*3.4. QTL Pyramiding Line for LTG*

We compared the mean low-temperature germination rates of the F2 QTL-pyramiding lines (QTL-PL) with those of the parental lines (*O. rufipogon*, Hwaseong, CR1517, and CR1518) (Table S2). QTL-PL represents *O. rufipogon* alleles at four QTLs, *qLTG1*, *qLTG3*, *qLTG8*, and *qLTG10.1*, and Hwaseong allele at *qLTG10.2* locus. QTL-PL had higher germination rates than CR1517, CR1518, and Hwaseong, and lower germination rates than *O. rufipogon* at 6 and 7 DAI under the 13 ◦C condition. Although an 87.3% germination rate was observed in QTL-PL, this rate was significantly higher than that of CR1517 but not different from CR1518 at 6 DAI. The results indicate that pyramiding the QTLs with the combination of *O. rufipogon* alleles at *qLTG1*, *qLTG3*, and *qLTG8*, *qLTG10.1* and Hwaseong allele at *qLTG10.2* led to robust germinability under low-temperature conditions (13 ◦C). The lines could be useful material for developing rice varieties with enhanced LTG capacity (Table S2).

### *3.5. Verification of LTG QTL Using F3 Population*

Two F2 plants (CR8017-4 and CR8020-6) were selected based on their genotypes to verify *qLTG10.1* and *qLTG10.2*, respectively. CR8017-4 is heterozygous at *qLTG10.1* and *qLTG8*, while *O. rufipogon* homozygous for *qLTG3* and Hwaseong homozygous for *qLTG1* and *qLTG10.2*. CR8020-6 is heterozygous at *qLTG10.2*, while *O. rufipogon* homozygous for *qLTG1* and Hwaseong homozygous for *qLTG3* and *qLTG10.1*. The F3 plants were grown in the 2019/2020 winter season and dormancy was broken for seeds harvested from each F3 plant. We used low-temperature germination rates of two F3 populations at 6 and 7 DAI for the QTL analysis. QTL analysis indicated that RM25633 explained 21.79% of the phenotypic variance in the CR8017-4 population, confirming the presence of *qLTG10.1* (Figure 3A). However, we failed to detect the effect of *qLTG8* (*p* = 0.21). Additional experiments using larger populations are needed to verify the effect of *qLTG8*. QTL analysis indicated that *qLTG10.2* explained 20.75% of the phenotypic variance in the CR8020-6 population (Figure 3B). The Hwaseong allele at *qLTG10.2* contributed to an increase in LTG. The results indicate that two linked QTLs (*qLTG10.1* and *qLTG10.2*) on chromosome 10 are associated with LTG. Because *O. rufipogon* allele at *qLTG10.1* and Hwaseong at *qLTG10.2* contributed to higher LTG, the *O. rufipogon qLTG10.1* could be selectively introgressed into *japonica* rice using MAS. Two QTLs, *qLTG10.1* and *qLTG10.2*, explained 21.79% and 20.75% of the phenotypic variance, respectively, in the near-isogenic background. The low contribution of the QTL is partly due to the small population size and other environmental factors. The F3 plants were grown in the 2019/2020 winter season. We recorded the heading date of each plant and harvested them 30 days after heading. However, some late-flowering F3 plants failed to set adequate seeds to be tested for LTG, leading to the small population size and uneven distribution of three genotypic classes in the F3 population.

**Figure 3.** Frequency distribution of LTG at 7 days after incubation in the two F3 population. Plants in each population are segregating in (**A**) RM25633 for *qLTG10.1* and (**B**) RM333 for *qLTG10.2* in the different genetic background. The numbers in parenthesis indicate plants number followed by mean germination rate ± standard deviation of genotype group. *p*-value and R<sup>2</sup> were determined by one-way ANOVA. HH: Hwaseong homozygoutes, RR: *O. rufipogon* homozygotes, HR: heterozygotes.

### *3.6. Coleoptile Length in Parental Lines and F2 Population*

*O. rufipogon* showed significantly higher coleoptile length during 2–3 DAI at optimal condition than Hwaseong, CR1517, and CR1518, while no significant difference was observed in coleoptile length among Hwaseong, CR1517, and CR1518 at 30 ◦C (Figure S4A). At 1 DAI, mean coleoptile lengths were 0.6 mm and 0.3 mm in *O. rufipogon* and Hwaseong, respectively, while CR1517 and CR1518 were 0.4 mm in length on average. Coleoptile length varied from 12.4 mm to 14.6 mm at 4 DAI among parental lines (Figure S4A).

Coleoptile length varied markedly among parental lines under the 13 ◦C condition (Figure S4B). The mean length of the coleoptile in *O. rufipogon* was longer than those of Hwaseong, CR1517, and CR1518. The average coleoptile length in *O. rufipogon* at 5 DAI was 1.5 mm, while it reached 9 mm at 10 DAI. In Hwaseong, it was close to zero at 5 DAI and attained 4 mm at 10 DAI. While the average coleoptile length in CR1518 was slightly greater than that in Hwaseong at 5 to 9 DAI, the difference was much greater at 10 DAI. CR1517 had a longer coleoptile than CR1518 and Hwaseong, and a shorter coleoptile than *O. rufipogon*. The differences in coleoptile length between parental lines were more significant at 9 and 10 DAI than at other scoring dates (Figure S4, Table 3).


**Table 3.** Mean coleoptile lengths (mm) in *O. rufipogon*, Hwaseong, two introgression lines, and F2 population.

(x) DAI: days after incubation. Data are presented as mean germination rate or mean ± standard deviation. (y) Different letters in each row indicate significant difference at *p* < 0.05 based on Tukey's test. (z) *n*: number of plants.

In the F2 population, the coleoptile length varied from 0 to 12 mm from 5 to 10 DAI. The mean coleoptile lengths of 149 F2 plants were 0.4, 0.7, 1.2, and 3.2 mm at 5, 6, 7, and 8 DAI, respectively. At 9 DAI, the average coleoptile length was 6.6 mm, and it reached 7.9 mm at 10 DAI (Table 4, Figure 4). The F2 population exhibited nearly normally distributed coleoptile elongation at 9 and 10 DAI (Figure 4).


**Table 4.** List of QTLs identified for coleoptile length under low-temperature condition.

(x) CCL: Coleoptile length, (y) Coefficient of determination, (z) *a*: Additive <sup>e</sup>ffect=(*O. rufipogon* homozygote-Hwaseong homozygote)/2, *d*: Dominant effect = Heterozygote − (*O. rufipogon* homozygote + Hwaseong homozygote)/2, *d*/*a*:Degreeofdominance.

**Figure 4.** Frequency distribution of coleoptile length in 149 F2 plants at 13 ◦C at 9 and 10 DAI. Arrows indicate mean coleoptile length of CR1517 and CR1518 lines.
