A large-scale analysis of seed natural aging revealed stronger aging resistance in weedy rice than coexisting cultivated rice for continuous 4 years across China

Increasing resistance to aging is conducive to seed storage and germination rate of crop. Meanwhile, the resistance to aging is one of the important adaptive mechanisms of weed to thrive in farmland. Weedy rice (Oryza sativa f. spontanea) and cultivated rice (Oryza sativa L.) provide a unique pair demonstrating a weed and conspecic model crop that can be used to study the aging resistance of plants across a diverse geographical range. Chinese weedy rice derived from de-domestication of cultivated rice has rapidly risen to malignant weeds, though the hazard has only been reported for about 20 years. Whether weedy rice rapidly evolves higher seed aging resistance than cultivated rice during the process of dedomestication, which is conducive to its persistence in rice elds, is still unclear. In this experiment, the seeds of weed rice populations and their co-existing rice varieties were collected from 61 regions of China and germinated under normal and high temperatures for consecutive four years (2013–2016). Our study found that the aging resistance of weedy rice was higher than that of the co-existing rice cultivars, and weedy rice may have evolved a different aging resistance mechanism than rice cultivars and could be used as a germplasm resource to cultivate aging-resistant rice. The indica-type has strong aging resistance and no dormancy, while the japonica type has weak aging resistance and a little weak dormancy. Thus, by introducing indica-type aging-resistant alleles into japonica rice, cultivation of aging-resistant japonica rice could be possible.


Introduction
Seed aging not only decreases the storage ability of seeds but also decreases seed vigor and affect the germination in the eld ( Under normal storage conditions, the quality of rice seeds begins to deteriorate after one year and strongly affects its eating quality. Due to poor storage potentiality, a large quantity of rice seed is often discarded every year, which leads to huge loss (Li et al., 2017). According to the annual yield of 200 billion kg in recent years, the loss of food is nearly 6 billion kg (Xu et al., 2015). In addition, decreased seed vigor of rice during the storage also causes huge economic losses. The construction of cold storage facilities might be an option but involves high cost. Conversely, cultivating age-resistant rice varieties is a convenient and feasible alternative (Sasaki et al., 2005; Zhang et al., 2019).
Aging resistance of seed is considered to be an important adaptive mechanism for the persistence of weeds in the eld (Baker, 1974). Weedy rice in paddy elds has become one of the most malignant weeds all over the world (Delouche et al., 2007). According to the genome-wide evolutionary analysis, Chinese weedy rice was more closely related to local cultivated rice and considered to be de-domesticated from cultivated rice (Qiu et al., 2017). In the process of de-domestication, weedy rice might have evolved some new germination mechanisms to adapt the local environment. Xia et al., (2011) found that the critical low temperature of seed germination was related to the local habitat temperature in weedy rice populations. Burgos et al. (2006) found that the nitrogen utilization rate was more e cient in weedy rice (Strawhull) of Southern United States than the high-yield rice cultivar (Drew). Puteh et al. (2013) reported that weedy rice cultivated under less water had higher drought tolerance capacity in comparison to the indica-type rice cultivars (MR219 and MR 232) as well as weedy rice accessions (Bertam and Ketara). Borjas et al. (2016) found that weedy rice and the rice cultivar got a higher rate of germination and seedling vigor at low temperature than the American weedy rice (PSRR-1) along with rice cultivar (Bengal).
The aging resistance of seeds is usually studied using the natural aging and arti cial aging (Hay et al., 2019). Relative to natural aging, arti cial aging is to accelerate seed aging using high temperature and humidity (Zeng et al., 2002;Zeng et al., 2006;Xu et al., 2013;Xu et al., 2015). By comparing the results of arti cial aging with natural aging, some researchers have found that arti cial aging does not fully re ect all the processes that occur during true natural aging (Zhang and Hu, 2015;Xu et al., 2013). The vitality of natural aging seeds can persist to affect the whole growth process and ultimately affect the rice yield. On the contrary, the vigor of arti cial aging seeds only affects the early production performance of the rice plant and ultimately has little effect on the production performance of later stage (Zhang and Hu, 2015). Xu et al. (2013) found that the results were not completely consistent through comparing seed germination rate of rice varieties through quantitative trait loci (QTL) analysis of indica-japonica recombination inbred line population under natural aging (2 years) and arti cial aging (high temperature and high humidity), respectively. Hang et al. (2015) found that qSSh-2-1 and qSSh-2-2 could be detected under natural and arti cial aging, while the other 4 QTLs (qSSh-4, qSSs-5-1, qSSs-5-2 and qSSh-12) and 7 QTLs (qSSh-1, qSSh-3-1, qSSh-3-2, qSSh-3-3, qSSh-7-1, qSSh-7-2 and qSSh-11) were only detected under natural aging and arti cial aging, respectively.
In this experiment, 61 weedy rice populations (each population included 8 to 33 accessions) and coexisting rice cultivars across China were collected and germinated under normal and high-temperature dormancy-breaking treatment for consecutive four years (2013-2016). The research aimed to solve the following issues: 1) whether the weedy rice populations have stronger aging resistance than coexisting rice cultivars; 2) to reveal the relationship between indica-japonica type and aging resistance; 3) to explore the relationship between seed dormancy and aging resistance.  Table S1. Mean Annual precipitation, Mean Annual temperature, Mean temperature in the sowing month, and Mean precipitation in the sowing month were collected from http://www.worldclim.org (Table S1).
Each population included 8 to 33 accessions sampled from the same eld and each population was separated from others by a distance of more than 10 km. When weedy rice was collected at each sample site, co-existing rice cultivar was also collected from the eld which was more than one hectare. These rice varieties are recommended to cultivate by local government in China. Most of farmers buy rice seeds every year, especially for those farmers who own more than one hectare of land in China. In order to eliminate maternal, environmental, varying seed age, and storage effects, all weedy rice populations and coexisting rice cultivars were planted at Jiangpu Experimental Station, Nanjing Agricultural University, Nanjing, Jiangsu, China, in 2013. At 30 days after sowing, 50 seedlings from each population were transplanted in a random block design with three repetitions. The experimental eld was managed with conventional rice cultivation methods. Mature seeds from each population were harvested 50 days later after the heading date. The harvested seeds dried under well-ventilated conditions to a constant weight (about 12 % seed moisture content) and were stored at room temperature (< 30 o C) and 40-60 % relative humidity in Nanjing from 2013 to 2016. At November each year seeds were germinated at 28°C constant

Normal and high temperature treatments
A group of 180 plump rice seeds representing each sampled population were divided into six samples and placed on two layers of lter paper in Petri dishes. Three samples under normal temperature treatment (25 o C, 48 h) were germinated with 6 mL distilled water while three samples under high temperature treatment (50 o C, 48 h) to break dormancy were germinated with 6 mL distilled water. Then, the dishes were wrapped in plastic bags and placed into incubators (SPX-400-GBH) at a steady temperature of 28°C with a light: dark cycle of 16:8 h. The number of germinated seeds was recorded each day at 7th days. Germination was evaluated visually by protrusion of the radicle from the hull by ≥ 3 mm (Baskin & Baskin, 1998). GR = (number of seeds germinated by Day 7 / total seed number) × 100%. characters, that is, hull color at heading stage, epidermal hair on leave at heading stage, epidermal hair on hull at heading stage, length between 1 st -2 st panicle node at mature stage, phenol reaction of seed at mature stage, and length-width ratio of seed at mature stage.

Data analysis
Analysis of Variance (ANOVA) was performed using SPSS 18.0 (IBM, New York, New York, USA). The geographic information map was drawn using ArcGIS 10.0 (ESRI, Redlands, California, USA). Other gures were drawn using Excel 2010 (Microsoft Corporation, Redmond, Washington state,) and Origin 8.0 (Originlab, Northampton, Massachusetts, USA).

Comparing seed aging resistance between weedy rice populations and coexisting rice cultivars
One-way ANOVA showed that there was no signi cant difference in the germination rate between weedy and cultivated rice in 2013 and 2014 under normal temperature treatment condition, while the germination rate of weedy rice was signi cantly higher than that of cultivated rice in 2015-2016 ( Fig.1; Table 1).
Under high-temperature dormancy-breaking treatment condition, the germination rate of weedy and cultivated rice was not signi cantly different in 2013, while the germination rate of weedy rice was signi cantly higher than that of cultivated rice in 2014-2016 ( Fig. 1; Table 1). The germination rate of weedy and cultivated rice under high-temperature dormancy-breaking treatment was signi cantly higher than that of normal temperature treatment in 2013, which showed that seeds harvested in the rst year (2013) had a weak dormancy, so breaking dormancy could increase its germination rate. In 2014, the germination rate of weedy and cultivated rice under high-temperature dormancy-breaking treatment was signi cantly lower, which indicated that high temperature accelerates seed aging. Two-way ANOVA between temperature treatments (normal temperature and high temperature treatment) and genotypes (weedy and cultivated rice) showed that there were signi cant differences in genotypes in 2013-2016 except for 2013 (Table 2). Furthermore, with the increase of aging, the germination rate of weedy rice was signi cantly higher than that of cultivated rice (Fig 1; Table 1).

Relationship between indica-japonica type and aging resistance of weedy rice and coexisting rice varieties
The germination rate along with the latitude gradient was plotted under normal temperature and hightemperature treatment, respectively ( Fig.2 A, B). The germination rate of weedy and cultivated rice decreased with latitude, which suggests that the indica-japonica type may affect the aging resistance.
One-way ANOVA of germination rates showed that under normal temperature treatment, germination rates of indica-type weedy and cultivated rice were signi cantly higher than japonica-type weedy and cultivated rice in 2013-2016 (Fig. 4). In the third year (2015), the germination rate of japonica-type weedy and cultivated rice was decreased to about 40% and 10%, respectively, while the germination rate of indica-type weedy and cultivated rice remained at about 90%. Correspondingly, under the high-temperature dormancy-breaking treatment, there was no signi cant difference in the germination rate of indica-japonica type weedy and cultivated rice in the rst year (2013). In 2014-2016, the germination rate of indica type was signi cantly higher than that of japonica type for both weedy rice and cultivated rice, and the germination rate of japonica-type weedy rice was signi cantly higher than that of japonica-type cultivated rice (Fig. 3). During the comparison of normal temperature treatment with high-temperature treatment, the weedy and cultivated rice showed signi cantly increased germination rate in 2013, but the high-temperature dormancy-breaking treatment signi cantly reduced the germination rate in 2014-2016 ( Fig. 4). Besides, two-way ANOVA between subspecies (indica and japonica) and genotypes (weedy and cultivated rice) under normal temperature treatment revealed that there was a signi cant difference in the subspecies but no signi cant difference was there in genotypes in 2013-2014 (  (Table 3).
In summary, the aging resistance of indica-type weedy and cultivated rice was both signi cantly higher than that of japonica-type weedy and cultivated rice (Fig. 3). In addition, indica-type weedy and cultivated rice showed no dormancy, while japonica-type weedy and cultivated rice showed weak dormancy (Fig. 3).

Discussions
3.1 The aging tolerance of weedy rice is higher than that of cultivated rice and the aging resistance of indica type is higher than that of japonica type Our results indicated that the germination rate of weedy rice was signi cantly higher than that of coexisting rice varieties during the natural aging (Table 1,  During the comparative analysis of seed germination rate with latitude in 2013-2016 (Fig. 2), it was noticed that the germination rate of weedy rice and cultivated rice was decreased with the increase in f latitude. Indica-type rice varieties are normally adapted to warm, low-latitude or high-altitude tropical environments, while japonica-type rice varieties are adapted to temperate environments with relatively high latitudes or high altitudes with relatively cool climatic conditions (Glaszmann, 1987). By analyzing the whole genome sequencing of 155 weedy rice accessions and 76 locally rice varieties in four representative regions of China, Qiu et al. (2017) found that the genetic distance between weedy rice and local varieties was lesser than other groups. Therefore, the variation in germination rate may be related to the distribution of indica-japonica type in weedy and cultivated rice. Additionally, by analyzing the relationship between indica-japonica type and aging resistance of weedy and cultivated rice (Fig. 2, 3), it was noticed that the germination of indica-type weedy and cultivated rice was signi cantly higher than that of japonica-type weedy and cultivated rice in  Roberts (1963) found no signi cant relationship between seed dormancy and aging while studying the seed germination rate of six rice cultivars. Siddique et al. (1988) found that the higher-dormancy varieties also had higher aging resistance while analyzing 30 rice varieties under arti cial aging treatment. In order to distinguish seed dormancy and aging resistance, normal treatment, as well as high temperature dormancy-breaking treatment, was set in this experiment. In 2013, the germination rate under high-temperature dormancy-breaking treatment was signi cantly higher than that of normal temperature, indicating that the seeds harvested in the rst year had weak dormancy. Further analysis of the indica-japonica type of weedy and cultivated rice showed that the japonica type had weak dormancy, while the indica type had no signi cant dormancy (Table 3). This is consistent with previous report of Xia et al. (2011) that Chinese weedy rice has weak dormancy or no primary dormancy. In 2014-2016, the germination rate of japonica type under high-temperature dormancy-breaking treatment was signi cantly lower than that of normal temperature treatment, indicating that high temperature became new stress, which accelerated aging. The outcome of this study clearly shows that the aging resistance, but not dormancy, in the weedy and cultivated rice is related to the indica-japonica type. The aging resistance of the indica type was high but with no dormancy, while the aging resistance of the japonica type was low with comparatively higher dormancy. Similarly, Nguyen et al (2012) found that the high storage ability was related to weak seed dormancy, and the low storage ability is associated with high seed dormancy in Arabidopsis.

Conclusion
The present study analyzed the germination rate of 61 weedy rice populations and coexisting rice varieties in consecutive four years (2013-2016), and found that the aging resistance of weedy rice was stronger than that of cultivated rice. Thus, weedy rice might be used as germplasm resources to cultivate aging resistant rice cultivars. The aging resistance in the weedy and cultivated rice is closely related to indica-japonica type but not to the dormancy. The introduction of aging-resistant alleles from indica type into japonica rice may be utilized to breed aging-resistant japonica rice.

Declarations
Ethics approval and consent to participate Not applicable Consent for publication: Not applicable.

Availability of data and material
Not applicable.

Competing interests
The authors declare that they have no competing interests.       One-way ANOVA of germination rate of indica / japonica type weedy rice and coexisting rice cultivars under normal and high temperature treatments for consecutive 4 years in 61 sites across China Note: Blue and red histograms represent indica / japonica type weedy rice, respectively. Blue diagonal and red diagonal histograms represent indica / japonica type rice cultivars. Green dash-line border represents normal temperature treatment. Black solid-line border represents high temperature treatment. Vertical bars indicate ± standard error of mean. Signi cant levels are indicated by letters (a, b, c), respectively.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. 0Supplementaltable1Theinformationof61samplesites.xlsx