Graft incompatibility between pepper and tomato can be attributed to genetic incompatibility between diverged immune systems

Abstract

Graft compatibility is the capacity of two plants to form cohesive vascular connections. Tomato and pepper are incompatible graft partners; however, the underlying cause of graft rejection between these two species remains unknown.We diagnosed graft incompatibility between tomato and diverse pepper varieties based on weakened biophysical stability, decreased growth, and persistent cell death using trypan blue and TUNEL assays. Transcriptomic analysis of cell death in the junction was performed using RNA-sequencing, and molecular signatures for incompatible graft response were characterized based on meta-transcriptomic comparisons with other biotic processes.We show that tomato is broadly incompatible with diverse pepper cultivars. These incompatible graft partners activate prolonged transcriptional changes that are highly enriched for defense processes. Amongst these processes was broad NLR upregulation and hypersensitive response. Using transcriptomic datasets for a variety of biotic stress treatments, we identified a significant overlap in the genetic profile of incompatible grafting and plant parasitism. In addition, we found over 1000 genes that are uniquely upregulated in incompatible grafts.Based on NLR overactivity, DNA damage, and prolonged cell death we have determined that tomato and pepper graft incompatibility is likely caused by a form of genetic incompatibility, which triggers a hyperimmune-response.

Keywords: Solanaceae; autoimmunity; graft compatibility; hypersensitive response; plant grafting; programmed cell death.

Figure 1:. Heterografted tomato and pepper combinations exhibit moderate survival, unstable stem integrity, and reduced growth.

The relationship between percent survival (y-axis) and percent break (x-axis) is shown for all graft combinations (a). Black dots denote self-grafts, grey dots denote heterografts. Self-grafted tomato is outlined in orange. Self-grafted pepper is outlined in green. Heterografts where the scion is tomato are outlined in purple. Heterografts where the stock is tomato are outlined in blue. The identity of each data point is labeled 1–13. Percent survival n=30; For bend test sample size see Table S1. The change in stem diameter 2cm above the graft junction between 30 and 0 DAG (scion) (b). The change in stem diameter 2 cm below the graft junction between 30 and 0 DAG (stock) (c). California Wonder abbreviated to Cali Wonder, Doux des Landes abbreviated to DDL. Biological replicates are depicted as jitter and described in Table S2. Kruskal–Wallis one-way analysis of variance was used to detect significant differences between self-and heterografted combinations. p-value <0.05.

Figure 2:. Heterografted pepper fails to form vascular connections and shows a significant decrease in size 30 DAG.

(a, c, e, g, i, k, m, o, q, s, w, u, w, y) Representative photographs and (b, d, f, h, j, l, n, p, r, t, v, x, z) confocal micrographs for self-grafted tomato (a-b), self-grafted habanero (c-d), tomato:Habanero (e-f), Habanero:tomato (g-h), self-grafted Doux des Landes (DDL) (i-j), tomato:DDL (k-l), DDL:tomato (m-n). self-grafted Cayenne (o-p), tomato:Cayenne (q-r), Cayenne:tomato (s-t), self-grafted California Wonder (CW) (u-v), tomato:CW (w-x), CW:tomato (y-z). Graft junctions were stained with propidium iodide and imaged on a confocal microscope. Pink arrows indicate a successful graft junction with a healed xylem, white arrows indicate a failed vascular reconnection and white Asterix highlight adventitious roots. All plant images have scale bars are 5 cm, and all micrograph scale bars are 1000 μm.

Figure 3:. Incompatible grafts contain persistent nonviable tissue over time.

(a-r) Representative images of 2.5 mm long graft junctions at 7, 14, and 21 DAG stained with Trypan Blue. A representative ungrafted tomato stem and the percent of non-viable tissue (NVT) are shown at 7 DAG (a, s), 14 DAG (g, y), and 21 DAG (m, ae). A representative ungrafted pepper stem and the percent of NVT at 7 DAG (b, t), 14 DAG (h, z), and 21 DAG (n, af). A representative self-graft tomato junction and the percent of NVT at 7 DAG (c, u), 14 DAG (i, aa), and 21 DAG (o, ag). A representative self-grafted pepper junction and the percent of NVT at 7 DAG (d, v), 14 DAG (j, ab), and 21 DAG (p, ah). A representative tomato:pepper junction and the percent of NVT at 7 DAG (e, w), 14 DAG (k, ac), and 21 DAG (q, aI). A representative pepper:tomato junction and the percent of NVT at 7 DAG (f, x), 14 DAG (l, ad), and 21 DAG (r, aj). Yellow arrows point to examples of deep tissue death; dashed lines signify the graft site; all junctions are 2.5 mm tall (a-r). (s-aj) The percent of cell death and (ak) the area of cell death in the junction of all graft combinations at 7, 14, and 21 DAG. Pink boxplots are 7 DAG, orange boxplots are 14 DAG, and purple boxplots are 21 DAG. Biological replicates are depicted as jitter (ak) as well as described in detail in Table S4.

Figure 4:. Developmental programmed cell death is present in all graft junctions regardless of compatibility.

A representative graft junction from (a,e) tomato:tomato, (b,f) pepper:pepper, (c,g) tomato:pepper, (d,h) pepper:tomato 14 DAG. (a-d) TUNEL fluorescein-12-dUTP-labeled DNA and autofluorescence are false-colored cyan. (e-h) the TUNEL fluorescence merged with propidium iodide (false-colored magenta) staining nucleic acid and cell walls. Pink arrows indicate a successful graft junction with healed xylem, and white arrows indicate a failed vascular reconnection. Examples of newly developed xylem are labeled (xy). All images are equal and the scale bar is 500 μm.

Figure 5:. Incompatible heterografts have prolonged differential gene regulation compared to self-grafts.

Differentially expressed genes (>1.5 or <−1.5, p-value<0.05) of each grafted tissue (compared to ungrafted) at each time point for tomato and pepper. Upregulated genes are shown in light colors and downregulated genes are shown in dark colors. Self-grafted scions are dark purple, self-grated stocks are light purple, heterografted scions are orange, and heterograft stocks are yellow. Each combination has 3–5 bio-replicates.

Figure 6.. Heterograft-specific upregulated genes are involved in defense response.

(a-b) Uniquely upregulated heterografted genes were determined by performing likelihood ratio testing (p<0.05) on ungrafted, self-graft scion, and heterografted scion as well as ungrafted, self-grafted stock, and heterografted stock tissue. The genes upregulated in only the heterograft tissue were used to perform GO enrichment. GO terms enriched in heterografted tomato tissue at 7, 14, and 21 DAG (a). GO terms enriched in heterografted pepper tissue at 7, 14, and 21 DAG (b). (c-d) Log-fold change of NLRs in grafted tissue compared to ungrafted tissue of tomato (c) and pepper (d). (e) The log-fold change of genes involved in hypersensitive response in grafted vs. ungrafted tissue. The log-fold change was scaled by row. The tissue is denoted by the colored columns where self-grafted scions are dark purple, self-grafted stocks are light purple, heterografted scions are orange, and heterografted stocks are yellow. The days after grafting were denoted by colored columns where 7 DAG are white, 14 DAG are grey, and 21 DAG are black. Asterisks denotes p-value<0.05 and log-fold change greater than |1.5|.

Figure 7.. Heterografted plants share many differentially expressed putative orthologs such as ERF114

(a) Putative orthologs upregulated at any given tissue/time point in both tomato and pepper. Orthogroups were determined between Solanum lycopersicum, Capsicum annum, and Arabidopsis thaliana using OrthoFinder, where each gene corresponded to an orthogroup. Upregulated genes for all graft combinations were determined in comparison to ungrafted stems. A shared ortholog was determined if upregulated genes (lfc >1.5, p-value<0.05) from both tomato and pepper at a common tissue/time point were linked to the same orthogroup. (b) Normalized read counts of SlERF114 and CaERF114 across time. Read counts for tomato and pepper were normalized, combined, and faceted by tissue type. Boxplot color denotes tissue origin; Ungrafted tissue is orange, scion tissue is purple, and stock tissue is pink.

Figure 8:. Grafting elicits unique and shared genetic processes with other biological stressors.

(a) Spearman Rank Correlation between 7 DAG samples, botrytis infection, herbivory, plant parasitism, and arbuscular mycorrhizal fungi (AMF) colonization. (b) Overlap of upregulated genes from four biological processes investigated. (c) Upset plot showing the overlap between upregulated genes from the biological processes: AMF, plant parasitism, insect herbivory, and fungal infection, scion or stock self-, and scion or stock heterografted tissue 7 DAG. (d-g) Overlap of upregulated genes from scion or stock of self- or hetero-grafted tissue at 7 DAG and all biological processes. (d) The overlap between grafting and AMF, (e) botrytis fungal infection, (f) herbivory, (g) and plant parasitism. The grey outline denotes the biological stressors, whereas AMF was used as a control.