Seed ageing can be defined as progressive deterioration of the structures and functions of the seed over time which ultimately leads to death of the organism. The quality of soybean seed during storage has been reviewed and seed deterioration has been identified as one of the basic reasons for low productivity in soybean
(Shelar et al., 2008). Accelerated ageing is the most commonly used stress test to assess the storability of crop seeds in a short period of time
(Mali et al., 2014). The genotypes were grouped on the basis of seed coat colour (Table 1).
Analysis of variance (ANOVA) and variability for seed physical and physiological traits
Seed physical traits of soybean genotypes were recorded in fresh seed lot, while seed physiological parameters were recorded in both freshly harvested seeds and seeds subjected to AA. ANOVA for seed physical parameters of fresh seed exhibited significant variation among all the genotypes for seed length, seed width, seed thickness and 100 seed weight. Significant variation was observed for the seed physiological traits viz percent reduction in germination, seedling length, seedling dry weight, seedling vigour index-I, seedling vigour index-II (Table S1, S2 and S3). Genotypic coefficient of variability (GCV) of soybean genotypes for seed, seedling traits and vigour ranged from 7.65% to 41.12%. The seed physical and physiological traits expressed high heritability estimates ranging from 82.3 to 99.5% (Table 2).
Correlation between seed physical/physiological traits and seed longevity
Physical parameters associated with seed storability like seed coat colour, seed size, seed weight can be helpful in making a decision on duration of soybean seed storage
(Mosavi et al., 2011; Tubic et al., 2011). Seed physical and physiological traits were recorded before and after AA to study longevity of soybean germplasm and phenotypic correlation coefficients were computed (Table 3). Per cent reduction in germination after AA exhibited significant positive correlation with seed traits associated with storability such as seed length (0.40*), seed width (0.58***), seed thickness (0.51***) and 100 seed weight (0.52***). Significant negative correlation was observed with SV-I (-0.55***) and SV-II after ageing (-0.50**). SV-I displayed significant positive correlation with seedling length (0.81 ***), while SV-II displayed significant positive correlation with seedling dry weight (0.55 ***) after AA (data not shown). Similar positive and significant correlations were reported by Kuchlan
et al. (2010b) for seed size and 100 seed weight in soybean. Black-seeded genotypes were better than yellow-seeded genotypes. Traits such as hard seed coated seeds; small seed size and black seed coat were identified to have better storability in soybean
(Subash et al., 2017). Negative correlation is observed in the present study for germination
(Naik et al., 2019) and seedling vigour indices after AA
(Pallavi et al., 2018).
Evaluation of genetic integrity using SSR markers
Genetic integrity is also of equal concern as much as the physiological integrity during seed storage. Decline in germination percent after accelerated ageing was observed in all genotypes and was drastic in NRC 130 and Karune (Table S2). Variation in the SSR profiles after ageing was observed and four markers
viz., Satt 434, Satt 281, Satt 286, Satt 390 were able to detect polymorphisms among AA seeds of soybean. A drastic decrease in the number of alleles has been observed after AA indicating loss of DNA integrity (Table 4). Only one allele was detected for SSR markers Satt 523, Satt 371 and SOYGPATR after AA, while 4-5 alleles were observed in freshly harvested seeds. Loss of alleles due to failure of PCR amplification may be attributed to loss in genetic integrity during AA. It is interesting to note that most of the genotypes with black/brown seed coat have amplified few alleles after AA, indicating loss of genetic integrity was more prominent in green and yellow green seeded genotypes. Reduction in the number of alleles after AA has also been reported in several crops such as rice
(Adeboye et al., 2015), barley
(Parzies et al., 2000), rye
(Chebotar et al., 2003), bread wheat
(Fu et al., 2015), Siberian wild rye
(Huang et al., 2019), soybean and safflower
(Vijay et al., 2009). Assessment of DNA integrity combined with a germination test efficiently characterizes the vigour of seeds after AA and may be a promising tool for long term conservation in seed banks.
Validation of SSR markers for seed longevity
SSR markers associated with seed longevity such as seed length, seed width, seed thickness and 100 seed weight (single marker analysis) were validated in our study. Seven markers (Satt 285, Satt 534, Satt 538, Satt 281, Satt 162, Satt 631, Satt 371) having significant association with seed storability traits have been identified (Table 5). SSR marker Satt 285 was significantly associated with seed physical parameters such as seed length, seed width, seed thickness and seed weight, while Satt 534 was significantly associated with seed thickness, seed width and 100-seed weight. Despite the small sample size and minimum number of markers used in our study we observed significant association of seven SSR markers with seed longevity. This is corresponds with previous findings
(Singh et al., 2008a, b;
Mian et al., 1996). We also identified a few candidate SSR markers that discriminated the good storer soybean genotypes from the rest of the genotypes based on allele size. These markers (Satt 371, Satt 162, Satt 463, Satt 281, Satt 285) were associated with seed specific traits such as seed coat colour, seed size and storability. Our results are in conformity with earlier reports where SSR markers Satt 371, Satt 453 and Satt 618 produced specific allelic bands making them candidate markers for linkage with seed storability and testa colour
(Jagadish et al., 2013a). SSR marker Satt 281 exhibited distinct banding pattern that could clearly differentiate good and poor seed longevity in soybean genotypes
(Naik et al., 2019). Markers Satt 162, Satt 523 and Satt 453 which are either linked with seed coat colour or seed permeability exhibited a specific size allelic fragments in soybean genotypes and crosses revealing better seed longevity
(Pawar et al., 2018). On the other hand, marker Satt 285 showed allelic variation based on seed size.
Candidate markers identified for seed storability
SSR marker Satt 371 amplified in black seed genotypes exclusively at 268-278 bp. In our study, 10 out of 11 good storer genotypes amplified within this size range. Thus, preliminary studies indicate that Satt 371 may have close linkage with good storability that may be attributed to black seed coat. Few other markers such as Satt 162 amplified at 300 bp in 10 out of 11 good storer genotypes and Satt 463 amplified five out of seven black seed genotypes at 110 bp. Eight out of 11 good storer genotypes were amplified at 190-225 bp using Satt 281 marker. Marker Satt 619 amplified at 125 bp in 8 out of 11 good storers, which were black seeded suggesting that Satt 619 is specific to seed coat colour. Marker Satt 534 amplified at 175 bp in extremely poor storer genotypes, while Satt 656 amplified in 8 out of 11 good storer genotypes at around 150 bp indicating its possible association with storability.
Genetic divergence (D2 statistics) of soybean germplasm
The genotypes were grouped into seven clusters on the basis of seed physical and physiological attributes with cluster I being the largest comprising of ten genotypes. Five out of six brown seeded genotypes were grouped in cluster I (Table 6). The average inter-cluster distances varied from 378.08 to 56,445.64 while intra cluster distance ranged from 0.00 to 1300.32. The highest inter-cluster distance was recorded between clusters IV and VII indicating greater diversity (Table 7). Cluster means were computed for six seed traits and four seedling traits (before and after AA) (Table 8). Monogenotypic cluster V (NRC 2755) recorded maximum cluster mean for seedling length, seedling vigour-I & II, seedling dry weight both before and after AA indicating better storability of the germplasm. The monogenotypic cluster III (PSPB-23 [green]) recorded lowest per cent reduction in germination and field emergence after accelerated ageing followed by cluster IV. Crosses may be attempted between genotypes of cluster III and IV with cluster V for enhanced seedling vigour and storability traits. Monogenotypic cluster (VII) consisted of bold seeded germplasm DS91-3 (black seeded) that recorded highest value for seed traits studied such as seed length, width, thickness and 100-seed weight. The genotype NRC 2755 with diverse genetic base is the most prominent contributor for seedling vigour- I and II after AA. The robust vigour after AA is attributed to seedling length and dry weight.
NRC 2755 can be used in crop improvement program for seedling vigour trait. Percent reduction in germination showed maximum contribution (34.5%) towards genetic divergence (Table S5).