**1. Introduction**

Rice is one of the most important crops, feeding more than a third of the global population. Global rice demand is projected to rise from 723 million tons in 2015 to 763 million tons by 2020 and is expected to increase further to 852 million tons by 2035 [1]. To feed the growing population, rice production needs to be improved and stabilized, globally and regionally. Improving potential yield and incorporation of biotic and abiotic stress tolerance mechanisms would facilitate the achievement of such goals. Wild relatives of rice are rich sources of desirable genes, not only with regard to yield but also with regard to disease resistance, stress tolerance, and other traits [2,3]. Exploring wild and exotic rice germplasm for desirable genes and transferring them into cultivars through crossing and

marker-assisted selection (MAS) has been demonstrated to be feasible approaches of increasing rice yield and augmenting stress tolerance [4].

Rice is cultivated in tropical, subtropical, and temperate regions of the globe. Low-temperature is one of the abiotic stress factors that lead to growth retardation and yield loss in rice cultivated at high latitudes and in the northern regions of the globe [5]. Throughout the rice cultivation period, decreasing temperatures to levels lower than the optimum could influence rice development significantly and can result in severe yield loss [6]. During the seed development period, various factors, such as environment and genetic background, influence low-temperature germination [7]. In rice cultivating regions, in areas where rice is irrigated with cold water, weak seedling growth and slow growth reduce yield significantly [8–13].

Along with rapid low-temperature germination (LTG), vigorous coleoptile growth is essential in direct seeding rice, when rice seeds are sown in flooded paddy fields and watered with cold irrigation water [14]. The rice coleoptile is a small and ephemeral organ that emerges first in imbibed rice seeds [15].

To determine the genetic basis of low-temperature germinability and coleoptile elongation in rice, numerous independent studies have been conducted using biparental populations and genome-wide association studies [6,8,9,12,16–21]. A total of 11 QTLs have been associated with low-temperature germination ability in the population derived from a cross between *indica* (N22) and *japonica* (USSR5) rice [12]. Three QTLs associated with low-temperature germination have been detected, and among them, *qLTG-9* was fine mapped to a 72.3-kbp region on chromosome 9 [22]. In addition, Lee et al. (2015) detected two colocalized QTLs for LTG and coleoptile length under a low-temperature condition (13 ◦C) [6]. Fukuda et al. (2015) also identified two QTLs associated with coleoptile length under 16 ◦C conditions [23]. Using the genome-wide association (GWAS) method, two QTLs associated with germination under cold tolerance, *qCTGERM8-1* and *qCTGERM8-2*, were identified in 421 accessions from a rice diversity panel (RDP1) [24]. In addition, Fujino et al. (2015) detected *qLTG8* on the RM5647 marker locus in 63 Japanese landraces [18]. Li et al. (2019) reported that Chinese Dongxiang wild rice (*O. rufipogon* Griff.) alleles at all five QTLs resulted in delayed germination rates, and QTL pyramiding line DX71 led to rapid germination and vigorous seedling survival under low-temperature conditions (15 ◦C and 8 ◦C) [25]. RNAseq analysis was conducted with two *indica* rice genotypes under cold treatment and detected 1361 differentially expressed transcripts [26].

Such studies indicate that low-temperature germination QTLs are distributed widely throughout the rice genome. However, the stability of such putative QTLs has not been tested in near-isogenic backgrounds, and studies analyzing interactions among QTLs are limited. Characterization of the QTLs underlying LTG and their interactions could provide more insights into the mechanisms of low-temperature germinability in rice and would be potentially more useful for pyramiding QTLs aimed at improving LTG in rice.

In our previous study, five QTLs for low-temperature germinability, *qLTG1*, *qLTG3*, *qLTG4*, *qLTG10*, and *qLTG11* were mapped using a BC4F8 population including 96 introgression lines (ILs) derived from an interspecific cross between a Korean elite line Hwaseong and *O. rufipogon* [27]. Recently, *qLTG1* was fine mapped using introgression lines and the new *qLTG3-1* allele of the *O. rufipogon* was identified [28,29]. In the present study, we used two introgression lines, CR1517 and CR1518, which consistently exhibit higher LTG than Hwaseong due to *O. rufipogon* introgression segments in the Hwaseong genetic background, as parents. Using 154 F2 plants derived from a cross between CR1517 and CR1518, we aimed to: (1) identify and characterize QTLs for low-temperature germinability and coleoptile length, (2) analyze interactions among LTG QTLs, and (3) develop QTL pyramiding lines with enhanced LTG and coleoptile length.
