Abstract: Seed dormancy is a naturally important biological process which can affect planting, germinating and harvesting in agricultural production. Variability in seed dormancy within the U.S. peanut mini-core collection had not been determined. Freshly harvested seeds in the same field from 103 accessions were tested for germination with two treatments (H2O only or 10 mM ethephon). The number of seeds that germinated or died was recorded. Significant variability in seed dormancy was observed among accessions and botanical varieties. In comparison with the H2O treatment, ethephon significantly promoted dormancy release, but the level of dormancy release was genotype dependent. The interaction effect (genotype x treatment) on seed dormancy release was also determined. Seed dormancy among four botanical varieties was compared. The botanical variety of hypogaea was more dormant than the other three botanical varieties evaluated. However, significant variability was also identified within botanical varieties. The accessions identified with greater dormancy within the same botanical variety would be useful in breeding programs to prevent preharvest sprouting. There are six botanical varieties in cultivated peanuts, but only four botanical varieties were included in the U.S. peanut min-core collection. The variability in seed dormancy for the other two botanical varieties needs to be investigated in future studies.
INTRODUCTION
Seed dormancy is defined as an intact viable seed delaying or preventing the completion of germination under favorable conditions (temperature/light/moisture). It can be classified into five categories: Morphological Dormancy (MD), Physiological Dormancy (PD), Morphophysiological Dormancy (MPD), Physical Dormancy (PY) and combinational dormancy (PY+PD). The length of seed dormancy varies from a few days to several years (Bewley, 1997; Baskin and Baskin, 2004). There are no universal mechanisms to trigger dormancy release which primarily depends on the dormancy type. Even seeds from the same species could have more than one type of dormancy (Nikolaeva, 1977). Seed dormancy is a naturally important biological process which can affect planting, germinating and harvesting in agricultural production. Some seed dormancy can be induced or overcome by environmental stimuli or plant hormones such as abscisic acid (ABA, germination inhibitor), ethylene or gibberellins (GA, germination activator) (Ketering and Morgan, 1971; Bewley, 1997; Hu et al., 2010).
Fresh seed germination can be directly affected by seed dormancy. Seed germination can be manipulated by the regulation of seed dormancy. For example, in order to speed up homozygosity of progenies from early to late generations by selfing, the dormancy from freshly harvested seeds needs to be released for early planting. Ethephon (2-chloroethylphosphoric acid) can be converted into ethylene by plants. It has been successfully used as a reagent for dormancy release in many crops including some legume species (such as Medicago, clover and peanut) (Toole et al., 1964; Ketering and Morgan, 1971; Globerson, 1978). On the contrary, in order to prevent preharvest sprouting, dormancy is required for freshly matured seeds (such as wheat grains on spikes aboveground and peanut seeds in pods underground). Preharvest sprouting can lead to great yield loss and poor seed quality for many crops. Peanut (Arachis hypogaea L.) is a warm-season legume. Either earlier or later harvesting of peanut seeds can affect the length of dormancy (Hull, 1937). If continuous rainfall occurs at harvest, delayed harvesting could lead to a 10-20% yield loss due to sprouting in the field (Gautreau, 1984), especially for Spanish peanut (Khalfaoui, 1991; Nautiyal et al., 2001). From limited data, monogene and multiple genes, additive, dominance and digenic epistatic effects for seed dormancy have been reported in peanut, depending on the types of parents used in crosses (John et al., 1948; Lin and Lin, 1971; Khalfaoui, 1991; Upadhyaya and Nigam, 1999; Issa et al., 2010). Determination of seed dormancy levels for some breeding materials (such as germplasm accessions, parents, progenies and cultivars) would be useful for breeding programs and agricultural production.
A peanut mini-core was established from the U.S. peanut germplasm collection (Holbrook and Dong, 2005) which encompassed divergent materials from four botanical varieties (fastigiata, peruviana, vulgaris and hypogaea) with a relatively high level of genetic diversity (Barkley et al., 2007; Kottapalli et al., 2007; Wang et al., 2011). Some accessions in the mini-core have been frequently requested by peanut breeders and geneticists and used in breeding programs and genetic studies. Seed dormancy is an important trait for biological studies and agricultural production. However, seed dormancy for accessions within the U.S. peanut mini-core has not been evaluated. In order to determine the variability for seed dormancy, freshly harvested seeds from the U.S. peanut mini-core were tested for germination using two treatments (either H2O only or 10 mM ethephon). The objectives of this study were to: (1) determine seed dormancy variability within the U.S. peanut mini-core collection; (2) determine ethephon treatment effect on seed dormancy release in peanut; (3) determine whether there is any difference in seed dormancy among botanical varieties and (4) determine whether there is any interaction between genotype x ethephon treatment on seed dormancy release.
MATERIALS AND METHODS
Planting and collection of seeds from the field: Seeds from 103 of the 112 accessions within the U.S. peanut mini-core (Table S1) were obtained from the USDA-ARS, Plant Genetic Resources Conservation Unit (PGRCU), Griffin, GA. Using a randomized complete block design, 20 seeds from each accession were planted in two-row, 10-feet long plots in Dawson, GA, 2010. Off-type plants were removed from the field when morphological variation was observed within an accession. In order to confirm the type of botanical varieties, five morphological traits including growth habit, main stem length, presence of flowers on the main stem, leaf color and stem pigmentation were recorded in the field at 10 and 14 weeks after planting, following the procedures used for standard peanut descriptors (http://www.ars-grin.gov/npgs). Peanut plants were harvested at physiological maturity by a small peanut harvesting combine.
| Table S1: | Peanut accessions within the U.S. mini-core tested for seed dormancy |
| W-dorm: Water treated dormancy, W-death: Water treated death, E-dorm: Ethephon treated dormancy, E-death: Ethephon treated death, Seed weight: Gram for 100 seeds | |
After drying, pods were photographed and scored for shape, constriction, reticulation and number of seeds per pod (http://www.ars-grin.gov/npgs). Based on plant observations and pod descriptors, the botanical variety classification in the GRIN database was confirmed and recorded when classification data were missing (Table S1). After shelling, harvested seeds were selected (based on size uniformity and maturity) and then used for germination tests within one week.
Testing seed germination under different treatments: Prior to germination testing, seed weight and seed-coat colors were recorded. Pre-selected seeds were tested for germination under two treatments: water only and 10 mM ethephon (2-chloroethylphosphonic acid) with two replicates. Twenty-five seeds were used for each replicate per accession. To prevent mold during germination, all the seeds were treated with Trilex Star® fungicide (Bayer CropScience). Two layers of standard germination paper were soaked in H2O or 10 mM ethephon. The peanut seeds were mixed well with fungicide powder. The soaked germination paper were spread out on a table and 25 seeds were laid onto it. Another layer of paper was soaked in H2O or ethephon and used to cover the seeds. The soaked paper covering the seeds was rolled up and banded with another replicate using a rubber band. The seeds in rolled-up paper towels were transferred into a germinator with 85% humidity set for 20°C for 12 h of darkness and 30°C for 12 h with 8 h of light. Two growth chambers were used: one for water-treated seeds and the other for ethephon-treated seeds. During the course of germination, tap water at room-temperature was added to the germination paper surrounding the seeds to keep the paper towels moist. Germination rates were recorded at 7, 14 and 21 days. Dead seeds (no any sign for germination but rotten during the process) were not included in the calculation of the germination rate.
Data analysis: Data were collected from the two replicates and two treatments (H2O and ethephon). An analysis of variance was performed and means were separated using Tukey’s multiple comparison procedure. Pearson’s correlation coefficient analysis was also performed to determine whether seed weight affects the rate of germination or dormancy.
RESULTS
The results from both types of statistical analysis (Type I SS and Type III SS) on seed germination, dormancy and death were very similar and only the results from Type I SS were listed in Table S2.
| Table S2: | Significance of variation source at 7, 14, and 21 days |
| Rep: Replicate, Pi: Plant introduction (i.e. accessions),
and “treatment*pi” for interaction between treatment and plant introduction |
|
| Table 1: | Ethephon treatment effect on seed germination, dormancy and death |
| H2O treatment results are in brackets, N: No. of samples, SD: Standard deviation, Means with different letters are significantly different | |
Overall, there were no significantly statistical differences between replicates (F = 0.01, p<0.9309 at 7 days; F = 0.40, p<0.5255 at 14 days and F = 1.05, p<0.3059 at 21 days) but there were significantly statistical differences between treatments (F = 5725.57, p<0.0001 at 7 days; F = 10374.2, p<0.0001 at 14 days and F = 7547.71, p<0.0001 at 21 days) and among accessions (F = 62.54, p<0.0001 at 7 days; F = 100.90, p<0.0001 at 14 days; F = 73.05, p<0.0001 at 21 days). Significant interaction between accessions and treatments were also identified (F = 47.57, p<0.0001 at 7 days; F = 87.71, p<0.0001 at 14 days and F = 62.90, p<0.0001 at 21 days). The order of factor’s impact on variation was treatment>accession>interaction between treatment and accession. The results for detecting variation on seed germination, dormancy and death from 7, 14 and 21 days were consistent. Seed weight did not correlate with seed germination, dormancy and death (data not shown).
Treatment effect on seed germination and dormancy: There were only two treatments (H2O only and 10 mM ethephon) applied in this study. The means and range of seed germination, dormancy and death rate under H2O or ethephon treatment are listed in Table 1 and displayed in Fig. 1. The average seed germination rate 97.8% (ranging from 24-100%) under ethephon treatment was significantly higher than 56.7% (ranging from 0-100%) under H2O treatment. The average dormancy rate under ethephon treatment was 1.0% (ranging from 0-68%), significantly lower than 41.6% (ranging from 0-100%) under H2O treatment. Clearly, ethephon can help overcome dormancy but not for all accessions which were tested.
| Fig. 1: | Ethephon treatment effect on seed germination, dormancy and death. If the letters above the bars for seed germination, dormancy and death are different, there is a significant difference |
| Fig. 2: | Number of accessions in each dormancy category. The numbers above the bars are the number of accessions for each dormancy category |
Interestingly, the seed death rate from the ethephon treatment was 1.2%, statistically higher than the one from H2O treatment (0.8%), but the value difference in these two death rates was little.
Genotype effect of accessions on seed germination and dormancy: Among 103 accessions, the level of seed dormancy with H2O treatment was classified into four categories: no dormancy (100% germination at 7 days), weak dormancy (over 80% germination at 14 days), strong dormancy (germinated accessions at 21 days) and complete dormancy (0% germination at 21 days). The number of accessions in each category was displayed in Fig. 2. There were two accessions (PI 475918 and PI 493356) with no dormancy showing 100% germination on day 7. Both accessions belonged to the botanical variety fastigiata. Forty-two accessions displayed weak dormancy and 54 accessions displayed strong dormancy. There were five accessions (PI 259658, PI 259851, PI 292950, PI 323268 and PI 355271) showing complete dormancy (no germination on day 21). These five accessions belonged to the botanical variety hypogaea and are good genetic materials to use for preventing preharvest sprouting in peanut breeding programs.
Botanical variety effect on seed germination and dormancy: Significant botanical variety effect on seed germination (F = 16.12, p<0.0001) and dormancy (F = 14.81, p<0.0001) was detected. Only the results from day 21 are listed (Table S3). The average levels of seed germination and dormancy among botanical varieties at day 21 are listed in Table 2 and displayed in Fig. 3.
| Table S3: | Significance of botanical variation effect on seed germination and dormancy |
| Rep: Replicate, Bt: Botanical variety | |
| Table 2: | The level of seed germination and dormancy at day 21 among different botanical varieties |
| Means with the same letter are not significantly different | |
| Fig. 3: | The levels of seed germination and dormancy among botanical varieties. “fp”, “fv”, “ff” and “hh” is for botanical variety fastigiata peruviana, fastigiata vulgaris, fastigiata fastigiata and hypogaea hypogaea, respectively. If the letters are the same above the bars for germination rate or dormancy rate, there is no significant difference between means |
| Fig. 4: | Interaction effect between genotype and ethephon treatment on dormancy release. The scale bar represents 1 cm |
There was no significant difference in the average seed germination (95.3, 89.2 and 84.7%) among three botanical varieties [subsp. fastigiata var. peruviana (fp), subsp. fastigiata var. vulgaris (fv) and subsp. fastigiata var. fastigiata (ff)]. However, the average seed germination (69.2%) for the subsp. hypogaea var. hypogaea (hh) was significantly lower than those of the three other botanical varieties. There was no significant difference in the average level of seed dormancy (29.0 and 14.3%) between subsp. hypogaea var. hypogaea (hh) and subsp. fastigiata var. fastigiata (ff). However, the average level of seed dormancy for subsp. hypogaea var. hypogaea (hh) was significantly higher than the levels (10.3 and 1.7%) for subsp. fastigiata var. vulgaris (fv) and subsp. fastigiata var. peruviana (fp). Therefore, there is a higher probability of identifying accessions with a high dormancy rate within botanical variety hypogaea than within the three other botanical varieties tested.
Interaction (genotype x treatment) effect on seed germination and dormancy: Although there was a significant ethephon-treated effect on seed dormancy release or seed germination enhancement, the response to ethephon treatment from different genotypes varied (i.e., interaction effect) (Fig. 4). For example, PI 370331 and PI 497359 were very sensitive to ethephon treatment. They did not germinate at all at day 7 if treated with water only, however they germinated at 100% if treated with ethephon. The accession, PI 157542 was sensitive to ethephon treatment. It did not germinate at day 14 if treated with water only, but it germinated 100% at day 14 if treated with ethephon. In Fig. 4, some of the germinated seeds from PI 157542 under the ethephon treatment had been removed at day 7 from the paper towel. Therefore, only five germinated seeds are shown on Fig. 4 at day 14. These five seeds did not germinate at day 7 but did germinate at day 14. The accession, PI 288210 was not very sensitive to ethephon. It could not reach 100% germination (only 30%) at day 21 with ethephon treatment. Some germinated seeds from PI 288210 under ethephon treatment had also been removed on day 14. Therefore, not all 25 seeds from PI 288210 were shown in Fig. 4 at day 21. The results from Fig. 4 clearly demonstrated that genotype interacted with ethephon treatment on seed dormancy release.
DISCUSSION
Significant differences in seed dormancy have been detected among botanical varieties. There are six botanical varieties within cultivated peanut; however, the U.S. peanut mini-core only contains four botanical varieties (Wang et al., 2011). The botanical varieties hirsuta and aequatoriana are not represented in the U.S. peanut mini-core collection. To completely determine the variability in seed dormancy within the U.S. peanut collection, these two botanical varieties need to be evaluated in future studies. Furthermore, the results from this study were obtained from seeds harvested from one year. More experiments need to be conducted by using seeds harvested from two or more years.
Seed coat can affect seed dormancy or germination. When the entire seed coat was removed seed germination increased 3.3 fold compared with the control (seeds with intact seed coat) in peanut (Toole et al., 1964). Variation in seed-coat colors (from tan to purple) exists among accessions within the U.S. peanut mini-core (Table S1). In other species (such as wheat), seed-coat color can have a significant effect on freshly harvested seed dormancy. In general, white grain wheat seeds have little dormancy whereas red grain wheat seeds have strong dormancy (Wellington, 1956). In this study, we did not observe this kind of correlation. Some accessions with a red-seed coat had strong dormancy whereas other accessions with a red-seed coat had no dormancy. For example, under H2O treatment, accessions, PI 475863 and PI 259851 with red-seed coats had strong dormancy (98 and 100 out of 100 seeds failed to germinate, respectively) whereas PI 475918 and PI 493356 with red-seed coats had no dormancy (all seeds germinated) (Table S1). Within the U.S. peanut mini-core collection, there were 29 accessions with red-seed coats and only 14 accessions had strong dormancy. The difference in seed-coat color on seed dormancy effect we observed between wheat and peanut may be explained by the following two main reasons. The seed-coat structure and composition is different between wheat (monocots) and peanut (dicots). Seed dormancy has been classified into five categories (Baskin and Baskin, 2004). The seed dormancy in wheat and peanut may belong to different categories. Furthermore, peanut seed-coat color is a complicated trait which can be affected by seed maturity. More accessions and research need to be investigated and conducted to draw a conclusion whether seed-coat color can significantly affect seed dormancy in peanut.
ACKNOWLEDGMENTS
The authors gratefully thank Mr. Jerry Davis for his assistance in statistical analysis, Mr. Brandon Tonnis and Mr. Sam Hilton, Mrs. Phiffie Vankus, Sonia Chesnut and Jessica Norris for their help on seed planting, harvesting and germination testing and Drs. Brad Morris and Zhenbang Chen for their suggestions and comments on the manuscript.