (GmSPL9a-SP2 and GmSPL9b-SP2), directed by the A. thaliana U6–1 promoter; the target adaptor SP3, targeting two sites (GmSPL9c-SP3 and GmSPL9d-SP3), directed by the A. thaliana U6–29 promoter; the target adaptor SP4, targeting one site (GmSPL9d-SP4), directed by the A. thaliana U3b promoter

Fig. 3

Results obtained from mutagenesis of four GmSPL9 genes by CRISPR/Cas9 technology. (a) Detailed sequence of the target site GmSPL9a-SP1 in the T0–10 line. (b) Detailed sequence of the target site GmSPL9b-SP1 in the T0–20 line. (c) Detailed sequence of the target site GmSPL9b-SP1 in the T1–10-3 line. (d) Detailed sequence of the target site GmSPL9c-SP3 in the T1–10-4 line. (e) Detailed sequence of the target site GmSPL9d-SP1 in the T3–10–1-3-2 line. Nucleotides in red and underlined represent the protospacer adjacent motif (PAM). The underlined nucleotides indicated the target sites. ‘-’ signs indicate the number of deleted nucleotides. W82 represents Williams 82 wild-type sequence

Fig. 4

The soybean CRISPR/Cas9-induced T4-generation spl9 mutant lines showed altered plant architecture. (a) Representative pictures showing the primary and secondary branches of wild-type (WT), GmmiR156b-overexpressing (GmmiR156b-OX) and spl9a (−/−)/spl9b-1 (−/−)/spl9c (−/−)/spl9d (−/−) quadruple mutant plants at 50th day after emergence, after the leaves were removed. Red arrows indicate the secondary branches on primary branches. (b) Node number on main stem, (c) total node number per plant, (d) branch number, and (e) dry weight were recorded after the experiment was completed at day 50th after emergence. (f) Dynamic changes in the trifoliate leaf number on main stem in different genotypes. (b-f) The data of WT, GmmiR156b-OX and the spl9b-1 (−/−) single mutant plants were obtained from six plants. The data of spl9a/spl9b-1/spl9c were obtained from five spl9a (−/−)/spl9b-1 (−/−)/spl9c (+/−) plants, the data of spl9a/spl9b-2/spl9c were obtained from nine spl9a (−/−)/spl9b-2 (−/−)/spl9c (+/−) plants, and the data of spl9a/spl9b-1/spl9c/spl9d were obtained from two spl9a (−/−)/spl9b-1 (−/−)/spl9c (−/−)/spl9d (−/−) and one spl9a (−/−)/spl9b-1 (−/−)/spl9c (+/−)/spl9d (+/−) plants. “a, b, c and d” indicate statistically significant differences among the genotypes (one-way ANOVA of variance, P < 0.05)

Fig. 5

Expression patterns of four GmSPL9 genes in wild-type (WT) and the spl9b-1 (Bar negative) mutant plants. (a) Relative expression of GmSPL9a, GmSPL9b, GmSPL9c and GmSPL9d in shoot apical meristem of WT and spl9b-1 plants. (B) Relative expression of GmSPL9a, GmSPL9b, GmSPL9c and GmSPL9d in leaves of WT and spl9b-1 plants. The plants were grown under artificial climate chamber (15-h light/9-h dark) conditions for 20 days after emergence. Relative transcript levels were assessed by quantitative RT-PCR and normalization to the expression level of the GmTUB gene. Expression levels shown are means ± SEs of three replicates (**P < 0.01; Student’s t-test)