Crop Science 41:1517-1521 (2001)
© 2001 Crop Science Society of America
CROP PHYSIOLOGY & METABOLISM
Flower Abortion Caused by Preanthesis Water Deficit Is Not Attributed to Impairment of Pollen in Soybean
M. Kokubun*,a,
S. Shimadab and
M. Takahashic
a Graduate School of Agricultural Science, Tohoku Univ., Aoba-ku, Sendai, 981-8555, Japan
b Tohoku National Agric. Exp. Stn., Kariwano, Akita, 019-2112, Japan
c National Agriculture Research Center, Kannondai, Tsukuba, 305-8666, Japan
* Corresponding author (kokubun{at}bios.tohoku.ac.jp)
 |
ABSTRACT
|
|---|
A large proportion of flowers in the soybean plant [Glycine max (L.) Merr.] abort during development. Water stress imposed during the development of flowers is a major factor increasing the rate of abortion, and long-term or frequent water deficits during this period might decrease yield. The physiological mechanism underlying this phenomenon remains unclear. The objectives of this study were (i) to discover whether a water deficit imposed on a soybean genotype IX93-100 prior to anthesis could cause the abortion of the proximal flowers, which are destined to produce seed-bearing pods at a high rate under optimal conditions, and (ii) to determine whether the abortion was due to the impairment of the pistil (ovule) or stamen (pollen) function. Water stress caused by restriction of watering for 3 d during the preanthesis stage significantly increased the abortion of the proximal flowers. The pistils of well-watered plants, pollinated with either stressed or nonstressed pollen, produced pods at a considerable rate, whereas only a small percentage of water-stressed pistils developed into pods, even when crossed with nonstressed pollen. These results suggest that flower abortion caused by a preanthesis water deficit is not attributed to an impairment of pollen, but was probably due to impairment of ovule function.
Abbreviations: DAF, days after flowering • PPFD, photosynthetic photon flux density
|
INTRODUCTION
|
|---|
THE YIELD of the soybean plant is determined by the number of seeds per unit area and individual seed weight. The number of seeds greatly depends upon the number of floral buds that initiate pods and attain maturity. Soybean plants produce an abundance of floral buds, but a large proportion of the ovaries abort prior to developing into mature pods (van Schaik and Probst, 1958; Kato, 1964; Brevedan et al., 1978; Wiebold et al., 1981).
The magnitude of abortion varies with the position on the plant, being greater in the branches, the lower part of the main stem and the top nodes of the main stem (Hansen and Shibles, 1978; Wiebold et al., 1981; Gai et al., 1984; Heindl and Brun, 1984). Within individual racemes, the proximal positions exhibit a higher pod-set percentage than do the distal positions (Huff and Dybing, 1980; Dybing et al., 1986; Spollen et al., 1986a; Carlson et al., 1987; Wiebold, 1990; Wiebold and Panciera, 1990; Kokubun and Honda, 2000). The physiological mechanism controlling reproductive abortion, however, remains unclear.
Possible physiological factors affecting flower abortion in the soybean include the quantity of certain plant hormones (Huff and Dybing, 1980; Beckmann, 1981; Heindl et al., 1982; Spollen et al., 1986b; Carlson et al., 1987; Yarrow et al., 1988; Kokubun and Honda, 2000), competition for carbohydrates and nutrients (Brevedan et al., 1978; Antos and Wiebold, 1984; Brun and Betts, 1984; Heitholt et al., 1986a,b), and the quality and quantity of light in the plant canopy (Heindl and Brun, 1983; Brun et al., 1985; Myers et al., 1987).
In soybean, most reproductive abortion occurs at an early stage of embryo development after fertilization. A deficient water supply during this time is a major environmental factor increasing the rate of abortion (Kato, 1964; Westgate and Peterson, 1993). Although increased abortion at one stage may be compensated by increased seed set at another stage of development, or by an adjustment of seed mass, frequent or long-term water deficits during flowering and early pod development decrease yield of soybean (Shaw and Laing, 1966; Sionit and Kramer, 1977). Water stress imposed during flowering reduces photosynthesis and the amount of photosynthetic assimilates allocated to floral organs, and might thereby increase the rate of abortion (Raper and Kramer, 1987). These observations raise the following question. Can water deficit prior to flowering cause an impairment of pistils or stamens which are undergoing development, leading to an increased rate of abortion? To address this question, experiments were conducted to determine (i) whether a water deficit imposed on a soybean genotype IX93-100 prior to flowering caused the abortion of the proximal flowers, which are destined to develop seed-bearing pods at a high rate under optimal conditions, and (ii) whether the abortion was due to the impairment of pistil (ovule) or stamen (pollen) function.
|
MATERIALS AND METHODS
|
|---|
Plant Material
Soybean plants were grown in pots (16-cm diam by 19 cm tall) each containing 3 kg of soil (Medial, mesic Hydric Hapludand), 0.6 g of N, 2.2 g of P, and 1.7 g of K. The soybean line used in this study was IX93-100. The seeds were provided by Dr. C.D. Dybing, South Dakota State University, Brookings, SD, via Dr. R.L. Nelson, USDA-ARS, Urbana, IL. This genotype has long racemes (up to approximately 10 cm). Within individual racemes, flowers open from the base to the tip at a rate of a few flowers per day. After emergence, plants were thinned to two plants per pot. The plants were grown in controlled environment chambers (30/20°C day/night temperature; 60 ± 10% relative humidity; 15-h photoperiod at 600 µmol m-2 s-1 photosynthetic photon flux density (PPFD) on the plant canopy, supplied by metal-halide lamps). Plants were irrigated several times a day except during periods of water-supply restriction. To ensure uniform plant growth, branches were removed whenever they emerged. Experiments were repeated in the two growth chambers of same capacity, with each containing 30 pots (60 plants). Each plant was considered as an experimental unit. Data are shown as averages of the two growth chamber trials.
Water Treatment
Plants were irrigated frequently and soil moisture was kept at a level that was 70% of the water-holding capacity of the soil, except during periods of water-supply restriction. When plants began to flower, the water supply was restricted to a level at which the water potential of the leaves fell below -1.5 MPa. This treatment was imposed on half of the plants (30 plants) and lasted for 3 d. After the treatment, the plants were rewatered to the original level. The water treatment was repeated twice in different growth chambers. Since there were many floral buds at different stages of growth within individual plants, these flowers received water stress at various growth stages prior to flower opening.
Measurement
For the plants grown without water restriction, the pod-set frequency of the flowers at different floral positions on a raceme was measured at maturity. Three racemes per plant at mid-canopy main stems nodes of 10 plants, for a total of 30 racemes, were used for this measurement.
In water-deficient plants and well-watered plants, the four most basal flowers (Position 1–4) on a raceme were tagged with colored tape, and their development and growth were monitored. The half of the monitored flowers were exposed to water-restriction treatment at stages ranging from -9 (9 d prior to flowering) to 2 d after flowering (DAF). The day of flowering (anthesis) was determined when the banner petal was fully extended in length (B2 stage according to Peterson et al., 1992). Pods that elongated between 10 and 15 d after anthesis were considered to be fertile pods. Pod-set percentage was calculated as the number of fertile pods divided by the number of flowers. Thirty racemes on 10 water-deficient and 10 well-watered plants were used.
The water potentials (
w) of leaves and flowers (pistils + stamens) were measured every day for 9 d after the initiation of water restriction treatment. Leaf discs 6 mm in diameter and flowers were sampled and immediately placed in a C-52-SF sample chamber (Wescor Inc., Logan, UT). The samples were allowed to equilibrate in the chamber for 3 h at 25°C, then measured with a Wescor model HR-33T dewpoint microvolt meter. For each sampling date, five terminal leaflets of recently expanded leaves and five newly opened flowers from five plants were used. Petals and calyxes were removed from the flowers. The measurement was carried out from 0900 to 1000 h.
Well-watered (WW) and water-deficient (WD) plants were reciprocally hand pollinated (WW x WD and WD x WW) daily for 10 d after the initiation of water restriction treatment. The hand pollination was made from 0800 to 0900 h (daytime started at 0600 h), at a stage when corolla became visible beyond calyx lobes, but not fully extended in length (B2 stage according to Peterson et al., 1992). Pod-set percentage was measured in the same way as described above. About twenty flowers from the basal position (Positions 1–4) of a raceme were used for single crosses every day. Pod-set percentage of the hand pollinated flowers was monitored in the same way as described above.
The apparent photosynthetic rates of the terminal leaflet of the most recently expanded leaves were measured with SPB-H Portable Photosynthesis and Transpiration Measurement System (Shimadzu Corp., Kyoto, Japan) at 0, 1, 3, and 5 d after water restriction treatment. Air for the leaf chamber, maintained at a flow rate of 500 mL min-1, was supplied through a mast installed outdoors that drew air from 5 m above the ground. The irradiance on the measured leaves was ca. 600 µmol m-2 s-1 PPFD. The measurement was carried out during 0900 and 1000 h.
|
RESULTS
|
|---|
Intraraceme Variation of Pod-Set Percentage
Pod-set percentage of well-watered plants was greater in the proximal positions (>70% at Positions 1–4), whereas distal flowers (Position 10 and above) never produced pods (Fig. 1).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1. Pod-set percentage of different positions on a raceme of well-watered plants. Soybean plants (IX93-100) were grown in pots in a controlled environment chamber (30/20°C day/night temperature, 15-h photoperiod). Each point represents the mean ± SE of 30 racemes.
|
|
Under self-pollination, the pod-set percentage of the proximal flowers was compared between well-watered plants and water-deficient plants. The percentage was lower in the water-deficient flowers than the well-watered flowers which retained a high percentage (Fig. 2). In particular, the flowers which suffered water deficit several days before flowering (-9 to -3 DAF) showed a marked decline of the pod-set frequency (less than 50%). The flowers which received the stress a few days after anthesis (1 and 2 DAF) showed no decline in frequency.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 2. Pod-set percentage at proximal positions (1–4) on racemes of self-pollinated flowers from well-watered and water-deficient plants, as a function of time of water restriction, expressed as days after flowering. Each point represents the mean ± SE of 30 racemes.
|
|
Water Potentials of Leaves and Flowers
Figure 3 shows the w of leaves and flowers which were monitored for 9 d after the initiation of water restriction treatment. In well-watered plants, the leaf w was -0.3 ± 0.1 MPa, while the flower w was -0.85 ± 0.15 MPa. The flower w was always lower than the leaf w, and the difference was nearly 0.5 MPa. Short-term water restriction caused a rapid decrease in the w of leaves and flowers. During the period of 3-d treatment, the w fell to -1.7 MPa in leaves and -2.0 MPa in flowers, respectively. As in the well-watered plants, the flower w was lower than the leaf w. Upon rewatering, the stressed plants recovered the control values of their leaf and flower w in 2 d.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. Effect of water restriction on water potentials of leaves and flowers in soybeans. WL, Well-watered leaf; WF, Well-watered flower; DL, Water-deficient leaf; DF, Water-deficient flower. Each point represents the mean ± SE of five plants.
|
|
Leaf Photosynthetic Rates
The apparent photosynthetic rates of leaves declined sharply with time after initiation of the water restriction treatment. Upon rewatering, the water-deficient leaves recovered the rates nearly to the control values (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Photosynthetic rates of leaves of well-watered (Control) and water-restricted (Restricted) soybean plants. After measurements on Day 3, plants of the water-restricted treatment were rewatered.
|
|
Pod-set Frequency of Flowers on Hand Pollinated Well-Watered and Water-Deficient Plants
Figure 4 shows the pod-set percentage of hand pollinated flowers. The crosses were made reciprocally between well-watered and water-deficient flowers. The pistils of well-watered plants, when pollinated with pollen from water-deficient plants, produced pods at a considerable rate (10–60%), whereas water-deficient pistils produced very few pods when crossed with pollen from well-watered plants.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 4. Pod-set percentage of hand pollinated flowers at proximal positions (1–4) on racemes. Reciprocal crosses were made between well-watered (WW) and water-deficient (WD) soybean plants (IX93-100). Each point represents the mean ± SE of 20 racemes.
|
|
|
DISCUSSION
|
|---|
The basal flowers (Positions 1–4) of the IX93-100 soybean showed a high pod-set percentage (>70%) when grown under well-watered condition (Fig. 1 and 2), in agreement with our previous observation (Kokubun and Honda, 2000). In a water-deficient environment, however, the percentage of pods produced from basal flowers decreased markedly (Fig. 2). The decrease was particularly significant when the plants experienced the deficit several days before their anthesis (-8 to -3 DAF). The period of several days before anthesis corresponds to a time when the reproductive organs, consisting of pistils and stamens, are being actively formed (Carlson and Lersten, 1987). The decreased pod-set percentage under conditions of a restricted water supply suggests that the structure and/or function of these organs was impaired as a result of a water deficit during their formation.
In soybean, water-supply restriction decreased leaf w, leading to decreased photosynthesis when the w fell below -1.1 MPa (Boyer, 1970). In the present experiment, the leaf w of water-deficient plants were below -1.1 MPa for a few days (Fig. 3), and photosynthetic rates of the stressed leaves were very low compared with well-watered plants (Table 1). Decreased photosynthate production might have reduced the allocation of assimilates to reproductive organs, which could have been a reason for the increased rates of flower abortion in water-deficient plants, as indicated by Raper and Kramer (1987). This appeared to be true during the period of 1 to 4 d after the initiation of water restriction treatment, because the leaf w of water-restricted plants was significantly lower than that of well-watered plants. However, the leaf w of the water-deficient plants recovered to the control value after rewatering (Fig. 3), and the photosynthesis did as well (Table 1). In spite of the recovery of leaf w and photosynthesis a few days after rewatering, the pod-set percentage of the water-deficient plants was significantly lower than that of the well-watered plants (Fig. 4). This result suggests that the decreased pod-set percentage could be caused by a factor other than a reduction of photosynthetic assimilation.
The w of flowers (excluding petals) were always lower than those of leaves. This phenomenon was previously reported by Westgate and Peterson (1993). The difference in w between flowers and leaves was nearly 0.2 to 0.3 MPa in their study, whereas it was 0.5 MPa or more in some cases according to our measurements (Fig. 3). Under conditions of water deficiency, the flower w fell to nearly -2.0 MPa (Fig. 3). This characteristic of having a low flower w also suggests that an impairment of the reproductive organs is likely to be induced by water stress imposed during the development of these organs.
For soybean, there have been no reports showing pollen and pistil w separately, probably because the measurement is technically difficult to gauge. In maize, pollen w was always found to be substantially lower than silk w, and silk w was found to vary with the water status in a plant, but pollen w did not (Westgate and Boyer, 1986a). Since soybean and maize have different reproductive structures, the evidence observed in maize may not hold true for soybean. Their suggestion that pollen desiccation should not be a factor limiting grain production in maize is, however, noteworthy, because pollen does not lose viability at w as low as -12.5 MPa (Westgate and Boyer, 1986b). The fact that pollen which received severe water stress during development could produce a considerable number of pods when crossed with nonstressed pistils (Fig. 4) may be explained by this suggestion.
It is difficult to evaluate directly whether pistils and/or pollen may be viable under conditions of water deficiency. Therefore, we indirectly evaluated this viability by observing the pod-set percentage of flowers that were made reciprocal crosses between water treatments. The pistils of well-watered plants that were pollinated with pollen from water-deficient plants, were able to grow pods at a considerable rate, whereas water-deficient pistils barely grew pods, even when crossed with nonstressed pollen (Fig. 4). These results suggest that flower abortion caused by a water deficit prior to anthesis is not attributed to an impairment of pollen, but was probably due to impairment of ovule function.
Stressed flowers, in which both pistils and pollen experienced a water deficit, showed a pod-set percentage of 20 to 60% when self-pollinated (Fig. 2), whereas stressed pistils crossed with pollen from well-watered plants (WD x WW) barely grew pods (Fig. 4). The reason for this unexpected result is not clear, since the present study lacks data on flowers that were hand pollinated between the same water conditions (WW x WW or WD x WD). One possible explanation is that a great discrepancy in water potential between pistils and pollen caused an incompatibility between the two. Another possibility is that physical disruption of petals and stamens by emasculation could induce water loss from the emasculated flowers, leading to a further decline of water potential of the stressed pistils.
|
ACKNOWLEDGMENTS
|
|---|
We thank Dr. C.D. Dybing and Dr. R.L. Nelson for the gift of IX93-100 seeds, and Dr. M. Westgate for helping to improve the manuscript. The technical assistance of Ms. Y. Kaneko is also gratefully acknowledged.
|
NOTES
|
|---|
Research financed by Ministry of Agriculture, Forestry and Fisheries of Japan.
Received for publication September 29, 2000.
|
REFERENCES
|
|---|
- Antos, M., and W.J. Wiebold. 1984. Abscission, total soluble sugars, and starch profiles within a soybean canopy. Agron. J. 76:715–719.[Abstract/Free Full Text]
- Beckmann, K.A. 1981. The internal control of flower abscission on soybeans. Ph D thesis (Diss. Abstr. 81-23613), Purdue University, West Lafayette, IN.
- Boyer, J.S. 1970. Leaf enlargement and metabolic rates in corn, soybean, and sunflower at various leaf water potentials. Plant Physiol. 46:233–235.[Abstract/Free Full Text]
- Brevedan, R.E., D.B. Egli, and J.E. Legget. 1978. Influence of N nutrition on flower and pod abortion and yield of soybeans. Agron. J. 70:81–84.[Abstract/Free Full Text]
- Brun, W.A., and K.J. Betts. 1984. Source/sink relations of abscising and nonabscising soybean flowers. Plant Physiol. 75:187–191.[Abstract/Free Full Text]
- Brun, W.A., J.C. Heindl, and K.J. Betts. 1985. The physiology of reproductive abscission in soybeans, p. 866–874. In R. Shibles (ed.) World Soybean Conference III: Proceedings. Westview Press, Boulder, CO.
- Carlson, D.R., D.J. Dyer, C.D. Cotterman, and R.C. Durley. 1987. The physiological basis for cytokinin induced increases in pod set in IX93-100 soybeans. Plant Physiol. 84:233–239.[Abstract/Free Full Text]
- Carlson, J.B., and N.R. Lersten. 1987. Reproductive morphology, p. 95–134. In J.R. Wilcox (ed.) Soybeans: Improvement, production, and uses (2nd ed.). ASA, CSSA, SSSA, Madison, WI.
- Dybing, C.D., H. Ghiasi, and C. Paech. 1986. Biochemical characterization of soybean ovary growth from anthesis to abscission of aborting ovaries. Plant Physiol. 81:1069–1074.[Abstract/Free Full Text]
- Gai, J., R.G. Palmer, and W.R. Fehr. 1984. Bloom and pod set in determinate and indeterminate soybeans in China. Agron. J. 76: 979–984.[Abstract/Free Full Text]
- Hansen, W.R., and R. Shibles. 1978. Seasonal log of the flowering and podding activity of field-grown soybeans. Agron. J. 70:47–50.[Abstract/Free Full Text]
- Heindl, J.C., D.R. Carlson, W.A. Brun, and M.L. Brenner. 1982. Ontogenetic variation of four cytokinins in soybean root pressure exudate. Plant Physiol. 70:1619–1625.[Abstract/Free Full Text]
- Heindl, J.C., and W.A. Brun. 1983. Light and shade effects on abscission and 14C-photoassimilate partitioning among reproductive structures in soybean. Plant Physiol. 73:434–439.[Abstract/Free Full Text]
- Heindl, J.C., and W.A. Brun. 1984. Patterns of reproductive abscission, seed yield, and yield components in soybean. Crop Sci. 24:542–545.
- Heitholt, J.J., D.B. Egli, and J.E. Legget. 1986a. Characteristics of reproductive abortion in soybean. Crop Sci. 26:589–595.[Abstract/Free Full Text]
- Heitholt, J.J., D.B. Egli, J.E. Legget, and C.T. MacKown. 1986b. Role of assimilate and carbon-14 photosynthate partitioning in soybean reproductive abortion. Crop Sci. 26:999–1004.[Abstract/Free Full Text]
- Huff, A., and C.D. Dybing. 1980. Factors affecting shedding of flowers in soybean (Glycine max (L.) Merrill). J. Exp. Bot. 31:751–762.[Abstract/Free Full Text]
- Kato, I. 1964. Histological and embryological studies on fallen flowers, pods and abortive seeds in soybean, Glycine max (L.). Tokai-Kinki Natl. Agric. Exp. Stn. Bull. 11:1–52.
- Kokubun, M., and I. Honda. 2000. Intra-raceme variation in pod-set percentage is associated with cytokinin content in soybeans. Plant Prod. Sci. 3:354–359.
- Myers, R.L., W.A. Brun, and M.L. Brenner. 1987. Effect of raceme-localized supplemental light on soybean reproductive abscission. Crop Sci. 27:273–277.[Abstract/Free Full Text]
- Peterson, C.M., C.O'H. Mosjidis, R.R. Dute, and M.E. Westgate. 1992. A flower and pod staging system for soybean. Ann. Bot. (London) 69:59–67.[Abstract/Free Full Text]
- Raper, C.D., Jr., and P.J. Kramer. 1987. Stress physiology, p. 589–641. In J.R. Wilcox (ed.) Soybeans: Improvement, production, and uses (2nd ed.). ASA, CSSA, SSSA, Madison, WI.
- Shaw, R.H., and D.R. Laing. 1966. Moisture stress and plant response, p.77–94. In W.H. Pierre et al. (ed.) Plant environment and efficient water use. ASA. Madison, WI.
- Sionit, N., and P.J. Kramer. 1977. Effect of water stress during different stages of growth of soybeans. Agron. J. 69:274–278.[Abstract/Free Full Text]
- Spollen, W.G., W.J. Wiebold, and D.S. Glenn. 1986a. Intraraceme competition in field-grown soybean. Agron. J. 78:280–283.[Abstract/Free Full Text]
- Spollen, W.G., W.J. Wiebold, and D.S. Glenn. 1986b. Effect of intraraceme competition on carbon-14-labeled assimilate and abscisic acid in soybean. Crop Sci. 26:1216–1219.[Abstract/Free Full Text]
- van Schaik, P.H., and A.H. Probst. 1958. The inheritance of inflorescence type, peduncle length, flowers per node, and percent flower shedding in soybeans. Agron. J. 50:98–102.[Abstract/Free Full Text]
- Westgate, M.E., and J.S. Boyer. 1986a. Silk and pollen water potentials in maize. Crop Sci. 26:947–951.[Abstract/Free Full Text]
- Westgate, M.E., and J.S. Boyer. 1986b. Reproduction at low silk and pollen water potentials in maize. Crop Sci. 26:951–956.[Abstract/Free Full Text]
- Westgate, M.E., and C.M. Peterson.1993. Flower and pod development in water-deficient soybeans (Glycine max L. Merr.). J. Exp. Bot. 258:109–117.
- Wiebold, W.J. 1990. Rescue of soybean flowers destined to abscise. Agron. J. 82:85–88.[Abstract/Free Full Text]
- Wiebold, W.J., D.A. Ashley, and H.R. Boerma. 1981. Reproductive abscission levels and patterns for eleven determinate soybean cultivars. Agron. J. 73:43–46.[Abstract/Free Full Text]
- Wiebold, W.J., and M.T. Panciera. 1990. Vasculature of soybean racemes with altered intraraceme competition. Crop Sci. 30:1089–1093.[Abstract/Free Full Text]
- Yarrow, G.L., W.A. Brun, and M.L. Brenner. 1988. Effect of shading individual soybean reproductive structures on their abscisic acid content, metabolism, and partitioning. Plant Physiol. 86:71–75.[Abstract/Free Full Text]
Related articles in Crop Science:
- This issue in Crop science
Crop Science 2001 41: 1379-1380.
[Full Text]
This article has been cited by other articles:
|
|
|
|
 
K. Sun, K. Hunt, and B. A. Hauser
Ovule Abortion in Arabidopsis Triggered by Stress
Plant Physiology,
August 1, 2004;
135(4):
2358 - 2367.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
