Developments at the graft interface in homo- and hetero-grafts

Abstract

Gene expression changes induced during graft union formation (the first month after grafting) in grapevine have been studied using whole genome microarrays. The genes differentially expressed between the rootstock and graft interface tissues of homo-grafts (Cabernet Sauvignon (CS) grafted onto CS) were compared at 3 and 28 days after grafting (dag). Graft union formation was associated with the upregulation of genes involved in secondary metabolism, cell wall, wound responses and hormone signaling. These gene expression differences were associated with the accumulation of lignin, cellulose and callose in the callus cells. Superimposed upon this, hetero-grafting between two different grapevine genotypes resulted in the further upregulation of stress and/or defense responses at the graft interface. Here we discuss the limitations of the techniques used to study the developments at the graft interface to date and future research directions to understand graft union formation in plants.

Grafting is widely used in the agriculture. Commercial varieties of many vegetables and fruit crops (such as grapevine, plums and apples) are made up of two different genotypes, the rootstock and the scion, which are grafted together to improve plant adaptation to agronomic traits and biotic/abiotic stress responses.^1,2 In Europe, where viniculture supports a full range of economic activities, the cultivation of grapevine has required the obligatory grafting upon rootstocks since the end of the 19th century. Rootstocks were initially used to overcome phylloxera, a soil-dwelling insect pest introduced from America. Today rootstocks are used worldwide because rootstocks participate in environmental adaptation, contribute to the control of grapevine shoot growth (conferring differences in scion biomass/vigour), and are important for fruit quality and yield.³ Despite the widespread use of grafting, the sequence of the complex biochemical and structural process that lead to successfully graft union formation remains largely uncharacterized. We have been studying the gene expression differences involved in graft union formation in grapevine homo-grafts (cabernet sauvignon (CS) grafted with CS⁴) and hetero-grafts (CS grafted with rootstocks⁵).

Transcriptional Changes During Graft Union Formation Overlap Considerably with Gene Expression Changes during the Activation of Stem Growth in the Spring in Grapevine

In herbaceous plants, global changes in gene expression during the process of graft formation have been analyzed 0, 3, 7 and 14 d after grafting (dag) in hickory (Carya tomentosa) with cDNA-AFLP. Some genes related to auxin, cell cycle, metabolism and signal transduction are differentially expressed.⁶ Graft union development in Arabidopsis thaliana hypocotyl grafts has also been recently studied at the histological and transcriptional level and graft union development was shown to involve wound and hormone signaling and the clearing of cellular debris.⁷

When transcriptome of grapevine rootstock and graft interface tissues sampled 3 and 28 dag were studied in homo-grafts (CS grafted with itself), we observed that homo-graft union development in grapevine involved the upregulation of many genes involved in cell wall synthesis, wound responses, secondary metabolism and signaling.⁴ These gene expression differences were accompanied by the accumulation of callose, lignin and cellulose in the callus cells at the graft interface (Fig. 1.).

Figure 1. Accumulation of cell wall components in the callus tissue at the graft interface of grafted grapevine 21 d after grafting. 100 um longitudinal sections prepared using a razor blade on a microtome imaged with an Axiophot binocular microscope, Zeiss. (A, C, and E). Bright field images of the graft interface, (B) fluorescent image of A stained with aniline blue, arrows indicates the accumulation of callose in the callus, (D) fluorescent image of C stained with caloflur, arrows indicate the accumulation of cellulose in the callus, and (E) bright field image of stained with phloroglucinol, arrows indicate the accumulation of lignin in the callus.

In woody plants such as grapevine, grafting is traditionally performed on over-wintering stems in the spring so that the reactivation of wood growth (with changes at cellular physiological, anatomical and metabolic levels) and graft union formation occur in parallel. In our work, many genes were differentially expressed over time, from 3 to 28 dag, in both tissues; these changes could be related to the activation of stem growth and metabolic activity in the spring. This hypothesis is supported by the upregulation of genes associated with cell wall synthesis, cell organization and phloem/xylem development. However, many genes differentially regulated in the graft interface are specific to graft union formation. Indeed, there was an upregulation of gene expression in the graft interface tissue compared with the rootstock, particularly genes involved in cell wall synthesis, secondary metabolism and hormonal signaling, but also associated with defense and wound responses.⁴

Hetero-Grafting Triggers the Up-Regulation of Stress and/or Defense Genes at the Graft Interface during the First Month after Grafting

Up until very recently, the only gene expression studies published on grafting concerned homo-grafting where processes presumably similar to simple wound-like responses^6,7,4 seem to occur. We have recently published a description of the gene expression differences induced at the graft interface when two different genotypes are grafted together.⁵ We have shown that superimposed upon the gene expression differences occurring in homo-grafts, additional stress and/or defense related genes are differentially expressed.⁵ These gene expression changes were not accompanied by alteration in the histology of the graft interface in terms of callose, cellulose and lignin accumulation (data not shown).

Conclusions and Future Perspectives

Our data suggests that the cells at the graft interface are capable of detecting the presence of the non-self grafting partner suggesting that there is some degree of self- and non-self recognition in grafted plants. However, one limitation of the gene expression study on hetero-grafts was the difficulty to separate genes differentially expressed in response to hetero-grafting from those genes constitutively differentially expressed between the two grafting partners, as the tissues at the graft interface are a mixture of both genotypes. This problem could be overcome by using RNA sequencing as long as there are sufficient differences in the sequences of the two grafting partners to assign the differentially expressed genes to one of the two genotypes. Alternatively, in scion/rootstock combinations that have visibly distinguishable cells, the cells of the two genotypes could be harvested separately using laser microdissection and tissue capture.

Moreover we believe that the tissue scale gene expression studies described here probably overlook the vital changes occurring at the cellular level such as biochemical and hormonal responses or the formation of plasmodesmata connections at the actual graft interface. There is also need to improve our understanding of the morphological developments at the graft interface; the graft interface develops in three dimensions so classical histology techniques do not accurately represent the macro-scale tissue developments. There have been a number of three dimensional studies of the graft interface,^8,9 but to date there has been no high resolution 3-dimensional imaging of the graft union formation in any plant species. In summary, further work is required to determine how the two different genotypes of grafted plants communicate and how graft union formation is organized at the cellular and tissue levels.

Cookson SJ, Clemente Moreno MJ, Hevin C, Nyamba Mendome LZ, Delrot S, Trossat-Magnin C, Ollat N. Graft union formation in grapevine induces transcriptional changes related to cell wall modification, wounding, hormone signalling, and secondary metabolism. J Exp Bot. 2013;64:2997–3008. doi: 10.1093/jxb/ert144.