Self‐incompatibility based functional genomics for rapid phenotypic characterization of seed metabolism genes

Reverse genetic characterization of plant gene function through technologies such as CRISPR/Cas, RNAi, or gene overexpression requires the ability to efficiently transform the plant species of interest. However, efficient transformation systems are not available for most plant species. Physaria fendleri [previously Lesquerella fendleri (Al-Shehbaz and O'Kane Jr., 2002)] is an emerging oilseed crop valued for its unusual hydroxylated fatty acids (HFA, e.g. lesquerolic acid) that accumulate up to 60% of seed oil (Azeez et al., 2022; Cruz and Dierig, 2015), and a non-toxic alternative to castor (Ricinus communis) as a source for HFA used in the chemical industry (Patel et al., 2016). Recently, multiple isotopic tracing studies have indicated that Physaria developing seeds utilize both non-conventional and novel pathways to partition carbon into fatty acid biosynthesis and assemble oil containing HFA (Bhandari and Bates, 2021; Cocuron and Alonso, 2023). To further understand these novel metabolic pathways and improve Physaria seed oil and HFA yields, we need to understand the in planta function of genes involved in these processes. Tissue culture-based transformation to generate stable transgenic Physaria (Azeez et al., 2022; Chen et al., 2021) suffers from challenges with chimeric tissues and self-incompatibility (endogenous to Physaria) making production of homozygous knockdown lines laborious, low-efficiency and time-consuming (~18 months to generate homozygous T1). Virus-induced gene silencing (VIGS; Burch-Smith et al., 2004) is a powerful functional genomics method to introduce transient gene silencing into somatic cells of T0 plant tissues. The spread of VIGS throughout the plant can allow functional genomic studies even within the developing T1 seeds of the T0 plant, thus providing rapid functional evaluation of seed metabolism genes in one generation. However, spread of VIGS throughout the plant is random indicating not each seed will have the gene of interest silenced. Additionally, Physaria is self-incompatible and will not produce seeds without outcrossing or hand pollination of immature flowers. Therefore, we developed a VIGS approach that utilizes induced self-compatibility as a selectable marker to rapidly identify and characterize suppressed gene functions in developing Physaria seeds. Initial testing of VIGS in Physaria targeted phytoene desaturase (PDS) at the four-leaf stage and demonstrated low silencing efficiency (mild photobleaching, Figure S1). To improve the silencing efficiency at whole plant level, we utilized vacuum agroinfiltration of germinated seeds (Zhang et al., 2017) (Figure S2). Through parameter optimization (inoculum density, incubation time), we found ~97% PDS silencing efficiency at an agrobacterium OD of 0.4 and 5 min of vacuum agroinfiltration time (Figure S3a). Among the plants with photobleaching phenotypes, 34% were completely photobleached (whole plant silencing), whereas 63% had partial photobleaching phenotypes (Figure S3b). Approximately 10% of the seeds produced a photobleached phenotype in the next generation (Figure S4), indicating inoculation of germinated seeds with transient VIGS constructs can produce a silencing effect within developing seeds of the mature plant, and the VIGS within the mature seed tissue can carry over to the next generation. To test the VIGS efficiency on developing Physaria seed lipid metabolism genes, we selected two genes involved in HFA biosynthesis: FATTY ACID HYDROXYLASE 12 (FAH12) and FATTY ACID ELONGASE 1 (FAE1). FAH12 hydroxylates oleic acid (18:1) esterified to phosphatidylcholine and produces ricinoleic acid (18:1-OH). After 18:1-OH removal and esterification to Coenzyme A, FAE1 elongates the ricinoleic acid to lesquerolic acid (20:1-OH) (Chen et al., 2021) that is then incorporated into seed oil. Germinating seeds inoculated with FAH12-VIGS and grown to maturity produced seeds with increased 18:1 and reduced 20:1-OH as expected (Figure 1a). FAE1-VIGS produced seeds with higher 18:1-OH and less accumulation of the 20:1-OH as expected (Figure 1b). Thus, transient VIGS employing germinated Physaria seed transformation can be utilized for developing seed gene functional evaluation, is rapid (only 4 months), and a less laborious tool than published tissue culture-based methods to produce homozygous stable transformants. Although, two drawbacks remained: (i) Due to Physaria self-incompatibility, laborious hand pollination at immature flower stages is needed for the VIGS plants to produce seeds; (ii) due to the variability of TRV transport throughout the plant into developing seed tissue, many individual seeds needed to be screened for fatty acid content to evaluate the metabolic effect of the gene silencing (Figure 1a,b required screening seeds from 40 to 60 pollinated flowers per line). Thus, self-incompatibility and lack of a marker for which seeds contain the VIGS construct are time-consuming and labour-intensive limitations to producing a high-throughput functional genomics system for Physaria seed metabolism. In various Brassicaceae members, the S-locus receptor kinase (SRK) and S-locus cysteine-rich (SCR) proteins are the major factors inducing self-incompatibility (Nasrallah et al., 2002). The female determinant SRK expressed in the stigma acts as a receptor for pollen-expressed SCR ligand protein, their interaction results in rejection of self-pollen (Nasrallah et al., 2002). Physaria is a closely related Brassicaceae species to Arabidopsis; therefore, we evaluated the role of these two crucial genes in Physaria self-incompatibility. Physaria female determinant SRK was expressed primarily in the stigma and male determinant SCR in pollen (Figure 1c). Interestingly, Physaria is self-compatible at an early stage (before the flower opens) (Chen et al., 2011), the SRK expression was low at very early stage 1 and then highly upregulated at stage 2 consistent with our ability to produce homozygous plants through hand pollination only with immature stage 1 flowers (Figure 1d,e). Later, in stages 3 and 4, the SRK expression level is reduced, but stigmas are less receptive to self-pollen likely due to protein produced during stage 2. To functionally characterize the role of these genes in Physaria self-incompatibility, we prepared VIGS constructs to knockdown SRK and SCR. The SCR-VIGS lines produced no seed pods after self-pollination, similar to controls (Figure 1f; Figure S5). However, ~30% of the SRK-VIGS lines produced healthy seed pods after self-pollination (Figure 1g), indicating SRK is the key self-incompatibility determinant in Physaria. Therefore, we developed a self-incompatibility-based functional genomics tool by co-targeting of SRK and our gene of interest (GOI) in a single VIGS construct, eliminating the need for hand pollination and utilizing self-compatibility as a novel selectable marker by visual selection of enlarged developing pods (Figure 1h,i). To verify, we co-targeted FAE1 and SRK. Approximately, 20% of SRK-FAE1-VIGS lines successfully developed seed pods without outcrossing, had reduced SRK expression (Figure 1j) and mature seeds had the expected changes in 18 and 20 carbon hydroxy fatty acids similar to FAE1-VIGS alone (Figure 1b) including significantly more 18:1-OH and less 20:1-OH (Figure 1k,l). Therefore, by combining VIGS of SRK with a GOI in a single construct we developed a novel tool for visual selection of GOI silenced seed pods, thus reducing both the pollination and seed phenotyping labour required to screen various genes for their role in Physaria seed development, such as lipid metabolism. The SRK-GOI-VIGS is a rapid transient functional genomic approach to evaluate seed metabolism genes in a single generation. This technology can be replicated in other self-incompatible crops by co-targeting the GOI with the endogenous self-incompatibility factors within that species (Fujii et al., 2016). This work supported by the United States Department of Agriculture National Institute of Food and Agriculture grants #2020-67013-30899, #2023-67013-39022. The authors declare no financial conflict of interest. P.D.B. and A.A. conceived the research and wrote the article. A.A. designed and executed the experiments. The data that supports the findings of this study are available in the supplementary material of this article. Data S1 Experimental procedures Figure S1 Agrobacterium leaf-infiltration of pTRV2-PfePDS construct. Figure S2 Agrobacterium-mediated vacuum-infiltration of P. fendleri (germinated) seeds to Virus Induced Gene Silencing of target genes. Figure S3 Optimization of silencing efficiency by changing Agrobacterium densities and vacuum-infiltration time. Figure S4 pTRV2-PfePDS VIGS plants shows heritable photobleaching phenotype. Figure S5 Virus induced gene silencing of P. fendleri SCR produced no seed without out-cross. Table S1 List of primers used in this study Table S2 Physaria fendleri gene sequences used in this study Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.