Planta (1985)166:32t-328
P l a n t a 9 Springer-Verlag1985
The control of seed germination in Trollius ledebouri A model of seed dormancy A. Hepher* and J.A. Roberts Department of Physiology and Environmental Science, University of Nottingham, School of Agriculture, Sutton Bonington, Loughborough, Leics, LE12 5RD, UK
Abstract. Treatment of Trollius ledebouri seeds with gibberellins A 4 + A 7 promotes germination. The efficacy of the treatment is dependent upon the duration of imbibition in distilled water prior to GA4+ 7 application. 'Presoaking' increases both the final percentage germination attained and also its rate of achievement. No presoaking effect is exhibited by seeds induced to germinate by testa removal in the absence of GA4+ 7. Active washing of Trollius seeds enhances the presoaking effect and the eluent from washed seeds is inhibitory to germination. The results support the hypothesis that the presoaking effect exhibited by Trollius is the result of the leaching of a germination inhibitor from the seeds which is antagonistic to GA4+ 7. Additionally, treatment of Trollius seeds with the gibberellin-biosynthesis inhibitor (2-chloroethyl)trimethylammonium chloride (CCC) prior to testa removal retards germination. The inhibitory effect of CCC on germination is overcome by GA4+ 7. Although CCC inhibits embryo growth during the presoaking of intact seeds, it does not affect the increased sensitivity of presoaked seeds to GA4+ v. Therefore, although endogenous gibberellins may be involved in the germination process, they do not contribute to the presoaking phenomenon. The expansion of isolated endosperm tissue is not affected by CCC. However, the chemical markedly inhibits endosperm expansion in intact seeds and implicates the embryo as both the site of production of the germination inhibitor and of gibberellin. These results are discussed in relation to previous studies and a model is presented to account for the characteristics of germination in Trollius.
Key words: Dormancy (seed) - Endosperm breakdown - Germination (seed) - Gibberellin and seed dormancy - Trollius.
Introduction
A dormant seed of Trollius consists of a small, rudimentary embryo situated at the micropylar end of the seed, surrounded by a living endosperm containing lipid and many protein bodies. Dormancy may be broken by testa removal or treatment with gibberellin (GA; Hepher and Roberts 1985). Germination is characterised by the swelling of the endosperm which causes the testa to rupture. The breakdown products of the food reserve are then consumed by the rapidly developing embryo, the radicle of which eventually penetrates the exposed endosperm. This process continues until the testa is shed from the cotyledons. In a previous paper it was demonstrated that the changes taking place in isolated endosperm tissue were comparable to those observed during germination of intact seeds (Hepher and Roberts 1985). It was also proposed that an inhibitor may be involved in the regulation of dormancy. In this paper further evidence is presented to support the hypothesis that an inhibitor, antagonistic to the promotive effects of exogenous GA, regulates germination in Trollius. Finally the role of endogenous GA in germination is investigated and a model presented to account for the characteristics of the dormancy mechanism in Trollius.
* Present
Material and methods
GA = gibberellin; CCC = (2-chloroethyl)trimethylammonium chloride
The germination procedure, method of GA application and statistical analysis were as described previously (Hepher and Roberts 1985).
address: Department of Botany, University of Durham, Durham DH1 3LE, UK
Abbreviations:
A. Hepher and J.A. Roberts: Control of seed germination in Trollius
322
Uptake of [3H]GA4. Approximately 74 kBq [3H]GA4 (specific activity 2MBq gM -1) dissolved in 10gl 100% methanol (obtained from Dr. V. Sponsel, Department of Chemistry, University of Bristol, UK) was made up to 500 gl with distilled water and divided into five 100-gl aliquots. Sixty seeds were imbibed in distilled water for 1 or 16 d. Testas were then removed from half the seeds in each treatment giving four lots of 30 seeds. Each treatment was placed into a micro test-tube containing 100 ~1 of the [3H]GA4 for 24 h at 25 ~C. After this period, each treatment was thoroughly washed in 100% methanol and divided into six replicates of five seeds. After removal and retention of the remaining testas each replicate was crushed in 200 gl Solvene-100 (Packard Instruments, Caversham, Berks., UK) and incubated overnight at 32~ before addition of 5ml scintillant (Dimilume-30, Packard Instruments). Both embryo-endosperm and excised testas were counted separately and results compared with previously unimbibed seeds incubated in the radioactive gibberellin under identical conditions. Active-washing procedure. The presence of an inhibitor was investigated by actively washing the seeds. Two grams of Trollius seed (approx. 1600 seeds) were immersed in 12.5 ml distilled water contained in a 20-ml glass tube supported on a sintered-glass base. Using a Dymax Mark 1 variable-speed pump (Charles Austin, Byfleet, Surrey, UK), air was passed through the glass base and served not only to aerate the mixture but also to agitate vigorously the imbibing seed in a cyclical manner. The medium was changed daily by switching off the pump and allowing it to drain into a Buchner flask below; it was retained for further study. One hundred and fifty seeds were removed after 1 and 2 soaking and treated with 10 4 M GA4+ 7 for 24 h. The subsequent germination was compared with controls soaked on moistened filter paper for equivalent times.
Inhibitor assay. The biological activity of the eluent obtained from the active-washing procedure was tested in a germination assay. After using a 0.45-gm filter (Millipore Corp., Bedford, Mass., USA), to remove seed debris as well as fungal and bacterial contaminants, 5 ml of the eluent was added to a sheet of Whatman 182 filter paper (Whatman, Springfield Mill. Kent, UK) in a Petri dish on which were placed four replicates of 10 seeds whose testas had been removed after 1 d imbibition. Subsequent germination performance, as assessed by radicle
-
-
---'-4
%
80
o
~ so
protrusion, was compared with a similar number of controls in distilled water.
Treatment with (2-chloroethyl)trimethylammonium chloride (CCC). The role of endogenous gibberellin was investigated using the inhibitor of gibberellin biosynthesis CCC (Column Chemicals, London, UK) as a 40% (w/v) aqueous solution.
Results
Unimbibed seeds were given a 24-h treatment with 10 -4, 10 .5 o r 10 6 M
G A 4 + 7. N o
stimulation
of
germination was observed after treatment with 10 6 M G A 4 + 7 , while plateaus of approx. 5% and 40% were recorded for 10 5 and 10-4 M G A 4 + 7 , respectively. This result was in contrast to that which had been obtained previously where 10 4 M G A 4 + 7 elevated germination in a similar experiment to 65% (Hepher and Roberts 1985). The only difference between the two treatments was that the seeds showing the greater response to applied GA4+ 7 had been imbibed for 24 h in water prior to gibberellin treatment. Consequently it was considered that this presoaking might have affected the subsequent response to exogenous GA. This was investigated by presoaking seeds on filter paper moistened with distilled water for 1, 4 or 14 d before a 24-h exposure to 10-4 M G A 4 + 7. Germination was assessed by both testa rupture (Fig. 1 a) and radicle protrusion (Fig. 1 b) and compared with controls unimbibed before gibberellin treatment . The results of Fig. 1 clearly demonstrate that presoaking not only increased the final percentage attained but also its rate of achievement. This effect of presoaking persisted if seeds were imbibed for similar periods and airdried to constant weight prior to gibberellin treatment. After 20 d presoaking no further
0
~N 10 e
O0
2
z,
6
8
10
12
DAYS AFTER GA TREATMENT
2
8 10 12 DAYS AFTER GA TREATMENT
Fig. 1 a, b. Germination of Trollius seeds as quantified by a testa rupture or b radicle protrusion after treatment with 10-4 M GA4+ 7. Seeds were imbibed in distilled water for 0, 1, 4 or 14 d prior to gibberellin treatment
A. Hepher and J.A. Roberts: Control of seed germination in Trollius
323
b :
:
4
.o
o
/
0.~ c~
a --
§ GA
/
..... c~
~= 0.3(
/
J
o Id PRESOAK / 9 10dP R E S O A K / /
/ /
u
5 el:
3(
"/ .o UO
1 2 3 DAYS AFTER TESTA REMOVAL
+GA
. . . . OA o 1d PRESOAK 9 10d PRESOAK
..';/ 9, :'.,'
0.1(
i
--
T#
,
P l a n t a 9 Springer-Verlag1985
The control of seed germination in Trollius ledebouri A model of seed dormancy A. Hepher* and J.A. Roberts Department of Physiology and Environmental Science, University of Nottingham, School of Agriculture, Sutton Bonington, Loughborough, Leics, LE12 5RD, UK
Abstract. Treatment of Trollius ledebouri seeds with gibberellins A 4 + A 7 promotes germination. The efficacy of the treatment is dependent upon the duration of imbibition in distilled water prior to GA4+ 7 application. 'Presoaking' increases both the final percentage germination attained and also its rate of achievement. No presoaking effect is exhibited by seeds induced to germinate by testa removal in the absence of GA4+ 7. Active washing of Trollius seeds enhances the presoaking effect and the eluent from washed seeds is inhibitory to germination. The results support the hypothesis that the presoaking effect exhibited by Trollius is the result of the leaching of a germination inhibitor from the seeds which is antagonistic to GA4+ 7. Additionally, treatment of Trollius seeds with the gibberellin-biosynthesis inhibitor (2-chloroethyl)trimethylammonium chloride (CCC) prior to testa removal retards germination. The inhibitory effect of CCC on germination is overcome by GA4+ 7. Although CCC inhibits embryo growth during the presoaking of intact seeds, it does not affect the increased sensitivity of presoaked seeds to GA4+ v. Therefore, although endogenous gibberellins may be involved in the germination process, they do not contribute to the presoaking phenomenon. The expansion of isolated endosperm tissue is not affected by CCC. However, the chemical markedly inhibits endosperm expansion in intact seeds and implicates the embryo as both the site of production of the germination inhibitor and of gibberellin. These results are discussed in relation to previous studies and a model is presented to account for the characteristics of germination in Trollius.
Key words: Dormancy (seed) - Endosperm breakdown - Germination (seed) - Gibberellin and seed dormancy - Trollius.
Introduction
A dormant seed of Trollius consists of a small, rudimentary embryo situated at the micropylar end of the seed, surrounded by a living endosperm containing lipid and many protein bodies. Dormancy may be broken by testa removal or treatment with gibberellin (GA; Hepher and Roberts 1985). Germination is characterised by the swelling of the endosperm which causes the testa to rupture. The breakdown products of the food reserve are then consumed by the rapidly developing embryo, the radicle of which eventually penetrates the exposed endosperm. This process continues until the testa is shed from the cotyledons. In a previous paper it was demonstrated that the changes taking place in isolated endosperm tissue were comparable to those observed during germination of intact seeds (Hepher and Roberts 1985). It was also proposed that an inhibitor may be involved in the regulation of dormancy. In this paper further evidence is presented to support the hypothesis that an inhibitor, antagonistic to the promotive effects of exogenous GA, regulates germination in Trollius. Finally the role of endogenous GA in germination is investigated and a model presented to account for the characteristics of the dormancy mechanism in Trollius.
* Present
Material and methods
GA = gibberellin; CCC = (2-chloroethyl)trimethylammonium chloride
The germination procedure, method of GA application and statistical analysis were as described previously (Hepher and Roberts 1985).
address: Department of Botany, University of Durham, Durham DH1 3LE, UK
Abbreviations:
A. Hepher and J.A. Roberts: Control of seed germination in Trollius
322
Uptake of [3H]GA4. Approximately 74 kBq [3H]GA4 (specific activity 2MBq gM -1) dissolved in 10gl 100% methanol (obtained from Dr. V. Sponsel, Department of Chemistry, University of Bristol, UK) was made up to 500 gl with distilled water and divided into five 100-gl aliquots. Sixty seeds were imbibed in distilled water for 1 or 16 d. Testas were then removed from half the seeds in each treatment giving four lots of 30 seeds. Each treatment was placed into a micro test-tube containing 100 ~1 of the [3H]GA4 for 24 h at 25 ~C. After this period, each treatment was thoroughly washed in 100% methanol and divided into six replicates of five seeds. After removal and retention of the remaining testas each replicate was crushed in 200 gl Solvene-100 (Packard Instruments, Caversham, Berks., UK) and incubated overnight at 32~ before addition of 5ml scintillant (Dimilume-30, Packard Instruments). Both embryo-endosperm and excised testas were counted separately and results compared with previously unimbibed seeds incubated in the radioactive gibberellin under identical conditions. Active-washing procedure. The presence of an inhibitor was investigated by actively washing the seeds. Two grams of Trollius seed (approx. 1600 seeds) were immersed in 12.5 ml distilled water contained in a 20-ml glass tube supported on a sintered-glass base. Using a Dymax Mark 1 variable-speed pump (Charles Austin, Byfleet, Surrey, UK), air was passed through the glass base and served not only to aerate the mixture but also to agitate vigorously the imbibing seed in a cyclical manner. The medium was changed daily by switching off the pump and allowing it to drain into a Buchner flask below; it was retained for further study. One hundred and fifty seeds were removed after 1 and 2 soaking and treated with 10 4 M GA4+ 7 for 24 h. The subsequent germination was compared with controls soaked on moistened filter paper for equivalent times.
Inhibitor assay. The biological activity of the eluent obtained from the active-washing procedure was tested in a germination assay. After using a 0.45-gm filter (Millipore Corp., Bedford, Mass., USA), to remove seed debris as well as fungal and bacterial contaminants, 5 ml of the eluent was added to a sheet of Whatman 182 filter paper (Whatman, Springfield Mill. Kent, UK) in a Petri dish on which were placed four replicates of 10 seeds whose testas had been removed after 1 d imbibition. Subsequent germination performance, as assessed by radicle
-
-
---'-4
%
80
o
~ so
protrusion, was compared with a similar number of controls in distilled water.
Treatment with (2-chloroethyl)trimethylammonium chloride (CCC). The role of endogenous gibberellin was investigated using the inhibitor of gibberellin biosynthesis CCC (Column Chemicals, London, UK) as a 40% (w/v) aqueous solution.
Results
Unimbibed seeds were given a 24-h treatment with 10 -4, 10 .5 o r 10 6 M
G A 4 + 7. N o
stimulation
of
germination was observed after treatment with 10 6 M G A 4 + 7 , while plateaus of approx. 5% and 40% were recorded for 10 5 and 10-4 M G A 4 + 7 , respectively. This result was in contrast to that which had been obtained previously where 10 4 M G A 4 + 7 elevated germination in a similar experiment to 65% (Hepher and Roberts 1985). The only difference between the two treatments was that the seeds showing the greater response to applied GA4+ 7 had been imbibed for 24 h in water prior to gibberellin treatment. Consequently it was considered that this presoaking might have affected the subsequent response to exogenous GA. This was investigated by presoaking seeds on filter paper moistened with distilled water for 1, 4 or 14 d before a 24-h exposure to 10-4 M G A 4 + 7. Germination was assessed by both testa rupture (Fig. 1 a) and radicle protrusion (Fig. 1 b) and compared with controls unimbibed before gibberellin treatment . The results of Fig. 1 clearly demonstrate that presoaking not only increased the final percentage attained but also its rate of achievement. This effect of presoaking persisted if seeds were imbibed for similar periods and airdried to constant weight prior to gibberellin treatment. After 20 d presoaking no further
0
~N 10 e
O0
2
z,
6
8
10
12
DAYS AFTER GA TREATMENT
2
8 10 12 DAYS AFTER GA TREATMENT
Fig. 1 a, b. Germination of Trollius seeds as quantified by a testa rupture or b radicle protrusion after treatment with 10-4 M GA4+ 7. Seeds were imbibed in distilled water for 0, 1, 4 or 14 d prior to gibberellin treatment
A. Hepher and J.A. Roberts: Control of seed germination in Trollius
323
b :
:
4
.o
o
/
0.~ c~
a --
§ GA
/
..... c~
~= 0.3(
/
J
o Id PRESOAK / 9 10dP R E S O A K / /
/ /
u
5 el:
3(
"/ .o UO
1 2 3 DAYS AFTER TESTA REMOVAL
+GA
. . . . OA o 1d PRESOAK 9 10d PRESOAK
..';/ 9, :'.,'
0.1(
i
--
T#
,
