We previously reported about the identification of novel
Sletr1 alleles (
Sletr1-1 and
Sletr1-2) showing ethylene insensitivity and impaired fruit ripening phenotype at different levels. Segregation analysis in the F
2 populations suggested an association between these alleles and ethylene insensitive phenotype (
Okabe et al. 2011). In this study, to confirm whether
Sletr1 mutations confer reduced ethylene sensitivity and to discuss the availability of Micro-Tom mutant library combined with TILLING in mutation breeding of tomato, we characterized transgenic tomato lines over-expressing mutant versions of
SlETR1,
Sletr1-1 and
Sletr1-2.
In the seedling ethylene triple response analysis, both of T
1 transgenic lines showed different levels of ethylene insensitive phenotype depending on the transgene type. No significant difference in the hypocotyl length was observed in the
Sletr1-1 transgenic lines compared to the
Sletr1-1 mutant line. Although the root length of the
Sletr1-1 transgenic lines showed significant difference, the values were close to that of the
Sletr1-1 mutant line. This phenotype is similar to what had been observed in transgenic tomato plants expressing
Nr (
Wilkinson et al. 1995). In contrast to the
Sletr1-1 transgenic lines, various levels of reduced ethylene sensitivity were observed in the
Sletr1-2 transgenic lines (). This observation is likely to correspond to the ethylene insensitivity level of the two
Sletr1 alleles, since the expression level of the transgene was similar in all tested transgenic lines as shown in . These results indicate that
Sletr1-1 confers nearly complete ethylene insensitive phenotype, whereas
Sletr1-2 confers reduced ethylene sensitive phenotype.
Regarding fruit phenotypes of the transgenic lines, although we initially expected that the
Sletr1-2 transgenic lines would display weaker ripening impairment compared to the
Sletr1-1 transgenic lines,
Sletr1-2 transgenic lines were similarly impaired in fruit ripening and shelf life (, ). These phenotypes resembled that of transgenic tomato plants expressing
Arabidopsis etr1-1 (
Wilkinson et al. 1997). It is suggested that the enhancement of ethylene insensitivity affected differently to various organs (i.e. fruit, hypocotyl and root), thus resulted in nonuniform ethylene sensitivity. It is also speculated that the fruit ripening process is more sensitive to ethylene, since the enhanced expression of the weaker allele (
Sletr1-2) resulted in the fully inhibition of ripening. Furthermore, one possible explanation for the difference of fruit ripening phenotype between
Sletr1-2 mutant and
Sletr1-2 transgenic lines is that fruit ripening inhibition in transgenic tomato fruits may depend on the
Sletr1 expression level, because
SlETR1 is expressed in all tissues including fruit at a constant level (
Kevany et al. 2007,
Lashbrook et al. 1998).
So far, it has been reported that genetic engineering of ethylene-mediated biological regulation including ethylene biosynthesis, perception and signaling is available for manipulating fruit ripening. Suppression of ethylene biosynthesis genes in previous studies have revealed that antisense transgenic lines of tomato
ACC gene had resulted in decreased ethylene production in fruits and impaired fruit ripening (
Oeller et al. 1991), RNAi transgenic lines of ACC oxidase (
ACO) gene, which is a rate-limiting enzyme of ethylene biosynthesis had resulted in decreased ethylene production, delayed fruit ripening, and prolonged fruit shelf life (
Xiong et al. 2005). Similarly, with respect to ethylene perception and signaling, transgenic tomato lines constitutively expressing
Arabidopsis etr1-1 exhibited reduced ethylene sensitivity and delayed fruit ripening (
Wilkinson et al. 1997), fruit specific suppression of
SlETR4 gene using the
E8-promoter resulted in early ripening phenotype (
Kevany et al. 2008), antisense transgenic lines of
ethylene response factor 1 (
ERF1) gene showed longer fruit shelf life (
Li et al. 2007). However in many cases, these transgenic plants displayed excessive ripening inhibition or undesired developmental effect by the ectopic expression or global suppression of transgenes under the CaMV 35S promoter. Our data also showed that ripening inhibition occurred even in the
Sletr1-2 transgenic lines (). From the standpoint of breeding, a material showing severe ripening inhibition is not suited for practical use. A case study was successful in generating transgenic tomato plants with altered ripening but not an agronomic penalty by using a fruit-specific
E8-promoter (
Kevany et al. 2008). Another study demonstrated that transgenic tomato plants expressing
Arabidopsis mutant ethylene receptor
etr1-1 using an inducible promoter conferred reduced ethylene sensitivity depending on the concentration of inducer (
Gallie 2010). Utilization of such approach is not widespread due to the limited availability of appropriate or flexible promoters that can induce the expression of transgene at expected level. Therefore, production of the numerous variations in target traits by transgenic approach could be realistically difficult.
Contrary to genetic engineering of agronomic traits by transgenic approach, TILLING allows obtaining a broad range of variants from EMS mutagenized population. As a beneficial property of EMS mutagenesis, the resulting population theoretically includes various levels of mutants, such as null and leaky mutant of a target gene. It is also considered that TILLING approach in mutation breeding is more efficient as a strategy compared to the transgenic approach in which selecting different kinds of promoter is necessary to create variations in targeted agronomic trait. We previously identified the
Sletr1-2 allele by TILLING as a potential breeding material for improving the fruit shelf life without agronomic penalties such as excessive ripening inhibition and apparent loss of color in ripen fruit (
Okabe et al. 2011). We consider that if we did not perform TILLING, the
Sletr1-2 allele probably would not have been selected from the Micro-Tom mutant population, since the fruit ripening phenotype of
Sletr1-2 plant is not easily distinguished from wild type when the mutant populations are grown in large scale.
Furthermore, to manage the field trials, the phenotype screening of 10,000 tomato M
2 families consisting of twenty individuals per family in common tomato cultivar would require approximately 20 hectares of field space, i.e., one individual/m
2 (
Giovannoni 2007). A mutant screen in such scale would not be realistic for majority of laboratories. In contrast, the plant size of Micro-Tom enables growing in limited spaces.
Meissner et al. (1997) showed that Micro-Tom can be grown at high density, up to 1,357 individuals/m
2. Thus, the combination of Micro-Tom mutant library and TILLING permitted to efficiently isolate a valuable material, such
Sletr1-2. This approach could be applied to other important agronomic traits.