Kazuma Okada

Efficient Genome Editing in Apple Using a CRISPR/Cas9 system.

Genome editing is a powerful technique for genome modification in molecular research and crop breeding, and has the great advantage of imparting novel desired traits to genetic resources. However, the genome editing of fruit tree plantlets remains to be established. In this study, we describe induction of a targeted gene mutation in the endogenous apple phytoene desaturase (PDS) gene using the CRISPR/Cas9 system. Four guide RNAs (gRNAs) were designed and stably transformed with Cas9 separately in apple. Clear and partial albino phenotypes were observed in 31.8% of regenerated plantlets for one gRNA, and bi-allelic mutations in apple PDS were confirmed by DNA sequencing. In addition, an 18-bp gRNA also induced a targeted mutation. These CRIPSR/Cas9 induced-mutations in the apple genome suggest activation of the NHEJ pathway, but with some involvement also of the HR pathway. Our results demonstrate that genome editing can be practically applied to modify the apple genome.

Identification of QTLs for fruit quality traits in Japanese apples: QTLs for early ripening are tightly related to preharvest fruit drop

Many important apple (Malus × domestica Borkh.) fruit quality traits are regulated by multiple genes, and more information about quantitative trait loci (QTLs) for these traits is required for marker-assisted selection. In this study, we constructed genetic linkage maps of the Japanese apple cultivars ‘Orin’ and ‘Akane’ using F1 seedlings derived from a cross between these cultivars. The ‘Orin’ map consisted of 251 loci covering 17 linkage groups (LGs; total length 1095.3 cM), and the ‘Akane’ map consisted of 291 loci covering 18 LGs (total length 1098.2 cM). We performed QTL analysis for 16 important traits, and found that four QTLs related to harvest time explained about 70% of genetic variation, and these will be useful for marker-assisted selection. The QTL for early harvest time in LG15 was located very close to the QTL for preharvest fruit drop. The QTL for skin color depth was located around the position of MYB1 in LG9, which suggested that alleles harbored by ‘Akane’ are regulating red color depth with different degrees of effect. We also analyzed soluble solids and sugar component contents, and found that a QTL for soluble solids content in LG16 could be explained by the amount of sorbitol and fructose.

Structural and functional analyses of two new S-RNase alleles, Ssi5 and Sad5, in apple.

Summary We determined the genomic DNA sequences of two new S-RNase alleles, Ssi5 and Sad5, in apple. Both consisted of 232 deduced amino acids, and contained five conserved domains (C1, C2, C3, RC4, and C5), and an intron specific to S-RNase genes of the Rosaceae, at the appropriate positions. We developed a PCR-RFLP method to identify these alleles, and clarified their occurrence in several Malus species, such as M. orientalis (W1-11-13; I.D. number 90104; Ssi5), M. sieversii (W1-10-49; ID number 90100; Ssi5), and M. sylvestris (ID number 88076; Sad5) and cultivars, such as ‘Adersleber Calville’ (Sad5) and ‘Cox’s Pomona’ (Sad5). Cross-pollination tests confirmed that Ssi5 and Sad5 were functionally distinct in spite of their identical rosaceous hypervariable region (RHV), which appears to be critical for S-allele discrimination in the Rosaceae. We recommend use of the next available numbers, S33 and S34, instead of Ssi5 and Sad5, respectively, since the allele annotations are only transitory. This is the first known case in apple in which two S-RNase alleles containing an identical RHV act as different alleles.

Characterisation of partial self-compatibility in the European pear cultivar, 'Grand Champion'.

Summary Most cultivars of European pear (Pyrus communis L.) exhibit S-RNase-based gametophytic self-incompatibility (SI), but the cultivar, ‘Grand Champion’, is partially self-compatible (SC). We used pollination and molecular genetic approaches to study the cause of the partial SC, and its effects on fruit set and quality. ‘Grand Champion’ was genotyped to SbSe by an S-RNase-based cleaved amplified polymorphic sequence marker system. Crossing ‘Grand Champion’ with pollen from an SI cultivar, ‘California’ (SbSe), showed that partial SC was caused by the disruption of pistil function. Of the Sb- and Se-RNase alleles cloned from ‘Grand Champion’, the Sb-RNase allele had two nonsynonymous nucleotide substitutions compared with the Sb-RNase allele previously cloned from ‘Doyenné du Comice’, but it retained the typical primary structure of S-RNases of the Maloideae. The sequence of the Sb-RNase allele from ‘Grand Champion’ was also obtained from two SI cultivars: ‘Joséphine de Malines’ and ‘Urbaniste’. Similar levels of Sb- and Se-RNase allele transcripts were found in the pistils of the partially SC and SI cultivars. Sb- and Se-haplotypes in the selfed progeny of ‘Grand Champion’ segregated in a 1:1 ratio. ‘Grand Champion’ fruits formed by self-pollination were the same size and quality as those formed by cross-pollination. Our results suggest that partial SC was not caused by a mutation in the S-RNase allele. The partial SC in ‘Grand Champion’ results in efficient fruit set and fruit of economic size and quality.

Identification and characterization of S-RNase genes in apple rootstock and the diversity of S-RNases in Malus species

We isolated and confirmed two S-RNases, denoted as mpS1 and mpS2, from apple rootstock ``Marubakaido`` (Malus prunifolia Borkh. Var. ringo Asami). These S-RNases contained and conserved five cysteine residues and two histidine residues, which are essential for RNase activity. The mpS1 showed high similarity to S5 (99.1%) of Malus spectabilis, whereas the mpS2 showed 99.5% nucleotide sequence similarity to S26 of (Malus × domestica) and 99.6% to S35 of (Malus sieversii) when compared with reported S-RNases. In amino acid sequences, the mpS1-RNase was almost similar to the S5-RNase of Malus spectabilis, and the mpS2-RNase was similar to the S35 of Malus sieversii, with only one bp being different from the S26-RNase of Malus × domestica. The 57 S-RNases of Malus species were renamed and rearranged containing the new S-RNases, as mprpS35 (mpS2) and mprpS57 (mpS1), for determining S-genotypes and identifying new alleles from apple species (Malus spp.).

Allelic composition of MdMYB1 drives red skin color intensity in apple (Malus × domestica Borkh.) and its application to breeding

Since apple fruit skin reddens poorly under warmer climates, new apple cultivars are desired that are adapted to global warming in terms of bearing well-reddened fruit. We developed a simple sequence repeat marker, Mdo.chr9.4, which is suitable for red skin color selection. It amplified four alleles (Mdo.chr9.4-R0, Mdo.chr9.4-Y−3, Mdo.chr9.4-Y−9, and Mdo.chr9.4-Y−15) distinguished by length. Mdo.chr9.4-R0 associated with MdMYB1-1 which confers red fruit skin. The presence of Mdo.chr9.4-R0 was consistent with empirical skin color in all 160 tested accessions. Mdo.chr9.4 was identified as the only significant marker that contributed to red skin color intensity by a genome wide association study (GWAS), and it accounted for 52.0% of phenotypic variation, confirming that MdMYB1 was the major and principal determinant of fruit skin color in apples. Individuals with a homozygous state of Mdo.chr9.4-R0 (dose 2) were significantly redder than those showing a heterozygote state (dose 1) in both the accession set and full-sib families, indicating a partially dominant effect of MdMYB1-1. Therefore, the selection of dose 2 individuals would target individuals with intensive red skin. We applied Mdo.chr9.4 to several application populations using a time and cost-efficient genotyping system developed in the present study. This system, along with Mdo.chr9.4, provide advanced marker-assisted breeding for intensive red skin color apples adapted to a global warming climate.

S-RNase genotypes of apple (Malus domestica Borkh.) including new cultivars, lineages, and triploid progenies

Summary We have determined the S-RNase genotypes of 33 new apple cultivars and lineages produced in Japan, 44 unknown cultivars and two lineages, and 22 triploid progenies. We have speculated on the putative parentage of new cultivars and lineages based on their S-RNase genotypes and also identified mistaken parents. In the case of the triploid progenies, the breeding of new cultivars using a triploid paternal parent may pose problems due to its low pollen viability. Nevertheless, diploid and triploid progenies were obtained using a triploid maternal parent. We have compiled a database of 516 apple S-RNase genotypes, including those previously investigated, which included a survey system for cultivar combinations, showing those that were fully-incompatible, semi-compatible, or fully-compatible, together with information on the PCR-RFLP method used for the identification of S-RNase genotypes and S-RNase allele designation (available at http://www.agr.nagoya-u.ac.jp/~0hort/apple/).

A root-localized gene in normal apples is ectopically expressed in aerial parts of columnar apples

The columnar apple ‘McIntosh Wijcik’, which is a mutation of ‘McIntosh’ shoots, has short internodes, thick stems, upright growth, poor lateral branches, and increased spur density. These columnar traits are controlled by a single dominant gene known as Co. We previously identified a putative dioxygenase gene (designated as 91071) as a promising Co candidate (expressed in the shoot apices of ‘McIntosh Wijcik’). However, tissue expression and function of the 91071 gene in noncolumnar apples is still not clear. In this study, we used reverse transcription polymerase chain reaction to demonstrate that the 91071 gene is mainly expressed in the roots of noncolumnar apples, whereas it is also expressed in shoot apices and leaves of columnar apples. In situ hybridization revealed that the 91071 gene is expressed at the primordium of lateral roots and root tips of both noncolumnar and columnar apple trees and in the shoot meristem and leaf primordium in the columnar apple ‘McIntosh Wijcik’. Grafting experiments of noncolumnar scion onto columnar rootstocks revealed that the columnar growth phenotype is not transmissible from rootstock to scion. These results indicated that ectopic expression of the 91071 gene in aerial parts causes columnar growth, whereas the expression of the 91071 gene in roots does not produce columnar growth. Furthermore, transgenic apples overexpressing the 91071 gene showed larger median adventitious root length and higher median number of lateral roots than control apples. Our result suggests that the 91071 gene may be related to root development.

Effects of temperature and solar radiation on sap flow in dwarfing apple rootstocks

Summary To elucidate the influence of temperature on water flow from soil to stem in a dwarfing rootstock, sap flow was measured under different temperature conditions in 1-year-old trees of apple rootstocks of different vigour. Sap flow was largely determined by solar radiation, and increased linearly as solar radiation increased when the temperatures above and below ground were held constant. The adjusted means of sap flow in the trees tested were not significantly influenced by the total leaf area, but were significantly influenced by the root mass. The adjusted means of sap flow at 30ºC were two- to three-times greater than those at 20ºC in all rootstocks tested, except one super-dwarfing rootstock, during the period of shoot extension. As the degree of rootstock vigour increased, the adjusted means of sap flow increased. The difference in sap flow between an invigorating rootstock and a dwarfing one increased under high temperature conditions. Because the degree of dwarfing was clearly expressed as the difference in sap flow under high temperature conditions, measurements of sap flow, as conducted in this study, will provide a useful tool for studying the mechanism of dwarfing.