A cassava protoplast system for screening genes associated with the response to South African cassava mosaic virus

Abstract

Background: The study of transient gene expression in cassava plants during virus infection using existing protocols is laborious and may take approximately fifteen weeks due to cassava's recalcitrance to transformation. The combination of a protoplast system with CRISPR-mediated gene editing promises to shorten the turnaround time from plant tissue culture to high-throughput gene expression screening for candidate genes. Here, we detail a protocol for screening genes associated with the response to South African cassava mosaic virus (SACMV) in cassava protoplasts, with reference to the ubiquitin E3 ligase gene, MeE3L.

Methods: Cassava protoplasts of model, and SACMV-susceptible and -tolerant genotypes, were transformed with SACMV infectious clones and/or a CRISPR-editing construct targeting the MeE3L using PEG4000-mediated transfection. DNA and RNA were extracted from transformed protoplasts at 24 h post-transfection. Relative SACMV DNA accumulation was determined via qPCR using DpnI-digested total DNA, MeE3L relative expression was determined via reverse transcriptase qPCR, and results were analysed using one-way ANOVA, Tukey's HSD test and the 2^-ΔΔCTstatistical method. The MeE3L exonic region was sequenced on the ABI 3500XL Genetic Analyzer platform; and sequences were analysed for mutations using MAFTT and MEGA-X software. Construction of a phylogenetic tree was done using the Maximum Likelihood method and Jones-Taylor-Thornton (JTT) matrix-based model.

Results: The differential expression of unedited and mutant MeE3L during SACMV infection of model, susceptible and tolerant cassava protoplasts was determined within 7 weeks after commencement of tissue culture. The study also revealed that SACMV DNA accumulation in cassava protoplasts is genotype-dependent and induces multiple mutations in the tolerant landrace MeE3L homolog. Notably, the susceptible cassava landrace encodes a RINGless MeE3Lwhich is silenced by SACMV-induced mutations. SACMV also induces mutations which silence the MeE3L RING domain in protoplasts from and tolerant cassava landraces.

Conclusions: This protocol presented here halves the turnaround time for high-throughput screening of genes associated with the host response to SACMV. It provides evidence that a cassava E3 ligase is associated with the response to SACMV and forms a basis for validation of these findings by in planta functional and interaction studies.

Keywords: Cassava mosaic disease; Geminivirus; Protoplast, ubiquitin E3 ligase.

Fig. 1

Cassava protoplast isolation from leaf mesophyll cells by 16 h-long enzymatic digestion. a M. esculenta 4-week old donor plants cultured on ½ Murashige and Skoog medium. b Protoplasts from model M. esculenta cv.60444 c Protoplasts from susceptible M. esculenta T200 d Protoplasts from tolerant M. esculenta TME3. Spherical protoplasts with chloroplasts around the edge of the cell membrane and central vacuole were observed (shown by red arrows). Protoplasts were visualised under bright field microscopy

Fig. 2

Analyses of viability, quality and transformation of cassava protoplasts. Viability of freshly isolated protoplasts was determined by Evans’ Blue Dye staining and visualisation under bright field microscopy. Analysis of protoplast quality was done by flow cytometric density measurement where events are discriminated by size and granularity, represented in log scale density plots. The size and shape of cassava protoplasts are measured by their effect on the forward scatter (FSC-A) and side scatter (SSC-A) of the laser. Stable transformation with the CRISPR construct was determined by fluorescence microscopy visualisation of eGFP fluorescence through the GFP filter and bright field. a Protoplasts from model M. esculenta cv.60444; b Protoplasts from susceptible M. esculenta T200 (c); Protoplasts from tolerant M. esculenta TME3. Non-viable cells are stained blue. d Plot of model M. esculentac v.60444 protoplast density (e) Plot of susceptible M. esculenta T200 protoplast density (f) Plot of tolerant M. esculenta TME3 protoplast density. Circled regions correspond to desirable protoplasts. g M. esculenta T200 protoplasts visualised through the GFP filter (h) M. esculenta T200 protoplasts visualised through both the bright field and GFP filters

Fig. 3

Primary structure, secondary structure and phylogenetic analysis of MeE3L and/or its protein product. Sequence alignment and agarose gel resolution of MeE3L partial gene and partial transcript respectively show a 53 bp insertion mutation in the lengthier susceptible T200 homolog that is absent in the reference AM560-2, model cv.60444 and tolerant TME3 homologs. Computational prediction of secondary, molecular and zinc-binding structures of MeE3L homologs shows significant differences between T200 structure and the other structures. N → C. Phylogenetic analysis shows significant evolutionary distance between susceptible T200 MeE3L and other plant MeE3L homologs. a Genomic nucleotide sequence alignment showing insertion mutation between nucleotides 397–398 and 422–423 in the susceptible T200 MeE3L homolog. b Agarose gel resolution of the PCR-amplified susceptible T200 and tolerant TME3 partial transcripts of MeE3L (c) Amino acid sequence alignment showing premature stop mutation at amino acid residue 141 in susceptible T200 MeE3L protein homolog (d) The reference AM560-2 MeE3L amino acid sequence. Asterisks denote stop codons in susceptible T200 (amino acid residue 141) and reference AM560-2 / model cv.60444 / tolerant TME3 (amino acid residue 200) homologs respectively. Red letters denote the first susceptible T200 MeE3L stop mutation at amino acid residue 141. Underlined letters denote the sequence adhering to the RING finger domain consensus sequence [CX₂CX_(9–39)CX_(1–3)HX_(2–3)CX₂CX_(4–48)CX₂X]. (ei) Reference AM560-2 MeE3L homolog predicted secondary structure (eii) Reference AM560-2 MeE3L homolog predicted tertiary molecular structure (eiii) Zinc binding in RING domain of reference AM560-2 MeE3L (eiv) Model cv.60444 MeE3L homolog predicted secondary structure (ev) Model cv.60444 MeE3L predicted tertiary molecular structure (evi) Zinc binding in RING domain of model cv.60444 MeE3L (evii) Susceptible T200 MeE3L homolog predicted secondary structure (eviii) Susceptible T200 MeE3L predicted tertiary molecular structure (eix) Predicted ligand binding structure of susceptible T200 MeE3L (ex) Tolerant TME3 MeE3L homolog predicted secondary structure (exi) Tolerant TME3 MeE3L predicted tertiary molecular structure (exii) Zinc binding in RING domain of tolerant TME3 MeE3L [Predictions were run on the I-TASSER On-line Server (https://zhanglab.ccmb.med.umich.edu/I-TASSER/; [97]] (f) Evolutionary analysis of plant MeE3L homologs using Maximum Likelihood method and Jones-Taylor-Thornton (JTT) matrix-based model in MEGA X [40]. Bootstrap support was calculated from 1000 replicates. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site

Fig. 4

Assessment of viral DNA accumulation, relative MeE3L expression, and predicted MeE3L primary structure in transformed cassava protoplasts. a Relative DNA accumulation of SACMV- and SACMV + CRISPR-Cas9-transformed cassava protoplasts under different transformation conditions. ΔMeE3L = mutant CRISPR-edited MeE3L. Real-time qPCR was performed in triplicate using DpnI-treated total DNA extracted from cassava protoplasts 24 hpt as template. b MeE3L relative expression levels in transformed cassava protoplasts. V = SACMV-transformed. ΔMeE3L = gene-edited MeE3L. C* = transformed with CRISPR construct lackingt gRNA duplex. RT-qPCR was performed in triplicate using total mRNA as template. c Stop mutation induced in SACMV-infected susceptible T200 MeE3L. d Stop mutation induced in SACMV-infected tolerant TME3 MeE3L. e The predicted amino acid sequence of tolerant TME3 MeE3L at reference sequence positions 2–148 showing multiple mutations in SACMV-infected variant. V = SACMV-infected. C = gene-edited. Sequence alignment was conducted in MEGA-X [40]. f Frequency and types of mutation at target gRNA sites from CRISPR-transformed protoplasts. Frequency of clones with altered sequence was obtained by expressing number of amplicons from a polyclonal mix with sequence alteration as a ratio of total amplicons (n = 10 per genotype) sequenced. Mutations were determined by aligning amplicon sequences with wild-type reference AM560-2 [14] and TME3 (RefSeq ID: RSFT01000007,GenBankassembly GCA_003957995.1 (unpublished data)) MeE3L homologs. Alignment was conducted on MEGA-X [40] using the CLUSTAL W algorithm for multiple sequence alignment [44]. g Timeline for rapid screening of genes associated with the response to South African cassava mosaic virus in cassava