Tải bản đầy đủ - 0 (trang)
IV. Regulation of Seed-Fill Duration

IV. Regulation of Seed-Fill Duration

Tải bản đầy đủ - 0trang



to the question—why does the seed stop growing?—can also be found at

both levels.

Many of the characteristics of seed growth, including regulation of seed

growth rate are consistent across crop species (Egli, 1998), in spite of variation

in seed morphology, size, shape, color and composition. Consequently, a single

set of mechanisms may also regulate the termination of seed growth of all

grain crop species.


Seed filling continues only as long as the mother plant supplies raw materials

to the seed and the seed has the ability to convert the raw materials into storage

compounds. If the regulatory mechanism that stops seed growth resides in the

seed, the termination of seed filling may not depend on the absence of raw

materials. Temporal measurements of canopy photosynthesis in soybean (Christy

and Porter, 1982) and maize (Pearson et al., 1984) suggest that seeds matured

when photosynthesis was still 10 – 20% of its maximum rate. Flag leaf

photosynthesis was 80% of its highest level when wheat kernels reached their

maximum dry weight (Sofield et al., 1977) and seeds on partially depodded

soybean plants matured when the plants were still green and photosynthetically

active (Munier-Jolain et al., 1996). Soybean seeds that had their growth limited

by physical restriction matured when leaves were still photosynthetically active

with high levels of Rubisco and chlorophyll (Crafts-Brandner, 1995, personal

communication) and mature pods can be found on soybean plants with green

leaves (Egli, 1998). Banziger et al. (1994) reviewed several reports of wheat

seeds maturing when plants still had green parts and the stem reserves were not

exhausted. Stay-green maize genotypes had high carbohydrate levels in the stem

when the seeds matured (Gentinetta et al., 1986). Death of grain sorghum plants

occurred after the seed reached PM (Rajewski and Francis, 1991). Pods and seeds

matured on pigeon pea [Cajanus cajan (L.) Millsp.], cowpea [Vigna unguiculata

(L.) Walp.] and rice plants that still had the capacity to produce a ratoon crop or a

second crop of pods (Bahar and De Datta, 1977; Sharma et al., 1978; Gwathmey

et al., 1992; Turner and Jund, 1993). Sucrose concentrations in wheat and

soybean seeds were near maximum levels at PM suggesting that assimilate

availability was not limiting when seed growth stopped (Jenner and Rathjen,

1975; Egli and Bruening, 2001).

These results from many crop species suggest that completion of leaf

senescence and death of the vegetative plant are not an absolute prerequisite for

termination of seed growth. Seeds cannot continue to accumulate dry matter

without a source of raw materials, but having a source does not guarantee that

growth will continue.



The termination of seed growth when assimilate is still available requires a

regulatory mechanism in the seed. Changes in tissue water status may play a

regulatory role in plant development (Walbot, 1978; Adams and Rinne, 1980;

McIntyre, 1987), so the widely observed desiccation of seed tissues that occurs

as the seed approaches PM (Westgate, 1994; Egli and TeKrony, 1997; Saini and

Westgate, 2000) may provide a mechanism. Desiccation occurs when cell

expansion and water movement into the seed stops but dry matter accumulation

continues (Ray, 1987; Egli, 1990, 1998). Cell expansion could be limited by

physical restriction by seed or fruit structures (Boshankian, 1918; Murata and

Matsushima, 1975; Scott et al., 1983), thus, the characteristics of the fruit or seed

(i.e., its maximum potential volume) would play a role in triggering the

termination of seed growth. The assimilate supply must also regulate cell

expansion and seed desiccation since it can affect seed-fill duration (Rajewski

and Francis, 1991; Andrade and Ferreiro, 1996; Munier-Jolain et al., 1996, 1998;

Egli and Bruening, 2001).


Since a seed cannot grow for long without a photosynthetically active plant

canopy, the progression of leaf senescence and seed-fill duration tends to change

in concert. Canopy photosynthesis begins to decline relatively early in the seedfilling period in most crops (Larson et al., 1981; Wells et al., 1982; Pearson et al.,

1984; Hall et al., 1990; Gent, 1995) as leaf senescence reduces the photosynthetic

capacity of the plant canopy. It seems ironic that soon after the preliminarily

events of the yield production process (development of the canopy, flowering and

seed set) are completed and the plant finally begins to accumulate dry weight in

the seeds—to actually produce yield—the productive capacity of the plant begins

to decline as the photosynthetic machinery is destroyed and nitrogen is exported

from the leaf.

Manipulation of leaf senescence frequently results in corresponding changes

in seed-fill duration. Complete defoliation to artificially reduce photosynthesis to

near zero results in a rapid cessation of seed growth (Jones and Simmons, 1983;

Hunter et al., 1991; Rajewski and Francis, 1991; Vieira et al., 1992). Inadequate

N supplies (Boon-Long et al., 1983a; Egli et al., 1985; Hayati et al., 1995) and

water stress (Gallagher et al., 1976; NeSmith and Ritchie, 1992; de Souza et al.,

1997; Davies et al., 1999) accelerated leaf senescence and shortened the seedfilling period.

Variation in leaf senescence patterns is often associated with genetic

differences in seed-fill duration. Senescence progressed faster in a maize hybrid

with a short EFP than in one with a long EFP (Crafts-Brandner and Poneleit,

1987). Hartung et al. (1989) selected for long and short seed-filling periods in a

recurrent selection program with maize and changed leaf senescence, as indicated



by photosynthesis and leaf N levels during seed filling, in concert with changes in

seed-fill duration (Crafts-Brandner and Poneleit, 1992). Longer seed-filling

periods in modern soybean cultivars were associated with a slower decline in

canopy photosynthesis, a good indicator of senescence (Wells et al., 1982). Leaf

N concentration declined faster in a soybean genotype with an exceptionally

short seed-fill duration (Egli et al., 1987b), but differences among genotypes

disappeared when leaf N concentration was related to the stage of seed

development. Cultivars with delayed leaf senescence (stay-green types)

frequently have higher yields (Tollenaar, 1991; Ma and Dwyer, 1998; Duvick

and Cassman, 1999) which are probably a result of longer seed-filling periods

(Frederick et al., 1989).

Delayed senescence will increase seed-fill duration only when the seed has the

ability to continue growth and increase in size. The seed, in some situations,

responds to treatments that simulated delayed senescence by increasing

assimilate supply per seed (depodding, degraining, increasing photosynthesis

after seed number is fixed) by increasing seed size in soybean (Egli et al., 1985;

Egli and Bruening, 2001), wheat (Fischer and Hille Ris Lambers, 1978; Winzeler

et al., 1989), sorghum (Kiniry, 1988), sunflower (Steer et al., 1988; Charlet and

Miller, 1993) and maize (Kiniry et al., 1990). Seed and fruit characteristics

apparently limited seed size increases in some instances in maize (Jones and

Simmons, 1983; Kiniry et al., 1990), barley (Scott et al., 1983; Dreccer et al.,

1997), rice (Murata and Matsushima, 1975; Kato, 1999), wheat (Millet, 1986; Ma

et al., 1995) and sunflower (Charlet and Miller, 1993). Developing soybean seeds

spilt pods in environments that resulted in abnormally small pods (Egli, 1998), an

example of potential physical restriction of seed growth when assimilate was

available. Simply delaying senescence, therefore, may lead to a longer seed-fill

duration and higher yield in some crops or some environments, but in other

situations the characteristics of the seed may also have to be changed.

The characteristics of the seed and leaf senescence are clearly intertwined as

seed filling must stop when the plant can no longer supply assimilate or when the

seed can no longer accumulate dry matter. Leaf senescence, the process that ends

the life of all annual grain crops, sometimes in spectacular fashion, has probably

received more attention than the characteristics of the seed, but either can limit

seed-fill duration and yield.


Leaf senescence comprises a series of events that result in cellular disassembly

in the leaf and mobilization of the materials released (Thomas and Stoddart,

1980). Organelles and membranes maintain their integrity such that cellular

compartmentation is maintained throughout the senescence process (Thomson

and Platt-Aloia, 1987).



Leaf senescence normally occurs during seed filling in all grain crops and the

nitrogen released is exported to the developing seed (Wittenbach, 1979;

Crafts-Brandner and Egli, 1987; Crafts-Brandner and Poneleit, 1987). Remobilized N can be a significant source of N for the seed, with estimates of its

contribution to the total seed N at maturity ranging from 11 to 100% in soybean

(Egli et al., 1978, 1983; Israel, 1981; McBlain and Hume, 1981; Zeiher et al.,

1982), 63 to 100% in wheat (Heitholt et al., 1990; Mi et al., 2000), 49 to 64% in

sorghum (Borell and Hammer, 2000) and 41 to 69% in maize (Below et al., 1981;

Rajcan and Tollenaar, 1999), while one cowpea cultivar averaged 60% (Peoples

et al., 1983). Stay-green maize genotypes redistributed less N and, therefore, took

up more N from the soil during seed filling than normal genotypes (Rajcan and

Tollenaar, 1999). Soil N levels also influenced the contribution of remobilized N

to total seed N (Pan et al., 1984).

The importance of redistributed N to the N budget of developing seeds may

have distorted our understanding of cause and effect relationships between leaf

senescence and seed-fill duration. The hypothesized need to remobilize N from

the leaf to sustain seed growth was thought to accelerate senescence and limit

seed-fill duration (Sinclair and de Wit, 1975, 1976; Frederick and Hesketh, 1994).

A presumed seed N “demand”—seeds cannot grow unless adequate N is

available—was the driving force behind the remobilization of leaf N. Models

based on this premise suggest that seed-fill duration could be sustained only by

(1) decreasing the total N requirement [total rate of N accumulation

(g m22 day21) of the seeds] (2) by a larger pool of remobilizable N at the

beginning of seed filling or (3) by taking more N from the soil during seed filling

(Frederick and Hesketh, 1994; Saini and Westgate, 2000).

The rate of N accumulation by the seeds, a function of total seed growth rate

and seed N concentration, is important only if seeds demand N from the

vegetative plant as a condition for growth. If seeds simply utilize the N made

available by the mother plant, seed growth would not necessarily affect the rate

of senescence, and variation in the supply of N may cause changes in seed N

concentration. The in vitro seed growth rate was relatively insensitive to N

availability in wheat (Barlow et al., 1983), maize (Singletary and Below, 1989)

and soybean (Saravitz and Raper, 1995; Hayati et al., 1996). In planta seedgrowth rate is also insensitive to N stress during seed filling (Egli et al., 1985;

Hayati et al., 1995). Seed N concentration is, however, sensitive to changes in N

supply (Hayati et al., 1995, 1996). Hayati et al. (1996) found that 17 mM N in

the media sustained maximum seed-growth rates, but 270 mM N produced the

highest seed N concentration. The preponderance of evidence suggests that there

is no active seed N demand; instead seeds seem to subsist on the N supplied by

the vegetative plant. Abandoning the concept of seed N demand eliminates a

primary mechanism by which seed growth could regulate senescence.

Models that assumed an active seed N demand (Sinclair and de Wit, 1975;

Frederick and Hesketh, 1994) predicted that changes in total seed growth rate



and, consequently, the total seed N requirement would affect the rate of

senescence. However, reducing seed number and total seed growth rate did not

lengthen the seed-filling period in some experiments (Frey, 1981; Jones and

Simmons, 1983; Kiniry, 1988) but it did in others (Konno, 1979; Egli et al., 1985;

Munier-Jolain et al., 1996, 1998; Egli and Bruening, 2001). Reducing

photosynthesis and total seed growth rate with shade also resulted in conflicting

responses, sometimes seed-fill duration increased and sometimes there was no

change (Frey, 1981; Simmons et al., 1982; Egli, 1999). Increasing photosynthesis

and total seed growth rate with elevated CO2 levels accelerated leaf senescence in

barley and wheat (Fangmeier et al., 2000), but there was no effect when high

radiation levels were used with soybean (Hayati et al., 1995). The high radiation

levels increased total seed growth rate, but seed N concentration was reduced and

the rate of senescence was not changed in non-nodulated soybean plants with and

without N available to the roots (Hayati et al., 1995). There was also no obvious

relationship between seed protein concentration, the concentration aspect of the

total seed N accumulation rate and seed-fill duration among species (Table I)

(Egli, 1981) or among cultivars within a species (Fig. 3). Salado-Navarro et al.

(1985), however, found that soybean genotypes with high seed protein

concentrations had shorter seed-fill durations and lower yields than genotypes

with normal seed protein levels. Experiments with several species following a

variety of treatment protocols did not establish a consistent relationship between

total seed growth rate and seed-fill duration, suggesting again that there may be

no active seed N demand driving senescence.

Figure 3 The relationship between effective-filling period and seed protein concentration for 11

soybean genotypes grown in the field at Lexington, Kentucky in 1998. Average size of seed from

small-seeded genotypes was 96 mg seed21 compared with 251 mg seed21 for the large seeded

genotypes. Egli and Bruening (1998), unpublished data.



A large supply of N in the plant at the beginning of seed fill, relative to that in

the seed at maturity, may slow senescence and extend seed filling (Sinclair and

de Wit, 1975; Triboi and Triboi-Blondel, 2002). Sinclair and Sheehy (1999) and

Sheehy (2001) argued that a larger pool of remobilizable N was needed to

increase yield of modern cultivars. There is no consistent relationship, however,

between the size of the N pool and seed-fill duration.

The variation in seed-fill duration among soybean cultivars was not related to

the size of the potentially remobilizable N pool at the beginning of seed filling

(Zeiher, 1980; Zeiher et al., 1982) (Fig. 4). Larger maximum vegetative mass and

a larger N pool are usually associated with long total growth duration, but seedfill duration does not necessarily increase as total growth duration increases

(Fig. 5) (Duncan, 1969; Egli, 1994, 1998). Stress during vegetative growth

frequently reduces plant size without reducing yield (Frederick and Hesketh,

1994; Jiang and Egli, 1995). There are reports of a significant association

between total leaf N at the beginning of seed fill and seed-fill duration (Shibles

and Sundberg, 1998) or yield (Loberg et al., 1984), but, in aggregate, the results

do not consistently support a model where seed-fill duration is determined by the

size of the remobilizable N pool.

Variation in pool size and the contribution of remobilized N must cause

offsetting changes in the supply of N from the nodules or from the soil solution.

This uptake during seed filling must be essentially zero when the contribution

from remobilizable N approaches 100%, but it must remain high when

remobilizable N makes only a small contribution. Variation in N fixation

Figure 4 The relationship between potentially remobilizable N and seed-fill duration (growth

stage R5 –R7) for eight soybean cultivars from maturity groups III to V grown in the field in 1978

(Zeiher, 1980). Potentially remobilizable N is the total N in the leaves, petioles, stems and pod walls at

beginning of seed fill (growth stage R5). Potentially remobilizable N was correlated ðr ¼ 0:99; n ¼ 8;

significant at P ẳ 0:001ị with the mass of the vegetative plant at beginning of seed-fill.



Figure 5 The relationship between the length of the total growth cycle (planting to PM) and the

duration of seed filling. (A) The effective filling period of four soybean cultivars grown in the field in

1989 (X) and 1990 (W) (Egli, 1993). (B) The days between growth stage R5 and R7 of eight soybean

cultivars grown in the field, 1978 (Zeiher et al., 1982). (C) Days from flowering to maturity for 38

diverse cowpea genotypes grown in the field at Riverside, California in 1983 (Dow el-madina and

Hall, 1986). Two genotypes with seed-fill durations near 30 day and one with a 15 day seed-fill

duration were not included in the regression analysis. (D) Days from silking to maximum seed mass

averaged across 1 or 2 years and several plant densities for 11 maize hybrids grown in the field

(Hanway and Russell, 1969).



patterns during seed filling in soybean that match these extremes has been

documented (Harper, 1974; Lawn and Brun, 1974; Denison and Sinclair, 1985).

If crops can obtain most of their seed N from uptake or fixation during seed

filling, the maximum rate of uptake/fixation may provide no meaningful

limitation to seed growth. Sheehy’s (1983) calculations supported this position

by suggesting that potential uptake rates were much greater than the

0.5 g m22 day21 assumed by Sinclair and de Wit (1975) and provided no

limitations to seed growth. Perhaps it is more reasonable to assume that the N

supply would be in balance with the productive capacity of the plant canopy, i.e.,

crop growth rate. If this is true, the N uptake rate may not be limiting, but N

availability in the soil or an unfavorable soil environment that inhibited nodule

function could limit N supply and affect seed-fill duration.

Seed growth requires a continuous supply of assimilate from the mother plant,

paradoxically, canopy photosynthesis starts declining early in seed-filling as the

leaves senesce and usually reaches zero when the seeds stop growing. Thus, the

seed-filling period is characterized by a steady decline in the supply of assimilate,

while the products of senescence are frequently a major source of N for the seed.

Senescence must be involved in the regulation of seed-fill duration and much

remains to be learned about this complex relationship, but describing leaf

senescence as a straight-forward interaction of N supply and seed N need is

probably a serious oversimplification. Although debate continues on the exact

metabolic mechanisms regulating the cessation of seed growth (Kermode, 1995;

Saini and Westgate, 2000) and leaf senescence (Buchanan-Wollaston, 1997), it is

clear that the characteristics of the seed and of senescence regulate seed-fill

duration. The duration of growth of a seed that normally reaches its maximum

potential size will not increase if leaf senescence is slowed without changes in

the characteristics of the seed and, conversely, there will be no benefit derived

from increasing the potential size of the seed if there is no delay in leaf

senescence to provide assimilate for continued seed growth. Slowing senescence

may, however, increase seed-fill duration in those species whose seeds do not

normally reach their potential size.




Yield of a grain crop is determined by the accumulation of dry matter by the


seeds. This accumulation has two components, a rate (g m22

) and a

land area day

duration, and either or both can contribute to variation in yield. There is ample

evidence from many crops that seed-fill duration can be associated with yield.



However, variation in total seed growth rate could also affect yield, or variation in

seed-fill duration could be compensated for by changes in total seed growth rate

to maintain a constant yield. Since yield is a function of a rate and a duration it is

unrealistic to expect that seed-fill duration will account for all variation in yield.

The plant’s environment affects seed-fill duration, as described previously,

and these effects are frequently translated into changes in yield. Water stress

shortened seed filling and reduced yield of chickpea (Davies et al., 1999),

soybean (Meckel et al., 1984; de Souza et al., 1997), maize (NeSmith and

Ritchie, 1992) and wheat (Frederick and Camberato, 1995). A shorter seed-filling

period played a role in yield reductions from N stress with wheat (Frederick and

Camberato, 1995) and soybean (Egli et al., 1978), and P and K stress with maize

(Peaslee et al., 1971). Nitrogen stress reduced leaf area duration and yield of

maize (Wolf et al., 1988), probably as a result of shortened seed-fill duration.

The effects of sowing date and irrigation on yield of pinto beans (Phaseolus

vulgarius L.) and field beans (Vicia faba L.) were expressed through changes in

leaf area duration which probably represented differences in seed-fill duration

(Husain et al., 1988; Dapaah et al., 2000). Wheat yield was closely associated

with leaf area duration across trials involving planting date, seeding rates and N

fertilizer rates (Fischer and Kohn, 1966).

Seed-fill duration is sensitive to temperature, and this variation frequently

translates into changes in yield. Artificially lowering night temperature increased

yield of wheat maize, and soybean, apparently as a result of a longer seed-filling

period (Peters et al., 1971). Lower temperatures and longer seed-filling periods

increased yield of oat (Hellewell et al., 1996) and wheat (Wardlaw et al., 1980).

Lower temperatures and longer seed-fill durations contributed to larger yields

at higher elevations (Cooper, 1979). However, compensatory effects of solar

radiation (Muchow, 1990) or seed-growth rate (Chowdhury and Wardlaw, 1978)

minimized changes in seed size and yield in other situations. Long seed-filling

periods may be partially responsible for exceptionally high yields in some

environments with moderate temperatures (Duncan et al., 1973; Muchow et al.,

1990; Sinclair and Bai, 1997; Hall, 2001).

Yield is frequently associated with genetic differences in seed-fill duration.

Positive associations between seed-fill duration and hybrid or cultivar yields were

found in maize (Daynard and Kannenberg, 1976; Bolanos, 1995), wheat

(Gebeyehou et al., 1982; Penrose et al., 1998), barley (Leon and Geister, 1994;

Dofing, 1997) and soybean (Hanway and Weber, 1971; Dunphy et al., 1979). The

higher yields of maize hybrids compared with inbreds were associated with a

longer seed-fill duration (Johnson and Tanner, 1972; Poneleit and Egli, 1979).

Dwyer et al. (1994), however, found a positive correlation between yield and

seed-fill duration in only two of nine combinations with early maturing maize

hybrids while a positive association occurred in only 2 of 4 years with oat

(Peltonen-Sainio, 1991). The yield advantage for tropical maize hybrids

over open-pollinated cultivars was not associated with seed-fill duration

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

IV. Regulation of Seed-Fill Duration

Tải bản đầy đủ ngay(0 tr)