1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References

Depending on the environmental conditions, imbibed seeds survive subzero temperatures either by supercooling or by tolerating freezing-induced desiccation. We investigated what the predominant survival mechanism is in freezing canola (Brassica napus cv. Quest) and concluded that it depends on the cooling rate. Seeds cooled at 3°C h−1 or faster supercooled, whereas seeds cooled over a 4-day period to −12°C and then cooled at 3°C h−1 to−40°C did not display low temperature exotherms. Both differential thermal analysis and nuclear magnetic resonance (NMR) spectroscopy confirmed that imbibed canola seeds undergo freezing-induced desiccation at slow cooling rates. The freezing tolerance of imbibed canola seed (LT50) was determined by slowly cooling to −12°C for 48 h, followed with cooling at 3°C h−1 to −40°C, or by holding at a constant −6°C (LD50). For both tests, the loss in freezing tolerance of imbibed seeds was a function of time and temperature of imbibition. Freezing tolerance was rapidly lost after radicle emergence. Seeds imbibed in 100 μM abscisic acid (ABA), particularly at 2°C, lost freezing tolerance at a slower rate compared with water-imbibed seeds. Seeds imbibed in water either at 23°C for 16 h, or 8°C for 6 days, or 2°C for 6 days were not germinable after storage at −6°C for 10 days. Seeds imbibed in ABA at 23°C for 24 h, or 8°C for 8 days, or 2°C for 15 days were highly germinable after 40 days at a constant −6°C. Desiccation injury induced at a high temperature (60°C), as with injury induced by freezing, was found to be a function of imbibition temperature and time.

Abbreviations – 

abscisic acid


differential thermal analysis


time to kill 50%


lowest temperature to kill 50%


low temperature exotherm


moisture content dry weight basis


nuclear magnetic resonance.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References

In temperate climates, seed survival in the soil at subzero temperatures is important for the propagation of both native and introduced species. In winter, soil temperatures as low as −10°C to −30°C in the surface (4 cm) can occur for periods of 2–3 weeks when the air temperature is −30°C or lower. These subzero temperatures result in the death of imbibed seeds.

Several reports are available on the freezing tolerance of mature seeds (Juntilla and Stushnoff 1977, Keefe and Moore 1981, Ishikawa and Sakai 1982, Becwar et al. 1983). Most seeds with a moisture content on a dry weight (MCDW) basis of 12% can survive direct immersion in liquid nitrogen (−196°C) because the water in these seeds is considered non-freezable (Burke et al. 1976). However, freezing tolerance decreases with increasing seed moisture and is reduced dramatically with the emergence of the radicle. For example, seed of Grand Rapids lettuce (at 5–13% water) survived−196°C while imbibed seed (13–16% MCDW) survived only to −40°C; yet after radicle emergence (greater than MCDW), injury occurred as soon as freezing was initiated (Juntilla and Stushnoff 1977). These authors demonstrated that lettuce seeds, with a moisture content greater than 20% (MCDW), can avoid freezing by supercooling as determined by differential thermal analysis (DTA). Low temperature exotherms (LTEs) were detected as low as −29°C, when imbibed lettuce seeds were cooled rapidly (>20°C h−1). Becwar et al. (1983) also noted imbibed seeds from 12 species avoided freezing injury by supercooling, as determined by the presence of LTEs, when cooled at 120°C h−1. Ishikawa and Sakai (1982) demonstrated that the supercooling ability of lettuce seeds was dependent on the cooling rate. Lettuce seeds (48% water) cooled at a rate of 5°C h−1 displayed LTEs at −14°C to −15°C, which coincided with the temperature at which the seeds were killed. However, lettuce seeds of a similar moisture content tolerated −50°C when cooled at a rate of 1.3°C h−1, and no LTE was observed. Most dry seeds (<12% MCDW) can survive direct immersion in liquid nitrogen (−196°C), yet dry seeds of Linum usitatissimum and Corylus cornuta may possibly die when cooled to−196°C in liquid nitrogen (Becwar et al. 1983). Air-dried seeds of Eucalyptus pauciflora and Eucalyptus camaldulensis cannot tolerate −32°C (Cremer and Mucha 1985). Also, coffee (Coffea arabica) seeds tolerated dehydration to 8% but did not survive exposure to liquid nitrogen (Becwar et al. 1983).

Water uptake of dehydrated seeds is triphasic (Bewley 1997). The first stage (imbibition) involves passive water uptake, generally to moisture contents that are less than 20% by weight, and is not dependent on seed viability. The length of the second- or plateau-stage is dependent on seed dormancy. In the third and final stage, cell division and elongation occur with water uptake postulated to occur by an active process (Chrispeels and Maurel 1994). During the third stage, seed moisture content increases from approximately 30 to 50% or greater, and the seed loses most of its stress tolerance following radicle emergence (Bewley 1997).

Nothing is known of the freezing characteristics of imbibed canola seeds, an important oil seed crop. Volunteer seedlings of canola often occur in previous canola stubble fields the following spring (Kirkland and Johnson 2000). When successful, fall-seeded spring cultivars of canola often have a greater seed yield, have a higher oil level, mature 1–3 weeks earlier, escape heat, drought and frost stresses, and have reduced insect damage, compared with spring seeding (Kirkland and Johnson 2000). Unfortunately, the window of success for fall seeding is very narrow and the time of seeding is unpredictable from year to year. The objectives of this study were to determine: (1) whether canola seeds supercool to avoid injury or whether they tolerate freezing-induced desiccation when exposed to subzero temperatures; (2) the influence of seed moisture content and temperature of water uptake on the freezing tolerance of imbibed canola seeds; and (3) the effect of abscisic acid (ABA) on the freezing tolerance of imbibed canola seeds. We present evidence that the loss of freezing tolerance in imbibed canola seeds is a function of imbibition level and imbibition temperature.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References

DTA of imbibed canola seeds

Seeds of canola (Brassica napus, cv. Quest) from the current harvest year were used throughout this study. The canola seeds were imbibed in either water or 100 μM ABA at 23°C, 8°C, and 2°C for various times to obtain a range of moisture contents and different stages of germination (Gao et al. 1999). The seed moisture content was calculated as percentage of dry weight following drying at 60°C in a forced-air oven for 2 days or until a constant weight was achieved.

Seeds were either bare or wrapped in moist tissue containing either ice or finely ground ‘Thermica’, which initiated freezing at −2°C to −4.5°C. The supercooling ability of imbibed seeds was determined by DTA as described by Gusta et al. (1983). Depending on the study, the imbibed canola seeds were cooled either at 5°C h−1 or 25°C h−1 or in a stepwise manner as follows: −3.5°C for 1.5 h, −5°C for 20 h, −7°C for 24 h, −12°C for 48 h, and then cooled at 3°C h−1 to−40°C. In another study, seeds cooled in the stepwise manner to −40°C h−1 were warmed to 1°C for either 15 or 60 min and then re-cooled at 5°C h−1 to −40°C. Exotherms, as a result of the crystallization of water, were recorded on potentiometric recorders at 0.5 μV, connected to samples with 28-gauge copper-constantan thermocouples. Each DTA analysis had five to six seeds and was replicated four to five times.

Nuclear magnetic resonance analysis of canola seed

Nuclear magnetic resonance (NMR) measurements were made at 20 MHz on a Bruker Minispec PC 20 pulsed NMR spectrometer, as described previously by Gusta et al. (1975). Free induction decay was measured 20 μs after the second and each subsequent radio frequency pulse. Pulses were 3 s apart to prevent saturation. The uncorrected free induction decay signal was assumed proportional to the liquid water content after the Boltzmann temperature correction was approximated by multiplying all NMR signals by the absolute temperature. Canola seeds were imbibed at 2°C for 2 days and then placed in a 3 mm NMR tube (three to five seeds per tube). The seeds were cooled at 2°C h−1 to−40°C, or to a constant −10°C for 48 h, and then cooled at 2°C h−1 to −40°C. The seeds were re-warmed at a rate of 2°C h−1.

ABA and temperature effects on germination and freezing tolerance

Canola seeds were imbibed at either 23°C, or 8°C, or 2°C in the dark on two layers of filter paper in 1.5 × 9 cm Petri dishes containing 4 ml of either water or 100 μM ABA. Each Petri dish contained 50 seeds, and each treatment was replicated four times. The LT50 (temperature at which 50% of the seeds were killed) and LD50 (time at −6°C to kill 50% of the seeds) were determined as described below. S-ABA, a generous gift from Toray Inc., Japan, was initially dissolved in 6 ml of 1 N KOH prior to dilution in water.

Determination of LT50 of imbibed seed

Canola seeds imbibed in either H2O or 100 μM ABA (25 per test-temperature) were placed on moist tissue in a 24 × 1.5 cm Pyrex test tube cooled in an ethylene glycol bath. Freezing was initiated at −2.5°C with ice crystals. Seeds were cooled in the following stepwise manner: −2.5°C for 1 h, −3°C for 4 h, 5°C for 3 h,−7°C for 7 h, then −12°C for 48 h, and cooled thereafter at 3°C h−1 to −80°C and then immersed directly in liquid N2. This cooling protocol did not display any LTEs as determined by DTA. A series of 10 test-temperatures, separated by either 3°C or 5°C intervals, were used to determine the lowest temperature that killed 50% of the imbibed seeds (LT50). Upon reaching the selected test temperature, the samples were removed from the ethylene glycol bath, thawed overnight at 4°C, and held in the original test tubes at 23°C for up to 2 weeks. The freezing tests were replicated three times. Germination and growth of the seedlings were used to evaluate the LT50 of the seeds.

Effect of the duration at −6°C on survival of imbibed canola seed (LD50)

Imbibed canola seeds were cooled in the stepwise manner as described above to −6°C and held isothermal for up to 90 days to determine the effect of prolonged freezing on germination. LTEs were not observed employing this temperature regime. After pre-selected times, the seeds were removed from the −6°C test-temperature, thawed slowly (2°C h−1) to 23°C, and held at this temperature for 7 days to determine 50% germination (LD50) and seedling development.

Desiccation tolerance of canola seed

Seeds were imbibed in either water or ABA at 8°C for 4 days or at 2°C for 15 days. Germination was recorded each day for seeds held at 8°C and every third day for seeds held at 2°C. Following 1, 2, 3, and 4 days at 8°C, and 6, 9, 12, and 15 days at 2°C, seeds were air-dried at 60°C for 2 days and then placed in Petri dishes as described above for the germination assays at 23°C.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References

DTA of imbibed canola seeds

DTA profiles of canola seeds imbibed at 2°C for 3 days and cooled at various rates are shown in Fig. 1. Similar results were observed for seeds imbibed at 23°C or 8°C, provided the seeds did not complete germination (data not presented). Imbibed canola seeds did not display a high temperature exotherm (HTE), as observed for lettuce (Juntilla and Stushnoff 1977, Ishikawa and Sakai 1982, Keefe and Moore 1980). LTEs, if any, were observed between −13°C and −24°C (Fig. 1). Seeds imbibed at higher temperatures, which allowed germination to proceed faster, had LTEs at higher temperatures (−7°C to −14°C, data not presented). Generally, the magnitude and temperature of the LTE was a function of the cooling rate (Fig. 1A vs B). In comparison with a cooling rate of 25°C h−1, imbibed seeds cooled at 5°C h−1 had smaller LTEs that occurred at a lower temperature. The lowest temperature that killed 50% of the imbibed seeds (LT50) cooled at 25°C h−1 was −15°C vs −21°C for imbibed seeds cooled at 5°C h−1. When imbibed seeds were held at −5°C for 20 h and then cooled at 3°C h−1, the LTE was smaller in magnitude compared with imbibed seeds cooled at 5°C h−1 (Fig. 1C). However, the temperature for the LTE and LT50 were similar for both cooling rates. LTEs were not detectable if the imbibed seeds were held at −3.5°C for 1.5 h, −5°C for 20 h, −7°C for 24 h, and −12°C for 48 h and cooled at 3°C h−1 to −40°C (Fig. 1D). Similar results were observed when the seeds were wrapped in moist tissues and nucleated at −2°C to −4°C (data not presented). When imbibed canola seeds were cooled slowly to −40°C (as outlined in Fig. 1D), warmed to 1°C for 15 min, a small LTE was evident when these seeds were re-cooled at 5°C h−1 (compare Fig. 1B, E). If the warming period was extended to 60 min, the LTE was larger and occurred at warmer temperatures (Fig. 1F). These results suggest that water migrates slowly from the embryo during slow cooling (as outlined in Fig. 1D), resulting in freezing-induced desiccation. Because this freezing event is slow, DTA did not detect the water freezing outside of the embryo. During thawing, water must migrate quickly to its original site(s), similar to what has been observed in the xylem vessels of hardwood trees (Gusta et al. 1983).


Figure 1. Differential thermal analysis of canola seeds imbibed at 2°C for 3 days cooled at a rate of (A) 25°C h−1 (LT50 = −15°C); or (B) 5°C h−1 (LT50 = −21°C) from 0°C to −30°C; canola seeds held at (C) −5°C for 20 h, and then cooled at 3°C h−1 (LT50−21°C), (D) −3.5°C for 1.5 h followed by −5°C for 20 h, followed by −7°C for 24 h, followed by−12°C for 48 h, and then cooled at 3°C h−1 to −40°C (LT50−196°C) (E) same as (D) except seeds were warmed (5°C h−1) to 1°C for 15 min and then re-cooled at 5°C h−1 (LT50−23°C) (F) same as (E) except warming period at 1°C extended 60 min (LT50−19°C).

Download figure to PowerPoint

Several studies have demonstrated that the LTE coincides with the killing temperature of the imbibed seed (Juntilla and Stushnoff 1977, Keefe and Moore 1981, Ishikawa and Sakai 1982). However, Bai et al. (1998), employing a cooling rate of 2.5°C h−1, did not observe a correlation between the LTE and seed death in fully hydrated winterfat (Eurotia lanata) seeds. Ishikawa and Sakai (1982) also noted that the LTE in several species was not detectable, if the cooling rate was less than 1.1°C h−1. It is speculated when seeds are cooled very slowly water migrates from the embryo to spaces within the seed that accommodate the growing ice crystal and the LTE becomes undetectable (Keefe and Moore 1981, Ishikawa and Sakai 1982). In several species, the endosperm envelope acts as a barrier to ice growth into the embryo (Gazeau and Dereuddre 1980, Keefe and Moore 1981). In this study, the LTE coincided with the killing temperature of the seeds; however, the seeds were killed at a lower temperature if they did not display a LTE.

Unlike many species of hardwood trees whose xylem ray parenchyma cells display supercooling to as low as −40°C, none of the imbibed canola seeds supercooled beyond −25°C. Water in the wood of hardwood trees, such as Quercus coccinea, can remain in a supercooled state for months (Gusta et al. 1983), even at temperatures approaching −40°C. The wood of Fraxinus pennsylvanica displays a LTE at −51°C when cooled at 40°C h−1; however, samples held at −30°C for several weeks do not display a LTE when cooled to −55°C (Gusta et al. 1983). It is assumed that during periods of prolonged exposure to subzero temperatures, water migrates from the supercooled cells to external sites of lower water potential (Gusta et al. 1983). The rate of water migration from the symplast to the apoplast is dependent on both cell wall flexibility and the vapor pressure difference between supercooled water and the air. Results suggest supercooled water in imbibed seeds migrates to external sites much quicker compared with woody species. This may be due in part to the flexibility of the cell wall.

Pulsed NMR spectroscopy of canola seed water

Canola seeds, imbibed in water at 2°C for 2 days, displayed a strong NMR water signal when cooled from 2°C to −18°C at 2°C h−1 (Fig. 2). At temperatures approaching −20°C, the water signal decayed rapidly due to the conversion of water from the liquid to the solid state (ice) (Fig. 2). In sharp contrast, when seeds were cooled at 2°C h−1 to −10°C and held constant at this temperature, the water signal slowly decreased over a 48-h period and became undetectable (open circle, Fig. 2). When the seeds were cooled to −40°C and then re-warmed at 2°C h−1, less than 3% of the original freezable water displayed a signal at −6.5°C, whereas all of the freezable water was in a liquid state at −2°C (Fig. 2). The conversion of ice to a liquid form at subzero temperatures has been demonstrated in both woody tissues (Gusta et al. 1983) and in herbaceous tissues (Gusta et al. 1975). The thawing curves of imbibed canola seeds were similar to those of cereal crowns, except 100% of the freezable water of imbibed canola seeds was in a solid state at −10°C compared with 75–80% for cereal crowns. The NMR data confirm the DTA observations that a LTE occurred, although a HTE did not.


Figure 2. Freezing and thawing curve of imbibed canola seeds (imbibed at 2°C for 3 days) exhibiting the effect of cooling and thawing rate and temperature on liquid water, as determined by nuclear magnetic resonance, as percentage of total water. Circles represent liquid water signal for seed held at −10°C for 48 h and then cooled at 2°C h−1 to −40°C. Squares represent seeds cooled from 0°C to −40°C at 2°C h−1. Solid units represent the freezing curve and open units represent the thawing curve. Dashed line represents loss of water signal for seeds held at −10°C for 48 h.

Download figure to PowerPoint

LT50 of canola seeds

In previous studies regarding seed-freezing tolerance, imbibition was generally conducted at room temperature. In this study, seeds were imbibed at either 8°C or 2°C in the presence or absence of 100 μM ABA (Table 1). After 5 days at 8°C, approximately 80% of the seeds imbibed in water completed germination compared with 50% of the ABA treated seeds. After 7 days at 2°C, 90% of the seeds imbibed in water completed germination; yet, none of the ABA treated seeds completed germination by 15 days. If the ABA-treated seeds (at 2°C) were washed in water briefly and transferred to water at 23°C, germination was completed within 48 h. ABA, other than delaying completion of germination, had no visible effects on seedling morphology.

Table 1.  Freezing tolerance (LT50) of canola (cv. Quest) seed imbibed in either water or 100 μm abscisic acid (ABA) at 8°C or 2°C. Seeds were wrapped in moist tissues, ice nucleated, and held at −2.5°C for 1 h, −3°C for 4 h, −5°C for 3 h, −7°C for 7 h, and −12°C for 48 h and cooled at 3°C h−1 to −40°C and then placed directly in liquid N2. Values in parenthesis indicate average percentage seed water as determined by drying at 60°C for 2 days or until a constant weight was achieved, ±sd. a 80% germination; b seeds not germinated, but 50% germination by 5 days; c 90% germination; d 25% germination; e 95% germination.
Imbibition time (days)Water (LT50°C)ABA (LT50°C)Water (LT50°C)ABA (LT50°C)
1>−196.0 (36.8 ± 0.7)>−196.0 (39.2 ± 6.3)>−196.0 (39 ± 1.7)>−196.0 (28.3 ± 0.9)
2−30.0 (41.0 ± 0.8)−30.0 (39.4 ± 1.0)  
3−25.0 (42.4 ± 1.5)−30.0 (42.3 ± 1.2)−196.0 (40.3 ± 2.1)>−196.0 (40.5 ± 0.4)
4−5.0a (46.0 ± 0.9)−10.0b (42.9 ± 1.7)  
5  −60.0 (41.4 ± 0.35)−196.0 (41.4 ± 1.4)
7  −30.0c (45.0 ± 0.35)−45.0 (41.9 ± 0.8)
15  −19.0 (56.6 ± 0.52)−45.0d (45.2 ± 0.5)
15 (1 day at 23°C)  −24.0 
15 (3 day at 23°C)  −3.0e 

Seeds imbibed in either water or ABA at 8°C for 1 day tolerated −196°C; however, the LT50 of seeds imbibed in either water or ABA increased the LT50 to −30°C after 2 days (Table 1). During this period, the seed water content ranged from 39 to 41%. After 4 days at 8°C, water-imbibed seeds had a LT50 of −5°C, whereas seeds imbibed in 100 μM ABA had a LT50 of −10°C. The water content of the seeds was 46% for the water-treated seeds and 43% for the ABA-treated seeds. At 2°C, the loss of freezing tolerance was much slower compared with 8°C even though the water content of these seeds was similar. For example, after 3 days of imbibition at 2°C, the seeds still tolerated −196°C although their water content was approximately 40% (Table 1). By day 5 at 2°C, water-imbibed seeds tolerated −60°C, whereas the ABA-treated seeds tolerated −196°C even though the seed water content was similar (Table 1). Surprisingly, seeds imbibed in water at 2°C for 7 days had an LT50 of −30°C, even though the radicle had emerged. At this point, the water content of the 2°C imbibed seeds was similar to the water content of seeds imbibed at 8°C for 4 days; however, the LT50 of the 8°C imbibed seeds was −5°C.

On day 15 at 2°C, water-imbibed seeds still tolerated −19°C, in contrast to −45°C for seeds treated with ABA. Therefore, LT50 is not entirely a reflection of water content but also dependent upon the temperature of water uptake and the presence of ABA. Andrews (1958) reported an LT50 of −15°C for hardy winter wheat seed imbibed at 0.5°C for 4 weeks. These results suggest that the embryo has the potential to cold acclimate when exposed to low acclimating temperatures or does not rapidly lose freezing tolerance under these conditions. Juntilla and Stushnoff (1977) reported freezing tolerance was completely lost upon radicle emergence of lettuce seeds. In that study, the seeds were imbibed at warm temperatures, whereas in this study, the seeds were imbibed at cold acclimating temperatures.

Canola seeds, imbibed in 100 μM ABA at 2°C for 15 days, tolerated −45°C. If the seeds imbibed in 100 μM ABA were held at 23°C for 1 day in the presence of ABA, the LT50 of the imbibed seeds increased to −25°C. After 3 days at 23°C in the presence of ABA, the imbibed seeds still did not complete germination; however, the LT50 of the seeds increased to −3°C (Table 1). It is speculated the loss in freezing tolerance was in part due to the uptake of water, reorganization of cells, and loss in cryoprotective compounds. These results suggest the loss of cryoprotective compounds is partially independent of radicle protrusion.

Duration at −6°C on survival of imbibed canola seeds (LD50)

Canola seeds imbibed in either water or 100 μM ABA at 23°C, 8°C, or 2°C lost viability during storage at −6°C. Canola seeds imbibed in water at 23°C for 4 h had 95% germination after 10 days of storage at −6°C (Fig. 3A). However, germination declined rapidly in succeeding days such that after 20 days, only 15% of the water-imbibed seeds completed germination (similar trends were observed for seeds imbibed for more than 4 h at 23°C). Seeds imbibed in ABA at 23°C for up to 24 h tolerated −6°C for 40 days with only a small loss in germination (Fig. 3A). Approximately 40% of the seeds imbibed in ABA for 48 h at 23°C completed germination after 20 days of storage at −6°C with no further decline observed up to 40 days.


Figure 3. The effect of abscisic acid (ABA) on the survival of imbibed canola (cv. Quest) seeds at −6°C for 0–90 days. Seeds were imbibed in either water or 100 μM ABA at 23°C for 0–24 h (A), 8°C for 0–8 days (B), 2°C for 0–15 days (C), before transferring to −6°C. Germination was conducted at 23°C for 7 days to determine the LD50.

Download figure to PowerPoint

In contrast to 23°C, seeds imbibed at 8°C in water tolerated longer periods at −6°C (Fig. 3B). For example, after 40 days at −6°C, seeds imbibed for 1 and 3 days at 8°C were 40 and 20% germinable, respectively. However, seeds imbibed for 6 or 8 days rapidly succumbed to the −6°C conditions. In comparison, seeds imbibed at 8°C in ABA for up to 8 days had a high germination percentage after 40 days of storage at −6°C.

Seeds imbibed at 2°C tolerated −6°C, the longest compared with warmer temperatures (Fig. 3C). Seeds imbibed at 2°C in water longer than 6 days had little freezing tolerance after 10 days at −6°C, whereas seeds imbibed in ABA for up to 15 days were highly germinable following storage at −6°C for 40 days (Fig. 3C). In addition, this study demonstrates a dramatic difference in survival of seeds as determined by the LT50 procedure which employs relatively fast freezing rates (2°C or faster) compared with those subjected to long-term subzero temperatures. For example, seeds imbibed at 2°C for 7 days had an LT50 of −30°C; however, seeds imbibed at 2°C for 6 days were not germinable after 10 days at constant −6°C. This demonstrates the difference in freezing tolerance between a short exposure to very low temperatures vs a prolonged exposure to a relatively warm subzero temperature.

Kirkland and Johnson (2000) reported canola, seeded in late October into standing stubble, had a higher survival rate compared with seeds sown into summerfallow. Soil temperatures are higher in standing stubble, which traps snow and effectively insulates the soil. The lower the soil temperature the higher the degree of desiccation potential, as calculated from the Clausius–Clapeyron equation (Maron and Prutton 1962). These results indicate that seeds can tolerate short periods of extreme desiccating conditions, although not long periods of moderate desiccation. Gusta et al. (1997) reported a similar phenomenon for winter cereals. Cold-acclimated hardy winter cereals tolerated −24°C when cooled at 2°C h−1, but succumbed after 11 days at a constant −12°C but tolerated 6 months at −3°C (Gusta and Fowler 1977). These results suggest that different mechanisms of injury occur under relatively fast rates of freezing compared with storage at a constant temperature and have demonstrated lethality of cereal crowns following prolonged storage at subzero temperatures was due to reactive oxygen species.

High-temperature desiccation of imbibed canola seed

Seeds imbibed in either water or ABA for 2 days at 8°C and then dried at 60°C for 2 days had a germination rate of 95% in subsequent germination tests at 23°C. However, after 3 days at 8°C, germination at 23°C was 88% for both water- and ABA-imbibed seeds (Table 2). These results were unexpected because 78% of the seeds imbibed in water had germinated prior to drying. It is generally assumed that once the radicle emerges, seeds lose their desiccation tolerance (Bradford 1995). By day 4 at 8°C, 95% of the water-imbibed and 50% of the ABA-imbibed seeds completed germination prior to the drying treatment, but only 10% of the seeds were germinable at 23°C.

Table 2.  The effect of duration of imbibition and temperature on germination of canola (cv. Quest) seed imbibed in water or 100 M abscisic acid (ABA), before and after drying at 60°C. Values in parenthesis indicate standard deviation. a Expressed on a dry weight basis. b Final germination was conducted at 23°C for 7 days. Seeds with the radicle emerged were considered as germinated.
   Germination (%)b
Days of imbibitionTreatmentSeed moisturea (%)Before dryingAfter drying at 60°C
At 8°C
 1Water36.8 (±0.7)095.0 (±2.0)
 1ABA37.0 (±0.7)095.0 (±2.0)
 2Water41.0 (±0.8)095.0 (±2.0)
 2ABA39.3 (±1.0)095.0 (±2.0)
 3Water40.5 (±0.1)78.0 (±5.0)88.0 (±4.0)
 3ABA40.8 (±0.2)088.0 (±2.0)
 4Water48.3 (±0.6)95.0 (±2.0)10.0 (±5.0)
 4ABA44.7 (±0.6)50.0 (±5.0)10.0 (±5.0)
At 2°C
 6Water41.4 (±1.2)095.0 (±2.0)
 6ABA41.5 (±1.5)095.0 (±2.0)
 9Water45.0 (±0.5)30.0 (±6.0)88.0 (±4.0)
 9ABA41.9 (±0.8)095.0 (±2.0)
 12Water50.2 (±2.0)90.0 (±4.0)45.0 (±3.0)
 12ABA43.6 (±0.7)095.0 (±2.0)
 15Water56.6 (±0.5)90.0 (±4.0)25.0 (±6.0)
 15ABA45.1 (±0.5)042.0 (±5.0)

In comparison with 8°C, seeds imbibed at 2°C were desiccation tolerant for a longer time following drying at 60°C for 2 days. For example, at 2°C seeds could be imbibed for up to 6 days without the 60°C drying treatment affecting subsequent germination at 23°C. By day 9 at 2°C, 30% of the water-imbibed seeds had completely germinated vs 0% of the ABA-treated seeds. However, 88 and 95% of the water-imbibed and ABA-imbibed seed, respectively, were still germinable following drying at 60°C. By day 12 at 2°C, 90% of the water-imbibed seeds germinated; however, only 45% of the seeds completed germination following drying at 60°C, whereas on day 15, only 25% of the dried seeds completed germination. Seeds imbibed in ABA for 12 days were 95% germinable following drying at 60°C; however, after 15 days, only 42% of the seeds completed germination following drying. Responses to freezing injury and drying at 60°C were similar; imbibition temperature and imbibition time and ABA all affected the stress tolerance of the seeds. Results suggest that ABA maintains freezing tolerance and desiccation tolerance for extended periods. Freezing is considered to be a form of desiccation because liquid water is converted to ice (Burke et al. 1976). These results suggest a strong relationship between cold- and warm-induced desiccation injury.

During seed maturation, a group of late embryogenesis abundant (LEA) proteins are produced (Baker et al. 1988). Also, these proteins are induced when immature seeds are treated with ABA (Galau et al. 1986) or subjected to desiccation (Hughes and Galau 1989). LEA proteins also include several of the dehydrins (Robertson and Chandler 1994) believed to enhance cell survival during desiccation (Bray 1993). Possible protective functions of dehydrins during periods of either freezing or desiccation include osmoregulation (Close 1996), neutralization of ions (Dure 1993), protection of DNA (Chiatante and Onelli 1993), and protection of membranes and proteins (Ingram and Bartels 1996). Egerton-Warburton et al. (1997) could not detect dehydrins in imbibed seeds of Zea mays after 30 h in water; however, high levels of dehydrin were still present after 40 h in 0.1 μM ABA. In addition, dehydrins increase during the cold acclimation of plants (Robertson et al. 1994) and when plants are treated with exogenous ABA (Close 1996).

The results in this study indicate that the rate of loss of these protective compounds in imbibed seeds is a function of temperature after a critical moisture content is achieved. At temperatures approaching 2°C, ABA is effective in preventing the loss of these protective compounds. In nature, freezing-induced desiccation is the major cause of lethality in overwintering imbibed seeds and winter cereals (Gusta et al. 1997). During imbibition of seeds at warm temperatures, protective compounds are rapidly degraded, resulting in a loss of freezing-induced desiccation tolerance. Seeds imbibed in water at 2°C for 7 days tolerated −30°C as determined by the LT50 test. In contrast, similarly treated seed were not germinable following storage at −6°C for 10 days. Gusta et al. (1997) demonstrated a similar phenomenon for winter wheat. Fully cold-hardened winter wheat seedlings could tolerate −24°C when cooled at 2°C h−1 but could not tolerate a constant −15°C for 2 weeks.

In summary, in order for imbibed seeds to tolerate the desiccating conditions of freezing, they must preserve the protective compounds synthesized during seed maturation. Results suggest that during imbibition the loss of these protective compounds is regulated in part by ABA at temperatures approaching 0°C. This may be due to a slower rate of degradation of ABA at temperatures approaching 0°C, or the ABA-binding sites are stronger at the lower temperatures. Studies on the freezing tolerance of imbibed seeds provide a simple model to study the freezing tolerance of more fully developed plants.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  • Andrews JE (1958) Controlled low temperature tests of sprouted seeds as a measure of cold hardiness of winter wheat varieties. Can J Plant Sci 38: 17
  • Bai Y, Booth DT, Romo JT (1998) Winter fat (Eurotia lanata (Pursh.) Mog.) seedbed ecology: low temperature exotherms and cold hardiness in hydrated seeds as influenced by imbibition temperature. Ann Bot 81: 595602
  • Baker J, Steele C, Dure L III (1988) Sequence and characterization of 6 LEA proteins and their genes from cotton. Plant Mol Biol 11: 277291
  • Becwar MR, Stanwood PC, Leonhardt KW (1983) Dehydration effects on freezing characteristics and survival in liquid nitrogen of desiccation-tolerant and desiccation-sensitive seeds. J Am Soc Hort Sci 108: 613618
  • Bewley JD (1997) Seed germination and dormancy. Plant Cell 9: 10551066
  • Bradford KJ (1995) Water relations in seed germination. In: KigelJ, GaliliG (eds) Seed Development and Germination. Marcel Dekker, Inc., New York, NY/USA, pp 351396
  • Bray EA (1993) Molecular responses to water deficit. Plant Physiol 103: 10351040
  • Burke MJ, Gusta LV, Quamme HA, Weiser CJ, Li PH (1976) Freezing injury in plants. Ann Rev Plant Physiol 27: 507528
  • Chiatante D, Onelli E (1993) Nuclear proteins and the onset of cell proliferation in root meristems of Pisum sativum: QP47 a novel acidic protein. Seed Sci Res 3: 3541
  • Chrispeels MJ, Maurel C (1994) Aquaporins: the molecular basis of facilitated water movement through living plant cells? Plant Physiol 105: 913
  • Close TJ (1996) Dehydrins: emergence of a biochemical role of a family of plant dehydration proteins. Physiol Plant 97: 795803
  • Cremer KW, Mucha SB (1985) Effects of freezing temperatures on mortality of air-dry, imbibed and germinating seeds of Eucalyptus and Radiata pine. Aust For Res 15: 243251
  • Dure L III (1993) A repeating 11-mer amino acid motif and plant desiccation. Plant J 3: 363369
  • Egerton-Warburton LM, Balsamo RA, Close TJ (1997) Temporal accumulation and ultrastructural localization of dehydrins in Zea mays. Physiol Plant 101: 545555
  • Galau GA, Hughes DW, Dure L III (1986) Abscisic acid induction of cloned cotton late embryogenesis abundant (Lea mRNA's). Plant Mol Biol 7: 155170
  • Gao YP, Young L, Bonham-Smith P, Gusta LV (1999) Characterization and expression in primed seed of Brassica napus during germination under stress conditions. Plant Mol Biol 40: 635644
  • Gazeau C, Dereuddre J (1980) Effects de refroidissement prolonges sur la teneur en eau et la resistance au froid des tissues de la Troene (Ligustrum vulgare L. Oleaceae). Compte Rendu l'Academie Des Sci, Paris, Series D 290: 14431446
  • Gusta LV, Fowler DB (1977) Factors affecting the old survival of winter cereals. Can J Plant Sci 57: 213219
  • Gusta LV, Burke MJ, Kapoor AC (1975) Determination of unfrozen water in winter cereals at sub-freezing temperatures. Plant Physiol 56: 707709
  • Gusta LV, O'Connor BJ, MacHutcheon MG (1997) The selection of superior winter-hardy genotypes using a prolonged freeze test. Can J Plant Sci 77: 1521
  • Gusta LV, Tyler NJ, Chen TH-H (1983) Deep undercooling in woody taxa growing north of the −40°C isotherm. Plant Physiol 72: 122128
  • Hughes DW, Galau GA (1989) Temporally modular gene expression during cotyledon development. Genes Dev 3: 358369
  • Ingram J, Bartels D (1996) The molecular basis of dehydration tolerance in plants. Annu Rev Plant Physiol Plant Mol Biol 47: 377403
  • Ishikawa M, Sakai A (1982) Characteristics of freezing avoidance in comparison with freezing tolerance: a demonstration of extra organ freezing. In: LiPH, SakaiA (eds) Plant Cold Hardiness and Freezing Stress, Vol. 2. Academic Press, New York, NY/USA, pp 325340
  • Juntilla O, Stushnoff C (1977) Freezing avoidance by deep supercooling of hydrated lettuce seeds. Nature 269: 325327
  • Keefe PD, Moore KG (1980) Freeze tolerance in hydrated Latuca sativa L. seed: a model to explain observed variation between seed lots. Ann Bot 51: 373383
  • Keefe PD, Moore KG (1981) Freeze desiccation: a second mechanism for the survival of hydrated lettuce (Lactuca sativa L.) seed at sub-zero temperatures. Ann Bot 47: 635645
  • Kirkland KJ, Johnson EN (2000) Alternative seeding dates (fall and April) affect Brassica napus canola yield and quality. Can J Plant Sci 80: 713719
  • Maron SH, Prutton CF (1962) Principles of Physical Chemistry. The MacMillan Co, New York, NY/USA pp 92–95.
  • Robertson M, Chandler PM (1994) A dehydrin cognate protein from pea (Pisum sativum L.) with an atypical pattern of expression. Plant Mol Biol 26: 805816
  • Robertson AJ, Weninger A, Wilen RW, Fu P, Gusta LV (1994) Comparison of dehydrin gene expression and freezing tolerance in Bromus inermis and Secale cereale grown in controlled environments, hydroponics and the field. Plant Physiol 106: 12131216