Development of a Highly Efficient Multiplex Genome Editing System in Outcrossing Tetraploid Alfalfa (Medicago sativa)

Abstract

Alfalfa (Medicago sativa) is an outcrossing tetraploid legume species widely cultivated in the world. The clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (CRISPR/Cas9) system has been successfully used for genome editing in many plant species. However, the use of CRISPR/Cas9 for gene knockout in alfalfa is still very challenging. Our initial single gRNA-CRISPR/Cas9 system had very low mutagenesis efficiency in alfalfa with no mutant phenotype. In order to develop an optimized genome editing system in alfalfa, we constructed multiplex gRNA-CRISPR/Cas9 vectors by a polycistronic tRNA-gRNA approach targeting the Medicago sativa stay-green (MsSGR) gene. The replacement of CaMV35S promoter by the Arabidopsis ubiquitin promoter (AtUBQ10) to drive Cas9 expression in the multiplex gRNA system led to a significant improvement in genome editing efficiency, whereas modification of the gRNA scaffold resulted in lower editing efficiency. The most effective multiplex system exhibited 75% genotypic mutagenesis efficiency, which is 30-fold more efficient than the single gRNA vector. Importantly, phenotypic change was easily observed in the mutants, and the phenotypic mutation efficiency reached 68%. This highly efficient multiplex gRNA-CRISPR/Cas9 genome editing system allowed the generation of homozygous mutants with a complete knockout of the four allelic copies in the T0 generation. This optimized system offers an effective way of testing gene functions and overcomes a major barrier in the utilization of genome editing for alfalfa improvement.

Keywords: CRISPR/Cas9; alfalfa; genome editing; multiplex; mutagenesis; outcrossing; polyploid.

Figure 1

Schematic illustration of the stay-green (MsSGR) gene structure and the construction of different multiplex gRNA-CRISPR/Cas9 vectors for genome editing in alfalfa. (A) Alfalfa MsSGR gene structure and four designed gRNAs (1, 2, 3, 4) on exons 1, 2, and 3. (B) Map of a multiplex MsSGR-gRNA-CRISPR/Cas9 vector in which Cas9 is driven by the CaMV35S promoter (version I). (C) Map of a multiplex MsSGR-gRNA-CRISPR/Cas9 vector in which the gRNA scaffold is modified (version II). Red line at the 5′ of gRNAs of vector II indicates point mutation made at TTTTT to TTGTT to abolish the pausing effect of continuous T’s. (D) Map of a multiplex MsSGR-gRNA-CRISPR/Cas9 vector in which Cas9 is driven by the AtUBQ10 promoter (version III). All vector versions were constructed using polycistronic tRNA-CRISPR-gRNA (PTG) clustering system.

Figure 2

Mutagenesis efficiency and phenotype of alfalfa stay-green (MsSGR) mutants obtained by genome editing using different vectors. (A) Mutagenesis efficiency of single and multiplex gRNA-CRISPR/Cas9 vectors (versions I, II, and III) in MsSGR gene editing. (B) Phenotype of detached leaves of MsSGR mutants obtained by using multiplex vector version I during dark treatment assay. (C) Phenotype of detached leaves of MsSGR mutants obtained by using multiplex vector version II during dark treatment assay. (D) Phenotype of detached leaves of MsSGR mutants obtained by using multiplex vector version III during dark treatment assay. (E) MsSGR mutants’ phenotypic classes (strong, mild and weak) within three vector versions (I, II, and III) based on stay-green retention capacity during dark treatment (DT) assay. Control-1: wild type; control-2: empty vector control. (F) Detached leaf phenotypic categories of different alfalfa mutant lines after 10 days of dark treatment. Control-1: wild type; control-2: empty vector control. The detached leaf dark treatment assay in B-D was carried out at the same time using control-1 and contol-2 as common controls. Error bars indicate SE

Figure 3

Molecular analysis of nucleotide deletion/insertion in MsSGR mutants generated by multiplex gRNA-CRISPR/Cas9 vectors. Mutation events detected at target sites (gRNA1, gRNA2, gRNA3, and gRNA4). Wild type (WT) sequence given for each target site (designed gRNA), deletions are indicated by red dashed lines with minus sign, insertions, or substitutions are indicated by red letters with plus sign and “s,” respectively, PAM indicated by red underlined italic letters. The four allelic copies of the MsSGR gene are designated as allele-1 (A1), allele-2 (A2), allele-3 (A3), and allele-4 (A4). Overall allelic copies mutated in percentage determine the homozygosity/heterozygosity of the mutant.

Figure 4

Genotypic and phenotypic analyses of MsSGR mutation by TA-clone sequencing and dark treatment. (A) Sequence analysis of different TA clones from MsSGR mutant #21 (MsSGR-21) at four target sites (gRNA1, 2, 3, and 4). Red arrow indicates the PAM of each gRNA, red dash indicates nucleotide deletion, red letter indicates insertion/substitution, colone-1, 2, … indicate corresponding plasmid sequences which represent allelic copies of the MsSGR gene in alfalfa. (B) Phenotype of detached leaves of MsSGR-21 mutant incubated in dark for 0, 5, and 10 days. Due to tetra-allelic homozygous mutation, all leaves stayed greenish. (C, D) Phenotype of detached leaves of wild type (WT) and empty vector control incubated in dark for 0, 5, and 10 days, all leaves became yellowish after ten days of dark incubation.

Figure 5

Illustration of phenotypic appearance of alfalfa MsSGR mutant harvested at first flower bud initiation stage and dried under the shade-induced condition for four weeks. (A) Phenotype of alfalfa wild type (WT) at vegetative stage (control), (B) Phenotype of MsSGR-21 mutant at vegetative stage, (C) Phenotypic appearance of alfalfa WT dried under shade for 4 weeks, (D) Phenotypic appearance of MsSGR-21 mutant dried under shade condition for four weeks shade, (E) Phenotypic appearance WT (control) (individual plant) dried under shade for four weeks, (F–M) Phenotypic appearance of eight independent lines of MsSGR mutant generated by multiplex gRNA-CRISPR/Cas9 vector version III after 4 weeks shade treatment. All MsSGR mutants showed strong phenotype due to their tetra-allelic homozygosity.