Conferring high‐temperature tolerance to nontransgenic tomato scions using graft transmission of RNA silencing of the fatty acid desaturase gene

Summary

We investigated graft transmission of high‐temperature tolerance in tomato scions to nontransgenic scions from transgenic rootstocks, where the fatty acid desaturase gene (LeFAD7) was RNA‐silenced. Tomato was transformed with a plasmid carrying an inverted repeat of LeFAD7 by Agrobacterium. Several transgenic lines showed the lower amounts of LeFAD7 RNA and unsaturated fatty acids, while nontransgenic control did not, and siRNA was detected in the transgenic lines, but not in control. These lines grew under conditions of high temperature, while nontransgenic control did not. Further, the nontransgenic plants were grafted onto the silenced transgenic plants. The scions showed less of the target gene RNA, and siRNA was detected. Under high‐temperature conditions, these grafted plants grew, while control grafted plants did not. Thus, it was shown that high‐temperature tolerance was conferred in the nontransgenic scions after grafting onto the silenced rootstocks.

Introduction

Plants are subjected to a wide range of temperatures. High temperatures increase the fluidity of the membrane, and the membrane lipids of plants are replaced by unsaturated fatty acids to maintain appropriate fluidity of the membrane (Los et al., 2013). Previous studies demonstrated that fatty acid desaturase 7 (FAD7) affects the composition of membrane lipids (Kim et al., 2001; Maienza et al., 2013; Murakami et al., 2000). Plants with an altered saturation level of lipids show different temperature sensitivities at various stages of growth. For example, fad7 and fad8 double silenced tobacco with a lower level of unsaturated fatty acids was better able to acclimate to higher temperatures (Murakami et al., 2000).

Tomato is a summer crop in temperate zones; however, high temperatures decrease tomato production. Some previous studies demonstrated that FAD affects the composition of membrane lipids and responsiveness to temperature stress (Maienza et al., 2013; Wang et al., 2010). LeFAD3 is the endoplasmic reticulum‐localized FAD gene. LeFAD3‐silenced tomato decreased 18 : 3 fatty acid and increased 18:2 fatty acid, and the fresh weight of silenced plants was higher than that of wild‐type plants after high‐temperature treatment (Wang et al., 2010). The suppressor of prosystemin‐mediated responses 2 (Spr2/LeFAD7) was isolated in tomato (Li et al., 2003). spr2 impaired jasmonate synthesis from 18:3 fatty acid and compromised defence against tobacco hornworm larvae (Li et al., 2003). More recently, FAD7‐silenced tomato (Micro‐Tom) using a portion of tobacco FAD7 as a gene‐silencing trigger improved seed germination, hypocotyl elongation and fruit yield under high‐temperature conditions (Maienza et al., 2013).

Post‐transcriptional gene silencing is a natural cellular process in a broad range of eukaryotic organisms including fungi, animals and plants (Mlotshwa et al., 2002). RNAi is a form of sequence‐specific gene silencing induced by introducing homologous double‐strand RNA (dsRNA) into a target gene. In the process of RNAi, small interfering RNA (siRNA; from 21 to 26 nucleotides in length) is produced from dsRNA and then suppresses the target gene expression due to binding to RISC, which degrades the target RNA (Baulcombe, 2004). The siRNA has been thought to move systemically via the phloem in plants (Waterhouse et al., 2001). This phenomenon was found in grafted plants of Arabidopsis, tobacco and others (Molnar et al., 2010). In a recent study, RNA silencing was induced in nontransgenic tobacco scions after grafting onto silenced rootstocks, and virus resistance was conferred to the nontransgenic scions (Ali et al., 2013).

In this study, we describe the production of LeFAD7‐silenced transgenic tomato lines and the graft transmission of RNA silencing from LeFAD7‐silenced rootstocks to nontransgenic, non‐LeFAD7‐silenced scions to confer temperature tolerance in scions.

Results

Characterization of LeFAD7

To characterize LeFAD7 (Li et al., 2003), the sequences of LeFAD were downloaded from the tomato genome database (http://solgenomics.net/) to complete a phylogenetic tree. Twelve genes were selected as LeFAD and 3 corresponded to ω‐3 fatty acid desaturase (Figure 1a). These proteins share the Δ(12) FAD domain that can introduce a second cis double bond at the Δ (12) position of the fatty acid (Figure 1b). LeFAD7ORF is 1305‐bp long, encoding 435 amino acids in length (Figure 1c). The protein sequences of LeFAD7, Solyc06 g007130.2 and Solyc06 g007140.2 share high identities (>80%).

Figure 1.

Phylogenetic tree of LeFAD genes and an open reading frame of the LeFAD7 gene. (a) The phylogenetic tree was constructed based on the protein sequences of LeFADs. The scale bar represents 0.1 substitutions per site. (b) Structure of LeFAD family. The yellow boxes indicate conserved domains in each FAD subunit. (c) Sequence of LeFAD7. Capital and lower‐case letters represent ORF and nontranslated regions, respectively. Underlined sequences represent the silencing trigger region of LeFAD7 ORF and UTR, respectively.

Production of LeFAD7 transgenic tomato

The regenerated plants in the presence of kanamycin were subjected to PCR analysis to confirm the transgene (Figure 2b). A total of 66 transgenic lines were obtained, including 10 and 6 for Micro‐Tom LeFAD7ORF and UTR, respectively, and 25 and 25 for ‘Block’ LeFAD7ORF and UTR, respectively. We selected 51 LeFAD7‐silenced lines of transgenic tomato: 7 and 5 lines for LeFAD7ORF and UTR of Micro‐Tom, respectively, and 18 and 21 for LeFAD7ORF and UTR of ‘Block’, respectively (Table S1). LeFAD7 expression and siRNA were tested using RT‐PCR and northern blot analysis, respectively (Figure 2c and d). Several transgenic lines showed less LeFAD7 RNA compared with control tomato, and siRNA was detected (Figure 2c and d). FAD7 is responsible for trienoic fatty acid formation in leaf tissues (Murakami et al., 2000). Unsaturation of monogalactosyldiacylglycerol (MGDG) of the 3 transgenic lines was tested (Figure 2e). As a result, in the transgenic lines, lower and higher amounts of MGDG 36 : 6 and MGDG 34 : 6 (black and horizontal stripe), and of MGDG 36 : 4 and MGDG 34 : 4 (white and vertical stripe) were observed compared with those in control tomato (Figure 2e, Table 1). Thus, these results reflect the effect of silenced LeFAD7 on fatty acid desaturation. Among them, 3 and 5 lines of Micro‐Tom and ‘Block’, respectively, were further investigated.

Figure 2.

Plasmid constructs for transformation and analysis of transgenic tomato. (a) Constructs for transformation. Sequences of LeFAD7 ORF and LeFAD7 UTR are cloned as an inverted repeat, respectively. Left, pLeFADORF RNAi and right, pLeFAD7UTR RNAi. (b) PCR analysis of transgenic lines for the presence of neomycin phosphotransferase II ( NPTII ). Lines are B7O1, B7O2 and B7O3; and B7U1, B7U2 and B7U3 for LeFAD7 ORF and LeFAD7 UTR, respectively. (c) RT‐PCR analysis of accumulation level of LeFAD7 ORF transcripts. Lines are B7O1, B7O2, B7O3, B7U1, B7U2 and B7U3 for LeFAD7 ORF. The expression level of the tomato ACTIN gene was used as an internal control. (d) siRNA detection in transgenic lines. siRNAs of LeFAD7 ORF and UTR were detected using a DIG‐labelled RNA probe specific to each silencing trigger region, respectively. Lines are B7O1, B7O2 and B7O3; and B7U1, B7U2 and B7U3 for LeFAD7 ORF and LeFAD7 UTR, respectively. The amount of tRNA was used as an internal control. (e) Fatty acid composition of transgenic lines. The degree of MGDG was determined by LC‐MS using the leaves of transgenic lines. The graph shows the average value of fatty acid composition from these lines; M7O3, M7O4 and M7O5; and M7U2, M7U3 and M7U4 for LeFAD7 ORF and LeFAD7 UTR in Micro‐Tom (left), respectively. The graph also shows the average value of fatty acid composition from these lines; B7O1, B7O2 and B7O3; and B7U2, B7U3 and B7U4 for LeFAD7 ORF and LeFAD7 UTR in ‘Block’ (right), respectively.

Table 1.

Fatty acid composition of transgenic plants and of scions grafted onto silenced rootstocks

Line	MGDG 34:4	MGDG 34:6	MGDG 36:4	MGDG 36:6
Micro‐Tom
WT	2.26 ± 0.56	30.12 ± 0.85	3.02 ± 0.139	42.07 ± 5.38
LeFAD7 ORF	13.57 ± 2.51*	1.19 ± 0.70*	45.98 ± 3.74*	2.53 ± 1.10*
LeFAD7 UTR	13.69 ± 4.45*	3.07 ± 2.44*	36.83 ± 10.26*	5.61 ± 4.01*
‘Block’
WT	2.07 ± 0.18	26.35 ± 1.58	3.55 ± 0.26	41.96 ± 1.93
LeFAD7 ORF	20.67 ± 3.48*	0.60 ± 0.26*	42.61 ± 3.27*	1.75 ± 0.60*
LeFAD7 UTR	17.20 ± 3.21*	2.73 ± 2.38*	37.29 ± 6.07*	4.70 ± 3.56*
Grafted plants
Momotaro‐WTa	1.43 ± 0.39	29.87 ± 1.43	2.59 ± 0.74	45.39 ± 3.65
Momotaro‐LeFAD7 ORF	11.5 ± 4.40*	1.28 ± 1.18*	29.22 ± 9.09*	2.57 ± 1.98*
Momotaro‐LeFAD7 UTR	16.5 ± 6.01*	1.35 ± 2.06*	44.94 ± 13.9*	3.49 ± 4.45*

An asterisk indicates a statistically significant difference between WT and transgenic plants and between grafted plants of control and LeFAD7 transgenic plants as rootstocks (Student's t‐test: P < 0.01).

^aScion, Momotaro; rootstock, WT. Data are from total number of plants: Micro‐Tom; 3 WT, 6 LeFAD7 ORF, 6 LeFAD7 UTR: ‘Block’; 3 WT, 9 LeFAD7 ORF, 9 LeFAD7 UTR, grafted plants: 3 Momotaro‐WT, 12 Momotaro‐LeFAD7 ORF, 9 Momotaro‐LeFAD7 UTR. Momotaro represents House Momotaro.

High‐temperature tolerance of LeFAD7‐silenced tomato

We investigated the high‐temperature tolerance of those silenced lines. First, seeds were sown on Petri plates and were transferred to soil for 2 weeks, and then, the seedlings were subjected to heat treatment of 40 °C/35 °C (day/night) for 2 weeks. Control Micro‐Tom showed yellowing, while the silenced lines (LeFAD7ORF or LeFAD7UTR) showed mild yellowing (Figure 3a). Moreover, the silenced lines grew by 2–3 cm during the high‐temperature treatment, while control did not (Figure 3b). The control and the silenced lines of ‘Block’ were also treated at 40 °C/35 °C for 2 weeks, and those plants showed severe yellowing or withered (data not shown). At 40 °C/35 °C for a week, the silenced lines (LeFAD7ORF or LeFAD7UTR) showed mild yellowing, while control showed severe yellowing or withered (Figure 3c). Moreover, the silenced lines grew by 3–5 cm during the high‐temperature treatment, while control did not (Figure 3d). Thus, the silenced lines of both cultivars showed the high‐temperature tolerance and could grow during high‐temperature treatment at 40 °C/35 °C.

Figure 3.

High‐temperature tolerance of LeFAD7‐silenced plants. (a) Silenced lines of Micro‐Tom were grown at 40 °C/35 °C (day/night) for 2 weeks. Photographs were taken immediately after the high‐temperature treatment. Lines are M7O3 and M7U3 for LeFAD7 ORF and LeFAD7 UTR, respectively. (b) Growth of the plants during high‐temperature treatment. The graph shows the average length of growth from these lines; M7O3, M7O4 and M7O5; and M7U2, M7U3 and M7U4 for LeFAD7 ORF and LeFAD7 UTR, respectively. (c) Silenced lines of ‘Block’ were grown at 40 °C/35 °C (day/night) for a week. Photographs were taken immediately after the high‐temperature treatment. Lines are B7O2 and B7U3 for LeFAD7 ORF and LeFAD7 UTR, respectively. (d) Growth of the plants during the high‐temperature treatment. The graph shows the average length of growth from these lines; B7O1, B7O2, B7O3, B7O4 and B7O5; and B7U1, B7U2, B7U3, B7U4, B7U6 and B7U7 for LeFAD7 ORF and LeFAD7 UTR, respectively. An asterisk indicates a significant difference between the samples (Student's t‐test: P < 0.01).

Graft transmission of RNA silencing of LeFAD7 from rootstocks to nontransgenic scions

Tomato scions of House Momotaro were grafted onto silenced rootstocks of ‘Block’. Six weeks after grafting, total RNA was extracted from the leaves of both scions and rootstocks followed by RT‐PCR analyses using the combination of primers: LeFAD7ORF‐F/LeFAD7ORF‐R for LeFAD7ORF and LeFAD7UTR, respectively. The scions of the grafted plants showed less of the target gene mRNA than that in the scions grafted onto the control rootstocks (Figure 4a). Furthermore, siRNA of LeFAD7ORF or LeFAD7UTR was detected using either specific probe ORF or UTR, although the amount of siRNA was less compared with that in each rootstock (Figure 4b). The unsaturation of MGDG in the scions onto the rootstocks of LeFAD7ORF or LeFAD7UTR was tested (Figure 4c). The scions showed less of MGDG 36:6 (black) and MGDG 34:6 (horizontal stripe), respectively, and more of MGDG 36:4 (white) and MGDG 34:4 (vertical stripe) compared with those in the control grafted plants (Figure 4c, Table 1). These results show that the degree of unsaturated fatty acids decreased in the scions after grafting.

Figure 4.

Analysis of grafted plants. (a) The accumulation of LeFAD7 transcript by RT‐PCR. Rootstocks and scions were analysed in grafted plants, respectively (indicated by asterisk). Lines B7O1 and B7U4 as rootstocks were used for LeFAD7 ORF. The expression level of the tomato ACTIN gene was used as an internal control. (b) siRNA detection. Rootstocks and scions were analysed in grafted plants, respectively (indicated by asterisks). siRNAs were detected using a DIG‐labelled RNA probe specific to each silencing trigger region. Lines B7O1 and B7U4 as rootstocks were used for LeFAD7 ORF and LeFAD7 UTR, respectively. The amount of tRNA was used as an internal control. (c) Fatty acid composition. Unsaturation of MGDG was determined by LC‐MS using the leaves of transgenic lines. Rootstocks and scions were analysed in grafted plants, respectively (indicated by asterisks). Lines B7O1, B7O2 and B7O3; and B7U2, B7U3 and B7U4 as rootstocks were used for LeFAD7 ORF and LeFAD7 UTR, respectively. Momotaro represents House Momotaro.

High‐temperature tolerance of scions after grafting

Because the silencing of LeFAD7 in grafted scions was induced, we investigated the high‐temperature tolerance of those grafted plants. After treatment of 40 °C/35 °C for a week, the scions grafted onto the silenced rootstocks showed mild yellowing, while the control scions grafted onto the nonsilenced rootstocks showed severe yellowing (Figure 5a). Moreover, the scions onto the silenced rootstocks grew up by 3–5 cm during the high‐temperature treatment, while the control scions did not (Figure 5b). These results show high‐temperature tolerance in the scions grafted onto the silenced rootstocks.

Figure 5.

High‐temperature tolerance of grafted plants. (a) Grafted plants were grown at 40 °C/35 °C (day/night) for a week. Photographs were taken immediately after the high‐temperature treatment. Lines B7O1 and B7U4 as rootstocks were used for LeFAD7 ORF and LeFAD7 UTR, respectively. (b) Growth of the plants during the high‐temperature treatment. An asterisk indicates a significant difference between the samples (Student's t‐test: P < 0.01). Lines B7O1, B7O2 and B7O3; and B7U2, B7U3 and B7U4 as rootstocks were used for LeFAD7 ORF and LeFAD7 UTR, respectively. Momotaro represents House Momotaro.

Discussion

In this report, transgenic tomato plants silenced with LeFAD7 were produced and showed high‐temperature tolerance. We further tested whether the nontransgenic tomato scions grafted onto the silenced rootstocks also showed high‐temperature tolerance. Previous reports have shown that Arabidopsis FAD7‐silenced tobacco showed high‐temperature tolerance (Murakami et al., 2000). In our data, the LeFAD7‐silenced plants reduced target gene expression and the higher amounts of MGDG 36 : 4 and MGDG 34 : 4 (white and vertical stripe) compared with those in control tomato (Figure 2e, Table 1) coincide with the results of the spr2 plant (Li et al., 2003). We produced and evaluated many LeFAD7‐silenced lines. However, no relationship was observed between the degrees of LeFAD7 silencing (siRNA level) and the degrees of unsaturation of MGDG. For example, lines M7O3 and M7O5 were different in LeFAD7 mRNA and the amount of siRNA: however, there was little difference in the amount of unsaturated MGDG and growth under high‐temperature (Table S1). Moreover, the nontransgenic scions grafted onto the silenced rootstocks were different in the amount of siRNA compared with the silenced rootstocks (Figure 4b). Nevertheless, the scions and the rootstocks were not different in the amount of MGDG and growth under high‐temperature (Figures 4c, and 5, Table 1). The decreased level of LeFAD7 mRNA clearly affected the degree of unsaturation of MGDG and high‐temperature tolerance, even with a slight decrease in LeFAD7 mRNA. These results implicate that even a slight reduction in LeFAD7 mRNA affects the unsaturation of fatty acids as well as temperature tolerance.

The high‐temperature tolerance of the LeFAD7‐silenced plants and the scions grafted onto silenced rootstocks were evaluated at the adult stage of development (Figures 3 and 5). High‐temperature tolerance at the adult stage could be beneficial for plant growth in summer. The transgenic tomato carrying the inverted repeat of tobacco FAD7 as a gene‐silencing trigger reduced LeFAD7 mRNA and high‐temperature tolerance (Maienza et al., 2013). Because FAD7 is conserved among some plant species (Figure 1a), the silencing plasmid construct in this study could be used to confer high‐temperature tolerance in other plant species. The variety of ‘Block’ used in this study has multiple resistances against plant diseases (http://www.sakataseed.co.jp/product/search/code068700.html). Therefore, it is thought that the scions grafted onto the silenced plants carry dual traits for high‐temperature tolerance and disease resistance. Furthermore, it is indicated that the spr2 plant was compromised in defence against attack by tobacco hornworm larvae (Li et al., 2003). Characterizing the defence in LeFAD7‐silenced plants against diseases and insects is the subject of further study. If other cultivars of tomato are used from transformation with this plasmid construct, it would be possible to produce transgenic tomato with other traits in addition to high‐temperature tolerance.

Grafting is a very old technique to confer desired traits such as disease resistance to scions in agriculture. Since graft transmission of RNA silencing (Palauqui et al., 1997) was reported. This technique has been more attractive for plant breeding. Lusser et al. (2012) reported that the grafting was one of the seven new plant‐breeding techniques and described the current situation of them from the point of view of commercialization and regulations. We previously reported the induction of high virus resistance in the nontransgenic tobacco scions grafted onto the rootstocks silenced with the tobacco endogenous genes, which are required for the virus multiplication (Ali et al., 2013). Here, the successful induction of high‐temperature tolerance in nontransgenic tomato scions grafted onto the rootstock silenced with LeFAD7. These studies may induce higher acceleration of this technique for crop improvements. More recently, the grafting causes the horizontal gene transfer of chloroplasts/plastids and even entire nuclear genome between cells at grafting junction (Fuentes et al., 2014; Stegemann et al., 2012; Thyssen et al., 2012). Although the underlined mechanism for gene transfer remains unclear, this technique could be more broadly useful for plant breeding, not only producing grafted plants with desired traits but also generation of different ploidy species with/without different set of chloroplasts.

Experimental procedures

Isolation of LeFAD7 and plasmid construction

The sequence of the LeFAD7 gene was downloaded from the tomato genome database (http://solgenomics.net/). A phylogenetic tree was constructed based on multiple alignment of protein sequences of LeFAD using ClustalW (http://clustalw.ddbj.nig.ac.jp/) and was displayed with Tree View (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html). The silencing trigger region spanning the sequence between +1 and +335 (the adenosine of the start codon was designated +1) of LeFAD7ORF (Figure 1b) and the sequence between +1309 and +1531 of LeFAD7UTR (Figure 1b) was amplified using PCR with DNA polymerase (KOD Plus; Toyobo, Osaka, Japan) and Micro‐Tom genomic DNA and each specific primers (LeFAD7ORF‐F/LeFAD7ORF‐R and LeFAD7UTR‐F/LeFAD7UTR‐R, respectively, listed in Table 2), and then, the amplified DNA fragments were cloned into pENTR^TM/D‐TOPO to make each entry clone following the manufacturer's manual (Invitrogen, Carlsbad, CA). The final RNA‐silencing vectors were produced by an LR clonase reaction between an entry vector and the pANDA vector (Figure 2a, Miki and Shimamoto, 2004). Those silencing clones were introduced into Agrobacterium tumefaciens MP90 strain (Chetty et al., 2013).

Table 2.

Synthetic oligonucleotides

Name	Sequence
LeFAD7ORF‐F	CACCATGGCAAGTTGGGTTCTATCAG
LeFAD7ORF‐R	CTCATGGACGTCCATGGATT
LeFAD7UTR‐F	CACCAAAACCCTGTTAGCTGATTATTT
LeFAD7UTR‐R	GATGGTGGGAACTGAAGTATAT
NPTII‐F	ATGATTGAACAAGATGGATTGCA
NPTII‐R	GAAGAACTCGTCAAGAAGG
ACTIN‐F	TGGCATCATACTTTCTACAATG
ACTIN‐R	CTAATATCCACCTCACATTTCAT
T7‐LeFAD7ORF‐R	TAATACGACTCACTATAGGGAGACTCATGGACGTCCATGGATTT
T7‐LeFAD7UTR‐R	TAATACGACTCACTATAGGGAGAGATGGTGGGAACTCAAGTATAT

Transformation of tomato plants

Transformation of tomato plants was performed according to the method of transformation of chrysanthemum (Shinoyama et al., 1998) with some modifications. Tomato, Solanum esculentum cv. Micro‐Tom and ‘Block’ (Sakata Seed Inc., Kanagawa, Japan), was used for the transformation. Seeds were germinated on germination medium (0.15% Hyponex, 0.5% sucrose and 0.5% gellan gum) and incubated. The explants were excised from the cotyledons of the 7‐day‐old seedlings and incubated on wet filter paper for a day. The explants were then mixed with a culture of Agrobacterium harbouring the plasmid and incubated on medium (MS, 3% sucrose, 1 mg/L zeatin, 0.1% casamino acid and 0.3% gellan gum) for 2 days. The explants were then transferred to a selection medium (MS, 3% sucrose, 1 and 2 mg/L zeatin for Micro‐Tom and ‘Block’, respectively, 25 mg/L meropenem, 100 mg/L kanamycin and 0.3% gellan gum) until the shoots appeared. The regenerated shoots were further transferred to a root‐inducing medium (1/2 MS, 1.5% sucrose, 25 mg/L meropenem, 100 mg/L kanamycin and 0.5% gellan gum) for 2 weeks.

DNA extraction and PCR amplification

DNA was extracted from the leaves of the plants using a cetyltrimethylammonium bromide‐based (CTAB‐based) extraction procedure (Doyle and Doyle, 1987). The quantity and purity of DNA were measured using a spectrophotometer (NanoDrop; Thermo Fisher Scientific, Waltham, MA). Genomic DNA (200 ng) was used as a template for PCR. The NPT II‐F and NPT II‐R primers (Table 2) were used to amplify NPT II. PCR products were analysed by electrophoresis on 1% agarose gel.

RNA extraction and RT‐PCR analysis

Total RNA was extracted from the leaves of the plants using a Sepasol‐RNA I Super G (Nacalai Tesque, Kyoto, Japan) following the manufacturer's manual, and the residual genomic DNA was removed using DNase I treatment (Promega, Madison, WI). Total RNA (1 μg) was reverse‐transcribed using RevertAid^™ reverse transcriptase (Thermo Fisher Scientific) following the manufacturer's manual. The cDNA was amplified using PCR under the following conditions: 94 °C for 30 s, followed by 27 cycles of amplification for LeFAD7 and 24 cycles for ACTIN (94 °C for 30 s, 56 °C for 30 s, 72 °C for 1 min) with each specific primers (Table 2).

Northern blot analysis

Total RNA (5 μg) was separated using acrylamide gel electrophoresis and electrotransferred to a membrane as described previously (Shimizu et al., 2009). Hybridization and siRNA detection were carried out with the DIG‐labelled FAD7ORF and FAD7UTR RNA probe constructs using a DIG RNA labelling kit (Roche Diagnostics, Basel, Switzerland) following the manufacturer's manual.

Fatty acid analysis

Sample preparation and mass spectrometry were performed as described previously (Okazaki et al., 2015). The data on the MGDG possessing 34 and 36 carbons were then analysed and graphed according to the number of double bonds between the carbons. Differences between the average of control and transgenic plants and grafted plants onto LeFAD7 silenced rootstocks were tested using a Student's t‐test (P < 0.01).

Plant growth conditions and high‐temperature treatment

Except for the high‐temperature treatment experiments, Micro‐Tom and ‘Block’ were cultured at 25 °C/20 °C under 16L/8D light conditions. In the high‐temperature treatment experiments, the tomato seeds were germinated in plant box on germination medium. Ten days after germination, the seedlings were transferred to soil (Protoleaf Inc., Tokyo, Japan). Four‐week‐old plants or 3‐week‐old herbaceous cuttings were measured a height of shoot and transferred to high‐temperature conditions (40 °C/35 °C under 16L/8D light conditions). The grafted plants onto ‘Block’ rootstocks and Micro‐Tom plants were exposed to high‐temperature conditions for one and 2 weeks, respectively. These plants (Micro‐Tom; 5 WT, 16 LeFAD7 ORF, 10 LeFAD7 UTR: ‘Block’; 12 WT, 13 LeFAD7 ORF, 19 LeFAD7 UTR: grafted plants; 8 Momotaro‐WT, 9 Momotaro‐LeFAD7 ORF, 10 Momotaro‐LeFAD7 UTR) were used for measuring a height of shoot before and after the high‐temperature treatment. Differences between control and transgenic plants and between scions of control and LeFAD7 rootstocks were tested using a Student's t‐test (P < 0.01).

Grafting

The tomato plants, cv. ‘Block’ and House Momotaro (Takii Seed Co., Ltd, Kyoto, Japan), were cultured at 25 °C/20 °C under 16L/8D light conditions. The seedlings were transferred to soil. Approximately 3‐week‐old plants were used for the cleft grafting, as described previously (Sonoda and Nishiguchi, 2000). Rootstocks were prepared by removing the shoots above at least the basal leaves, and then, the centre of the stem was cut vertically (2 mm). Scions were prepared by removing leaves and trimming the base of the scion to a wedge. The scion/stock junction was wrapped with Parafilm and a clip. Grafted plants were covered with plastic bags to avoid dehydration for 2 weeks or until the graft was completed.

Supporting information

Table S1 List of transformants.