Pyramiding of four QTLs for low-temperature germinability using MAS
The QTL pyramiding lines with the Akitakomachi genetic background were developed by crossing NILs carrying each of the qLTG3-1, qLTG3-2, and qLTG11-1 Maratteli alleles using MAS. Akikei770, carrying qLTG3-1, was crossed with Akikei793, carrying qLTG3-2 and qLTG11-1. The resulting F2 population of 576 individuals was subjected to MAS. Initially, 137 individuals homozygous for the Maratteli allele of qLTG3-1 were selected using the G1320 DNA marker. From these, 36 individuals homozygous for the Maratteli allele at RM7277 were selected, and subsequently, 28 individuals homozygous for the Maratteli allele qLTG11-1 were identified using RM206. The remaining eight individuals were recombinants between RM7277 and RM206. Finally, eight individuals homozygous for the qLTG3-2 Maratteli allele, flanked by RM6329 and RM3436, were selected from the 28 individuals. There were four recombinants between these markers. The following year, eight lines carrying the Maratteli alleles of qLTG3-1, qLTG11-1, and qLTG3-2, along with the qLTG1-1 Akitakomachi allele, were cultivated in the field. From these, the QTL pyramiding lines Kei2161, Kei2162, Kei2171, and Kei2172 were selected. Finally, Akikei812, harboring all four QTLs for low-temperature germinability, was established from Kei2161 (Fig. 1, Table 1).
The segregation ratios for each of qLTG3-1, qLTG11-1, and qLTG3-2 fit the theoretical 1:2:1 ratio (Table 1). Eight plants carrying three QTLs were observed among the 576 F2 individuals, which fits the 1:63 theoretical segregation ratio for three independent genes, possibly due to the distant location of qLTG3-1 and qLTG3-2 on chromosome 3 (Table 1, Fig. 1). This result suggests that no segregation distortion occurred during the pyramiding of QTLs for low-temperature germinability. While stepwise DNA marker selection reduced the number of analyses, increasing the number of pyramided QTLs or segregation distortion may considerably increase the number of plants analyzed, thus potentially limiting the QTL accumulation via MAS.
Improvement of low-temperature germinability of Akitakomachi through QTL pyramiding
Figure 2 shows the germination rate on day five after sowing at 15°C for the QTL pyramiding lines with the Akitakomachi genetic background. The germination rates of Akitakomachi and Maratteli were 3% and 81%, respectively. The germination rates of the two QTL pyramiding lines, Akikei770 (qLTG1-1 + qLTG3-1) and Akikei775 (qLTG1-1 + qLTG11-1) were 8% and 13%, respectively, which was significantly higher than that of Akitakomachi at the 1% level, as determined by the Student’s t-test. Conversely, the germination rate of Akikei771 (qLTG1-1 + qLTG3-2) was significantly lower than that of Akitakomachi at the 5% level, as determined by the Student’s t-test. These results indicate that qLTG3-1 and qLTG11-1 from Maratteli increased the germination rate of Akitakomachi at low temperatures, while qLTG3-2 did not.
The three QTL pyramiding lines, Akikei791 (qLTG1-1 + qLTG3-1 + qLTG3-2), Akikei792 (qLTG1-1 + qLTG3-1 + qLTG11-1), and Akikei793 (qLTG1-1 + qLTG3-2 + qLTG11-1) had germination rates of 47%, 40%, and 41%, respectively, at 15°C on day five after sowing, which was significantly higher than those of the two QTL pyramiding lines, Akikei770 and Akikei775. In addition, the germination rates of the four QTL pyramiding lines, Akikei812 and Kei2171 (qLTG1-1 + qLTG3-1 + qLTG3-2 + qLTG11-1), were 73% and 72%, respectively, which was significantly higher than those of the three QTL pyramiding lines. Another four QTL pyramiding line, Kei2161, exhibited a slightly higher germination rate than the three QTL pyramiding lines (Fig. 2).
These results demonstrated that the germination rate at low temperatures increases proportionally with the number of accumulated QTLs. Consequently, the low-temperature germination rate of Akitakomachi was remarkably improved by pyramiding QTLs for low-temperature germinability, which approached the level of Maratteli (Fig. 3).
Germination profile at low temperature of the QTL pyramiding lines
Figure 4 shows the time course of the germination rate at 15°C for the QTL pyramiding lines. Maratteli began germinating 4 days after sowing, with a rapid increase in germination rate, reaching 100% by day seven. Akitakomachi began germinating five days after sowing, completing germination by day 10. The germination rates of the three QTL pyramiding lines, Akikei791, Akikei792, and Akikei793, increased faster than Akitakomachi but later than Maratteli. The four QTL pyramiding lines, Akikei812 and Kei2171, exhibited a similar germination profile to that of Maratteli. The germination profile of another four QTL pyramiding line, Kei2161, was similar to that of the three QTL pyramiding lines. These results demonstrate that the low-temperature germination profile of Akitakomachi was enhanced to be closer to that of Maratteli by pyramiding four QTLs for low-temperature germinability.
Low-temperature seedling establishment of the QTL pyramiding lines
There were no significant differences in plant height, root length, and leaf age at low temperatures between Akitakomachi and the pyramiding lines with two or three QTLs for low-temperature germinability. However, in Akikei2161 and Akikei2162, which harbor four QTLs, these three traits were significantly increased compared to Akitakomachi. Similarly, these traits in Akikei2172 were also higher than those in Akitakomachi. In contrast, there was no difference in these traits between another four QTL pyramiding line, Akikei2171, and Akitakomachi (Fig. 5a-c). Since Akikei2171 and Akikei2172 are sibling lines with the same genotypes, these differences might be caused by differences in the physiological status of the harvested seeds.
The seedling establishment at low temperatures, which was evaluated by the percentage of individuals with a leaf age of 2.0 or more, was 8% for Akitakomachi. In contrast, this was 15.4% and 25.5% for Akikei770 and Akikei790, respectively, which accumulated two QTLs. The seedling establishment rates for the three QTL pyramiding lines ranged from 30.4–33.6%, which were higher than that of Akitakomachi. Furthermore, the seedling establishment rates at low temperatures for the four QTL pyramiding lines, Kei2161, Kei2162, Kei2171, and Kei2172, ranged from 38.6–54.5%, which were significantly higher than that of Akitakomachi (Fig. 5d).
The European varieties Italica Livorno and Arroz da Terra exhibited the highest plant height but did not significantly differ from Akitakomachi regarding the other traits. Plant height is a useful trait for evaluating low-temperature seedling establishment according to a previously used evaluation method (Matsumoto et al. 2001). The two European varieties were confirmed to be superior for low-temperature seedling establishment.
Conversely, Maratteli showed a tendency to be lower than Akitakomachi for all traits (Fig. 5). Maratteli has been reported to vary in plant height from 1999 to 2001 using the same evaluation system, which resulted in similar plant height to Akitakomachi (Matsumoto et al. 2001). Maratteli may be strongly influenced by the climatic conditions of the harvest year, possibly having characteristics that make seedling establishment unstable at low temperatures.
There was a high positive correlation (R² = 0.940) between the germination rate and seedling establishment at low temperatures among the QTL pyramiding lines with the Akitakomachi genetic background, which indicates that seedling establishment at low temperatures increased proportionally with the number of QTLs for low-temperature germinability (Fig. 6). The QTL pyramiding lines exhibited accelerated initiation and progression of germination at low temperatures (Fig. S2). This acceleration in germination was followed by faster seedling growth, resulting in higher seedling establishment. Therefore, accumulating QTLs that enhance low-temperature germinability is likely to improve low-temperature seedling establishment in elite varieties.