Soybean Seed Discoloration and Cracking in Response to Low Temperatures During Early Reproductive Growth

Ancha Srinivasan* and Joji Arihara


* Present address: Regional Science Institute, 4-13, Kita 24 Nishi 2, Kita-ku, Sapporo, 001 Japan. -----

ABSTRACT


Prevalence of low temperatures during flowering adversely affects appearance and quality of soybean [Glycine max (L.) Merr.] seeds at maturity. Our objective was to describe the nature of damage and factors influencing seed quality in a sensitive soybean cv. Kitakomachi, and examine genotypic variation in tolerance to discoloration and cracking. The damage was most evident in plants exposed to 15oC (day/night) for 2-3 weeks commencing from anthesis or one week thereafter. Kitakomachi was sensitive to both discoloration and cracking, while Tokei 795 and Tokei 782 were tolerant to discoloration and cracking respectively. Discoloration was first seen when seeds were about 70-80% of maximum size. Qualitative histochemical tests indicated a greater presence of phenolics in the palisade layer of the discolored seed coat. Cracking was confined to the upper layers of the seed coat and was associated with damage to osteosclereids. More damage to parenchyma in the funiculus and to osteosclereids in the seed coat was noticed in sensitive genotypes following exposure to cold stress. Within a pair of isolines differing in nodulation, nodulating line showed more damage than the non-nodulating line. Additional N fertilization showed varied effects on discoloration depending on duration of chilling and genotype, but it increased cracking. In contrast, excess soil water status increased discoloration. Further improvements in crop tolerance and agronomic practice are necessary to produce quality soybeans in cool climates.

Abbreviations: cv, cultivar; DAF, days after flowering;



-----

INTRODUCTION


Production of soybeans with good appearance and color is essential in preparation of processed foods for human consumption (Taira, 1990). In Japan, cultivars with yellow hilum and seed coat bring a higher price than those with black hilum and colored seed coat, especially for confectionery use. Most yellow hilum genotypes grown in northern Japan yield reasonably well even in cool summers but their seed quality varies considerably. About 28% of seeds produced in Hokkaido in 1987 showed discoloration (manifested as browning of the hilum and surrounding regions) and/or cracking symptoms (Hokkaido Department of Agriculture, unpublished data, 1988). Climatic data suggest that low temperatures during flowering are associated with deterioration of physical quality (unpublished data, 1992).

Discoloration makes seeds unappealing and can indicate undesirable external quality, the presence of toxic metabolites, etc. (Sinclair, 1992). Cracking can lead to a poorer storage stability by allowing a rapid entry of moisture and providing an easy access to microorganisms (Wolf and Baker, 1980). Split seeds yield oil of lower quality than sound seeds (Mounts et al., 1979) and are not favored for making products such as flours. Fungi and viruses cause seed discoloration. The production environment may influence mottling and the expression of seed color (Taylor and Caviness, 1982) but the influence of temperature on discoloration has not been studied. Cracking is caused by external stresses such as mechanical impact during harvesting, handling, and artificial drying (Ting et al., 1980). Internal stresses like rapid cotyledon expansion, and alternate wetting or drying of seeds also favor cracking. However, the effects of low temperatures during seed development on cracking have not been reported.

Pod and seed set in soybean are sensitive to chilling temperatures of 10 to 15oC, and genotypic differences have been reported (Hume and Jackson, 1981; Michailov et al., 1989). Efforts to identify genotypes with a better physical quality at low temperatures have been intensified in Japan recently (S. Yumoto, personal communication, 1992) but the progress has been limited. The effects of agronomic practices on physical quality are unknown. It is not known, for example, if additional N application during flowering affects seed quality. It is also not clear whether isolines differing in nodulation and seed N content vary in their response to chilling, or if excess soil moisture has an influence on seed quality.

This study was undertaken to characterize chilling sensitivity in cv. Kitakomachi, assess genotypic variation in tolerance to discoloration and cracking, and examine the influence of nodulation, N supply and excess soil moisture on seed discoloration and cracking.


-----

MATERIALS AND METHODS

Plant Culture

Four experiments were conducted from May to October 1992 at the Hokkaido National Agricultural Experiment Station, Sapporo, Japan (43o03'N, 141o20'E, 18.6 m above sea level). Six soybeans were planted in plastic pots (15.5 cm diam., 30 cm high) filled with 5.5 kg soil (andisol) supplemented with 1 g ammonium sulphate, 2 g mono calcium phosphate (70 g P kg-1 ), 1 g potassium sulphate, and 4 g fused magnesium phosphate. One week after emergence, seedlings were thinned to two per pot. The pots were irrigated uniformly at regular intervals and pests controlled when necessary. Plants were grown outdoors initially (20 May - 1 September) when mean daily temperatures fluctuated between 18 and 24oC. Canopy temperature was monitored at hourly intervals using copper-constantan thermocouples. Plants were either left outdoors (controls) or transferred 7 days after flowering (DAF) to a phytotron set at 15+0.5oC (day/night) and 75+5% relative humidity for exposing to low temperature. The mean photosynthetic photon flux density (400-700 nm) at canopy level in the phytotron was about 750 µE m-2 s-1. Since night temperatures at this location fall sharply from the first week of September, all plants were shifted on 2 September to a glasshouse with maximum and minimum temperatures set at 25 and 18oC respectively.

Experiments

In Expt 1, plants of Kitakomachi were exposed to 15oC (day/night) for two or three weeks at 0, 7, 14, 21 and 28 DAF to determine the sensitive phase of pod development. In Expt 2, the responses of Kitakomachi and six germplasm lines, Toiku 218 (F15), Toiku 223 (F9), Tokei 776 (F10), Tokei 782 (F9), Tokei 794 (F7) and Tokei 795 (F9), were compared after exposure to 15oC for three or six weeks commencing at 7 DAF. In Expt 3, the responses of a near-isogenic pair of genotypes differing in nodulation (To1-0, non-nodulating and To1-1, nodulating) to chilling were compared under normal and increased N supply. Additional N was applied as 4% ammonium nitrate solution at 30 ml pot-1 day-1 for five times at two-day intervals starting 7 DAF. In Expt 4, plants of Kitakomachi were exposed to normal and excess soil moisture with and without additional N under normal and low temperatures. In the excess water treatment, pots were saturated daily for three weeks, commencing 7 DAF.

Seed Discoloration and Cracking

Pods harvested at maturity were hand-threshed and the number of seeds counted. Seeds were dried at room temperature, weighed and scored. The severity of discoloration in each seed was determined subjectively on a 0-4 scale [0 = no discoloration, 1 = slight (discoloration around hilum only), 2 = moderate (discoloration spreading to other regions but <5% area of the seed coat discolored), 3 = severe (5-25% of the seed coat area discolored), and 4 = very severe (>25% of the seed coat area discolored)]. Cracking was scored on a similar scale [0 = no cracks; 1 = slight (cracking at one place i.e., either micropyle or on one side of the hilum only), 2 = moderate (cracking at two places, i.e., either micropyle and on one side of the hilum or on both sides of the hilum), 3 = severe (cracking at micropyle and on both sides of the hilum), and 4 = very severe (cracking at several places in the seed coat)]. Split seeds were given a cracking score of 4, in view of economic significance.

Histology, Histochemistry and Microscopy

Seeds from the control and low temperature treatments in Expt 1 were immersed in 1M NaOH for 10 min at room temperature. Seeds were rinsed in water five times to remove most of the NaOH and to inhibit further hydrolysis. Free-hand sections of the seed coat were made to examine the reaction of tissues with NaOH. Tissues in discolored and noncolored regions were macerated separately to test differences in the intensity of reaction of cells. It was necessary to complete microscopy within 1 h of treatment as the residual NaOH continued the deteriorative action on cellular integrity.

In Expt 2, green pods were excised at varying intervals. Free-hand or microtome-assisted transverse sections of pods, cut at funicular junction to the seed, were used for the following histochemical tests: Ruthenium red, for pectins (Rae et al., 1985); Phloroglucinol in HCl for lignin (O'Brien and McCully, 1981); Toluidine Blue O (pH 4.4) to detect poly-anionic pectins and phenolics (Feder and O'Brien, 1968); P-nitrobenzene diazonium tetrafluroborate (Harris et al., 1982), FeCl3 in ethanol (Ling-Lee et al., 1977), and NaNO2 (Mace, 1963) for phenolics. For all tests, control reactions were made.

Plant Arrangement and Data Analysis

In all experiments, pots were randomized both in the phytotron and outdoors and repositioned at least twice a week. Treatments were repeated on at least two cycles over time. Data were pooled from all cycles as error variances were homogeneous, and no significant (P < 0.05) cycle by treatment interactions were found. All data were subjected to analysis of variance. Mean discoloration and cracking indices were computed on a plant basis as the weighted means of individual scores of all seeds in a plant as follows: Discoloration or Cracking index = [(Seeds with score 4 x 4) + (Seeds with score 3 x 3) + (Seeds with score 2 x 2) + (Seeds with score 1 x 1) + (Seeds with score 0 x 0)]/ total number of seeds per plant.
-----

RESULTS

Analysis of temperatures during the past 10 cropping seasons of cv. Kitakomachi at Memuro, a representative location of the soybean-producing region in Hokkaido, showed no clear association between mean seasonal temperatures and seed quality. Temperatures during the first 35 DAF, however, showed that deterioration of external seed quality in 1983, 1987, 1988 and 1991 was associated with low temperatures in those years (Table 1). Temperature in these years fell below 15oC for as much as 12-13 h within a day, especially during the first 2-3 weeks after flowering (data not shown).

Stage of Sensitivity to Low Temperature and Reactions with the Alkali
The controls showed no seed damage; however, exposure to 15oC, led to reductions in seed number, weight and quality (Table 2). In general, severity of damage increased with duration of exposure. Plants exposed during the first 14 DAF showed more damage than those exposed later. The discolored area was <25% of the seed coat in most cases, but it was as high as 40-50% in plants cold-stressed for three weeks commencing 7 DAF. Cracking was seen mostly in the discolored regions and near the micropyle, and was confined to the upper layers of the seed coat with palisade and osteosclereid tissues missing.

Seeds from the controls showed no differential reaction with NaOH. In discolored seeds, however, the boundary between discolored and noncolored areas was distinct. Cells within the discolored area were strongly stained by NaOH, and those in the noncolored area remained unstained. Staining was confined to the palisade tissue and to a few osteosclereids in discolored regions. In the hilum region, however, staining was observed in the palisade and counter palisade cells, sclerenchyma, and stellate cells (Table 3).

Genotypic Variation in Tolerance to Seed Discoloration and Cracking
At normal temperatures, all genotypes produced seeds of good quality with Tokei 776 and Tokei 782 yielding more than the others. Distinct genotypic differences in discoloration and cracking were, however, found upon exposure to 15oC. While Kitakomachi was the most sensitive, Tokei 795 and Tokei 782 were tolerant to discoloration and cracking respectively (Table 4). The nature of cracking varied with genotype. A few but large cracks were found in Kitakomachi, while many small cracks were present in Tokei 795.

Genotypic variation was evident mainly during R6, when seeds were 70-80% of their maximum size. At this stage, brown discoloration near the hilum and a few cracks in the seed coat were found in Kitakomachi. Splitting was also conspicuous, especially in plants cold-stressed for six weeks. Tolerant genotype Tokei 795, however, showed little discoloration. The cell walls of palisade cells reacted with ruthenium red indicating pectins but no genotypic variation was found. Since the function of lignin is primarily structural and protective, a qualitative test with phloroglucinol-HCl was used but no clear differences were found. With toluidine blue and p-nitrobenzene diazonium tetrafluroborate, however, a greater intensity of staining was observed in the palisade layer of the discolored seed coat, indicating a greater presence of phenolic compounds. The tests with FeCl3 and NaNO2 also showed an intense staining of palisade layer in the discolored seed coat. Thus, cold-affected seeds of Kitakomachi showed a broken cuticle, rough epidermis, damaged parenchyma in the funiculus and osteosclereids in the seed coat, and a greater intensity of reaction with the stains specific for phenolics. In contrast, Tokei 795 seeds showed less cellular damage in the funiculus and no clear differences in the intensity of reaction for phenolics in the seed coat (Table 5).

Responses of Isolines to Low Temperature and Nitrogen
The nature of seed quality response to cold in both isolines was similar to that seen in other genotypes. The damage was more, however, in the nodulating than in the non-nodulating isoline. Additional N increased seed yields of both lines but the effects on seed quality varied. Additional N increased discoloration in the 3-week stress but not in the 6-week stress. It enhanced cracking in both chilling treatments however (Table 6).

Responses to Saturated Water Status and Nitrogen

Quality response of Kitakomachi to water status and N was temperature-dependent. Excess soil moisture did not affect physical quality at normal temperatures but it enhanced mean discoloration index at low temperature. In contrast, N had no influence on discoloration but it increased cracking at low temperatures. The damage was most severe when both excess soil water and additional N were applied during cold stress (Table 7).
-----

DISCUSSION

Both discoloration and cracking are undesirable for confectionery use of soybeans. Weather factors per se were presumed not to cause discoloration (Sinclair, 1992), but this study has shown that low temperature, especially during early flowering, substantially reduces physical quality. The findings are relevant not only to Hokkaido and other parts of northern Japan, but also to regions in Canada, China, northern USA and Europe, where temperatures around 15oC during flowering are not uncommon at least in some years (Holmberg, 1973).

The sensitivity to chilling was evident only during early stages. As the damage was most pronounced when plants experienced cold stress from 7 DAF, it seems that the sensitive stage coincides with the period of transition from division to differentiation of cells in the seed coat. Palisade and hourglass cells of the seed coat in soybean differentiate from 6-8 and 12-14 days after fertilization respectively, and complete differentiation by 18-20 days (30% max. seed size) after fertilization (Baker et al., 1987).

Significant genotypic variation was found in seed quality response to cold stress. Tokei 795 was tolerant to discoloration and Tokei 782, to cracking. The parents of these genotypes (Toyomusume and Tsurukogane; and Toiku 207 and Kitahomare respectively) were, however, sensitive to discoloration and cracking (Unpublished data, 1993). The responses thus seem to be cases of transgressive segregation and tolerance to discoloration and cracking may be a quantitative trait under the control of two or more genes. Discoloration and cracking were closely related in Kitakomachi (r=0.79***, n=100) but not in other genotypes (e.g., Tokei 795, r=0.09, n=100). This suggests that both symptoms may be regulated independently. Other studies indicate that the genes associated with tolerance to discoloration are linked with those for late maturity (R. Takahashi, personal communication, 1993), which complicates production in regions with a relatively short frost-free growing season.

Treating the seed coats with NaOH, which results in oxidation of phenols, led to a greater intensity of browning in the discolored regions. Werker et al. (1979) indicated that the appearance of dark brown coloration after NaOH treatment was due to oxidation of phenolics to quinones. Further, palisade tissue in the discolored seed coat showed a greater intensity of reaction with stains specific for phenolics. The results thus suggest that discoloration is probably due to a greater accumulation and/or oxidation of phenolics in the seed coat. An increased production of phenolics following exposure to chilling (e.g., Chalker-Scott and Fuchigami, 1989), and discoloration of plant products following oxidation of phenolics by polyphenol oxidase is well documented (e.g., Walter and Purcell, 1980). The rupture of cell membranes (as evidenced by increased cracking in the discolored regions) may also contribute to discoloration. Chalker-Scott and Fuchigami (1989) postulated that compartmentalized phenolics and enzymes are released into the cell simultaneously following a rupture of membranes, and that the phenolics lose their esterified moieties and freely react with proteins to form insoluble brown complexes. Genotypic differences in discoloration may be due to variation in production of substrate (e.g., catechol) and/or enzyme (polyphenol oxidase) as reported in other crops (e.g., sweet potato, Scott and Kattan, 1957). Discoloration response was not seen however, in plants experiencing cold during later stages of flowering. The reasons for this are not entirely clear but it is likely that development of pod walls at the time of exposure to cold play some protective role. Discoloration of pod wall without any effects on seed was more in the later treatments than in the early treatments (data not shown).

The nodulating isoline showed more damage than the non-nodulating line, irrespective of variations in N supply. This is contrary to others' findings that N-deficient plants produced more phenolics than those supplied with N (DiCosmo and Towers, 1983). Discoloration response to additional N varied with genotype. While additional N increased discoloration at low temperatures in To1-1 in Expt 3, it did not show any influence in Kitakomachi in Expt 4. Further research is warranted before any conclusion can be made.

Excess soil moisture status increased discoloration at 15oC. A great deal of evidence exists in the literature to suggest that ethylene production is enhanced under waterlogged conditions (e.g., Jackson and Campbell, 1976) and by chilling (e.g., Wang, 1989) and that ethylene production and phenolic synthesis are closely related (e.g., Chalker-Scott and Fuchigami, 1989). It is possible that both chilling and anaerobic conditions favor an increased production of phenolics in developing seeds. Baker and Minor (1984) suggested that spongy osteosclereids provide soybean seed coat with an elasticity that may aid in the prevention of cracking. Microscopic observations showed a greater damage to osteosclereids near the hilum in sensitive genotypes. Their seed coats may therefore lose elasticity and are easily prone to cracking. Additional N supply during chilling increased cracking probably due to a more rapid cotyledon expansion following transfer to normal temperatures (Wolf and Baker, 1980).

The above experiments indicate the interactive influence of different factors on seed quality at low temperatures. While enhancing genotypic tolerance to chilling stress, it seems desirable to make improvements in agronomic practice simultaneously. Top-dressing of N during flowering increased seed yields in this and other studies (Kuwahara et al., 1986), but it showed adverse effects on quality with reference to cracking. It seems desirable therefore to modify the time of N application, especially when low temperatures prevail. Adverse effects of excess soil water on quality are relevant to several regions in Hokkaido (e.g., Sorachi, Kamikawa and Abashiri districts), where heavy clay (>50% clay) soils predominate. Efforts to improve drainage in these soils, especially in cool years, are likely to have positive effects on quality. Cropping history also appears to influence seed quality response to low temperatures. The damage to a sensitive genotype, Toyomusume, was less in a field where soybean was grown continually than in a field where paddy was grown before (unpublished data, 1993). The observations are important as converted paddy fields account for nearly 60-70% of the acreage under soybeans in Japan and up to 40% in Hokkaido (Kuwahara et al., 1986). The influence of soil factors on seed quality at low temperatures needs to be examined further and testing of new genotypes in various soils is necessary to improve their overall adaptability.
-----

ACKNOWLEDGMENTS

A.S. is grateful to the Japan International Science and Technology Exchange Center and Research Development Corporation of Japan for awarding a Science and Technology Agency Fellowship. We thank S. Yumoto and I. Matsukawa for providing seed samples, M. Oda for climatic data, and H. Hamaguchi, M. Moriyama, Y. Takahashi, M. Sasaki and S. Sato for general assistance.
-----

REFERENCES

  1. Baker, D.M., and H.C. Minor. 1984. Ultrastructure and function of the soybean (Glycine max) seed coat and hilum: An SEM approach. Agron. Abstr. American Society of Agronomy, Madison, WI. p. 141.
  2. Baker, D.M., H.C. Minor, and B.G. Cumbie. 1987. Scanning electron microscopy examination of soybean testa development. Can. J. Bot. 65:2420-2424.
  3. Chalker-Scott, L., and L.H. Fuchigami. 1989. The role of phenolic compounds in plant stress responses. p. 81-90. In P.H. Li (ed.) Low temperature stress physiology in crops. CRC Press, Boca Raton, FL.
  4. DiCosmo, F., and G.H.N. Towers. 1983. Stress and secondary metabolism in cultured plant cells. p.97-121. In B.N. Timmermann, C. Steelink, and F.A. Loewus (ed.) Phyto-chemical adaptations to stress. Plenum Press, New York.
  5. Feder, N., and T.P. O'Brien. 1968. Plant microtechnique: some principles and new methods. Am. J. Bot. 55:123-142.
  6. Harris, P.J., R.D. Hartley, and G.E. Barton. 1982. Evaluation of stabilized diazonium salts for the detection of phenolic constituents of plant cell walls. J. Sci. Food. Agric. 33:516-520.
  7. Holmberg, S.A. 1973. Soybeans for cool temperate climates. Agric. Hort. Genet. 31:1-20.
  8. Hume, D.J., and A.K.H. Jackson. 1981. Pod formation in soybeans at low temperatures. Crop Sci. 21:933-937.
  9. Jackson, M.B., and D.J. Campbell. 1976. Waterlogging and petiole epinasty in tomato: the role of ethylene and low oxygen. New Phytol. 76: 21-29.
  10. Kuwahara, M., S. Hoshi, and T. Yoshida. 1986. Soil management and nitrogen fertilization for increasing the yield of soybean in Japan. Tech. Bull. No. 98, Food Fert. Tech. Cent. Taiwan.
  11. Ling-Lee, M., G.A. Chilvers, and A.E. Ashford. 1977. A histochemical study of phenolic materials in mycorrhizal and uninfected roots of Eucalyptus fastigiata Deane and Maiden. New Phytol. 78:313-328.
  12. Mace, M.E. 1963. Histochemical localization of phenols in healthy and diseased banana roots. Physiologia Pl. 16:915-923.
  13. Michailov, V.G., V.F. Marjushkin, and V.I. Sichkar. 1989. Cold tolerance during reproductive growth of soybean. Soybean Genet. Newslett. 16:211-213.
  14. Mounts, T.L., G.R. List, and A.J. Heakin. 1979. Postharvest handling of soybeans: Effects on oil quality. J. Am. Oil Chem. Soc. 56:883-885.
  15. O'Brien, T.P., and M.E. McCully. 1981. The Study of plant structure: Principles and selected methods. Termacarphi Pty, Melbourne.
  16. Rae, A.L., P.J. Harris, A. Bacic, and A.E. Clarke. 1985. Composition of the cell walls of Nicotiana alata Link et Otto pollen tubes. Planta 166:128-133.
  17. Scott, L.E., and A.A. Kattan. 1957. Varietal differences in the catechol oxidase content of the sweet potato root. Proc. Am. Soc. Hort. Sci. 69:436-442.
  18. Sinclair, J.B. 1992. Discoloration of soybean seeds - An indicator of quality. Plant Dis. 76:1087-1091.
  19. Taira, H. 1990. Quality of soybeans for processed foods in Japan. Japan Agric. Res. Quarterly 24:224-230.
  20. Taylor, B.H., and C.E. Caviness. 1982. Hilum color variation in soybean seed with imperfect black genotype. Crop Sci. 22:682-683.
  21. Ting, K.C., G.M. White, I.J. Ross, and O.J. Loewer. 1980. Seed coat damage in deep-bed drying of soybeans. Trans. ASAE 23:1293-1296, 1300.
  22. Walter, W.M. Jr., and A.E. Purcell. 1980. Effect of substrate levels and polyphenol oxidase activity on darkening in sweet potato cultivars. J. Agric. Food Chem. 28:941-944.
  23. Wang, C.Y. 1989. Relation of chilling stress to ethylene production. p. 177-192. In P.H. Li (ed.) Low temperature stress physiology in crops. CRC Press, Boca Raton, FL.
  24. Werker, E., I. Marbach, and A.M. Mayer. 1979. Relationship between the anatomy of the testa, water permeability and the presence of phenolics in the genus Pisum. Ann. Bot. 43:765-771.
  25. Wolf, W.J., and F.L. Baker. 1980. Scanning electron microscopy of soybeans and soybean protein products. Scanning Electron Microsc. III:621-634.
-----

If you have comments or need further information, please email me at ancha_s@yahoo.com
-----

updated 15 January 1999 1