Published in Crop Sci. 43:1760-1763 (2003).
© 2003 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
CROP BREEDING, GENETICS & CYTOLOGY
The Effect of Grafting on the Flowering of Near-Isogenic Lines of Soybean
Tomasz Przepiorkowski and
Steven K. St. Martin*
Dep. of Horticulture and Crop Science, Ohio Agric. Res. and Dev. Ctr., The Ohio State Univ., Columbus, OH 43210-1086
* Corresponding author (stmartin+{at}osu.edu).
 |
ABSTRACT
|
|---|
Grafting of soybean [Glycine max (L.) Merr.] genotypes differing in time to flower has been used to study translocation of flowering hormones and also as a tool to facilitate crossing. Our objective was to use Y-shaped, grafted soybean plants to examine the mobility of the flowering stimulus. We grafted plants of ‘Harosoy’ (L58-266, homozygous for the e1e2E3e5 genotype) and four isolines of Harosoy [L62-667 (e1e2e3e5), L71-802 (E1e2e3e5), L84-307 (e1E2e3e5), and L84-337 (e1e2e3E5)], differing by single maturity genes. The scion was reciprocally grafted on one of the branches, leaving the second branch as a stock. Ungrafted and self-grafted controls were included, and plants were grown in growth chambers under short- (12-h) and long-day (20-h) photoperiods. All treatments flowered at the same time under short-day photoperiod. Under long-day photoperiod, the greatest delay of flowering was observed in lines containing the E1 and E3 alleles. The stock and scion flowered independently of each other and no stock x scion interaction occurred. Hypothesizing that leaving full foliage on the plants could have been a reason for the lack of mutual effect of scion and stock, we performed a second experiment where one, two, four, eight, and all leaves were maintained on the scion. The scion and stock affected each other only slightly, and only when one or two leaves were present on a scion. The location of the first flowering node suggested the promotion of flowering in the isogenic line containing the late E3 allele.
|
INTRODUCTION
|
|---|
SINCE THE DISCOVERY of photoperiodic sensitivity in soybean by Garner and Allard (1920), many investigations have been performed concerning this subject. Most of the Asian soybean cultivars introduced to the USA at the beginning of the 20th century were not fully adapted to the growing conditions of North America. Through selection and hybridization, soybean cultivars have been developed to best fit growing conditions of specific regions and divided into 13 maturity groups (Bernard et al., 1987). There is a great diversity among soybean cultivars as to their sensitivity to photoperiod. Generally, early maturing cultivars are less sensitive to the daylength than late-maturing cultivars (Byth, 1968; Criswell and Hume, 1972). Criswell and Hume (1972) also found that in photoperiodically sensitive cultivars (00 to IV maturity groups), increasing the photoperiod from 12 to 24 h increased time from germination to the first bloom. Even though soybean is generally a short-day plant, the early maturing and photoperiodically sensitive cultivars have longer critical photoperiods than later maturing cultivars (Borthwick and Parker, 1938, 1939).
Many physiological and morphological changes within the soybean plant have been reported under short photoperiod. Zhang and Du (1999) found that photoperiodically sensitive cultivars display shortened vegetative stages (Ve to Vt) (Fehr and Caviness, 1977), as well as shortened reproductive stages when grown under 8- vs. 16-hr days. The plants grown under short-day conditions were shorter and had fewer nodes and branches in comparison to long-day counterparts. Thomas and Raper (1976), however, found that the number of branches and development of flower primordia were determined not only by photoperiod, but rather by interaction of photoperiod and temperature.
The development of isogenic lines prompted studies on the effect of different alleles controlling flowering. Five loci control flowering and maturity in soybean: E1 and E2 (Bernard, 1971), E3 (Buzzell, 1971), E4 (Buzzell and Voldeng, 1980), and E5 (McBlain and Bernard, 1987). The dominant allele at each locus conditions delays in flowering and maturity. McBlain and Bernard (1987) described the E1 allele as delaying flowering and maturity in isogenic lines of ‘Clark’ and Harosoy. Both the E2 and E3 alleles delayed pre- and postflowering periods in the same cultivars. The E5 allele was similar in responses to the E2 allele. The E4 allele delayed flowering by 1 to 6 d and delayed maturity by 8 to 20 d in several different genetic backgrounds (Saindon et al., 1989). Cober et al. (1996a)(b) found the E1 allele to be most sensitive to light quality, and flowering was delayed by a decrease in the ratio of red to far red light. The same authors found the E3 to be the least photoperiodically sensitive. With near-isogenic lines, they also found no significant differences in time of flowering and maturity under a 12-h daylength but that major differences occurred under a 20-h daylength. The E1, E3, and E4 alleles all delayed flowering under those conditions.
Reciprocal grafting has been used for investigating many aspects of flowering in soybean. Kiyosawa and Kiyosawa (1962) used side-grafting techniques in their studies on genetic differences. The reciprocal technique was improved by Bezdicek et al. (1972). Kiihl et al. (1977) used grafting as an effective tool in breeding, successfully inducing flowering of late-flowering cultivars of soybean by grafting them on early maturing cultivars. The use of two branched, Y-shaped soybean plants has also been described at length (Asian Vegetable Research and Development Center, 1977). Borthwick and Parker (1938) first used Y-shaped plants for photoperiodic studies of soybean, where each branch received a different daylength.
The question arises whether the flowering promoting or inhibiting substances can be translocated through the graft. The translocation of the flowering hormones through the graft in both directions was shown by Kiyosawa and Kiyosawa (1962). They discovered that an early flowering cultivar grafted on the stock of a late-flowering cultivar produced enough flowering hormone to stimulate flowering of the stock. The flower-promoting substances produced by the early cultivar overcame the effect of flowering inhibitor(s) produced by the late cultivar. Shanmugasundaram et al. (1979) showed that flower-inducing substance(s) can be translocated from the donor to the receptor branch in Y-shaped soybean, but only when the receptor branch had either no or at most two trifoliolate leaves. When four or more leaves were present on the receptor branch, either the photosynthates produced by the leaves diluted the flowering-inhibiting substance passed across the graft, or the flowering inhibitor(s) produced by the receptor countered the flowering-inducing substances from the donor branch. Experiments with 14CO2 performed by Gent (1982a)(b) supported the possibility of putative branch translocation of carbohydrates. By measuring the final yield, he concluded that the uptake of carbohydrates was not limited by the resistance to long distance translocation especially with a strong sink (such as developing seeds). But he detected some resistance to translocation of photosynthates between branches.
Grafting of near-isogenic soybean lines should be more precise than grafting different cultivars. With the lines, which differ only by one allele, we can also test the strength or effect of those alleles on each other. The objective of this study was to further investigate the mobility of the flowering stimulus in Y-shaped soybeans through grafting.
|
MATERIALS AND METHODS
|
|---|
Five near-isogenic lines were used in the experiment: Harosoy (L58-266) homozygous for the e1e2E3e5 genotype, L62-667 (e1e2e3e5), L71-802 (E1e2e3e5), L84-307 (e1E2e3e5), and L84-337 (e1e2e3E5), all of which were developed by Bernard et al. (1991). Line L62-667 contained all recessive alleles, and the other isogenic lines differed only in being homozygous dominant at one locus. The seeds were planted in 10-cm pots into a standard greenhouse peat-lite Metro Mix 360 (peat moss/perlite/vermiculite 1:1:1 v/v/v) (Scotts-Sierra Horticultural Products, Marysville, OH) media and placed in the growth chamber. The main photoperiod consisted of 11.5 h of light provided by incandescent and fluorescent lamps with average PPFD of 220 µmol s-1 m-2, followed by 0.5 h (for 12-h photoperiod) and 8.5 h (for 20-h photoperiod) of low intensity light from incandescent lamps with average PPFD of 30 µmol s-1 m-2. The temperature regime was 26°C day and 22°C night. Plants were fertilized once a week with 0.2 g kg-1 N with a 20-10-20 N-P-K (Peter's Complete, J. R. Peters, INC, Allentown, PA) liquid fertilizer. Two photoperiod treatments were planted; one was exposed to 12-hr, while the second was grown under a 20-h photoperiod. The plants were then decapitated above the cotyledons, allowing two simultaneous branches to develop from each cotyledon's axillary bud. About a one-month-old plant (two trifoliolate leaves) was then reciprocally grafted with each of the other near-isogenic lines. Also included as controls were ungrafted and self-grafted plants of each line. Self grafted plants were grafted with the scion of the same genotype as the stock. Only one branch was grafted, leaving the second one as a stock. The grafting site was wrapped with a wax film to prevent drying and covered for 48 h with a plastic bag. The scions were 
10 cm long, and defoliated before grafting. Multiple grafts were performed for each treatment, and only the four best-growing plants were used for the experiment. The rate of grafting success was 95%. There were four replications of each treatment and the experiment was set up as a completely randomized design. Plants under the 12-h photoperiod remained in the growth chambers for the duration of the experiment, while the plants from the 20-h photoperiod were moved to the greenhouse 1 wk after grafting. The natural light in the greenhouse was supplemented by high-sodium lamps, providing a 20-h photoperiod. The entire experiment was repeated. A combined ANOVA was performed, with repetitions considered as a random factor. The treatment x repetition mean square was used as the error term for comparing treatment. The time from grafting to the opening of the first flower was recorded for the scion and stock of each plant. The time of flowering for ungrafted controls was measured from time of grafting of all treatments within the photoperiod until the first open flower.
An additional experiment was performed with the two Harosoy near-isogenic lines most different (among pairs differing at a single locus) in their flowering time: e1e2e3e5 and e1e2E3e5. The seeds were planted in growth chambers under a 20-hr photoperiod. About 1-mo-old Y-shaped plants were then reciprocally grafted on each other. One week later they were moved to the greenhouse and grown under a 20-h photoperiod. The experimental treatments consisted of different numbers of leaves maintained on a scion, while leaving all the leaves on the stock part of the plant. The treatments were one, two, four, eight, and all leaves left on a scion in addition to the ungrafted and self-grafted controls. The time from grafting to bloom and the number of the plant node with the first flower were recorded. Like the first experiment, the design was completely randomized with four replications, and the entire experiment was repeated, with treatments x repetition mean square used as the error term.
|
RESULTS AND DISCUSSION
|
|---|
No significant differences were found between treatments under the 12-h photoperiod. Stocks and scions of all lines flowered within 14 to 16 d after grafting, as did the ungrafted controls. This confirms the findings of Cober et al. (1996a)(b), that no differences occur between the alleles in time of flowering under short-day photoperiod. The soybean plants were short and lacked lateral branches under a 12-h daylength.
Differences among treatments were observed under the 20-h photoperiod. For the time of flowering of the stock, the F test was significant for the stock only, the scion having no influence. For example, the E1 and E3 alleles in the scion did not delay flowering of the e1e2e3e5 stock. This shows that the genotype of the scion did not influence the flowering of the stock. Also, the scion x stock interaction was not significant. Because of the lack of scion x stock interaction, the flowering date of each stock was averaged across the five isogenic scions grafted to it. For each line, the flowering time of ungrafted plants was a few days shorter than that of their grafted counterparts (Table 1). In both ungrafted and grafted near-isogenic lines exposed to the long-day conditions, only the E1 and E3 alleles significantly delayed flowering by 10 d (E1) and 20 to 24 d (E3) in comparison with the L62-667 line, which had all recessive alleles. The E3 allele was described by Cober et al. (1996a) as the most photoperiodically responsive, providing the greatest delay of flowering. The E2 and E5 alleles did not significantly delay flowering. These results could have been affected by the natural light conditions in the greenhouse. Some of the alleles have strong photoperiodic reactions only under certain light spectra (Cober et al., 1996b), and the response we observed for the E2 allele might not have been characteristic.
View this table:
[in this window]
[in a new window]
|
Table 1. Time to bloom from grafting until first flower opened of stocks averaged across the grafts of five different isogenic soybean lines used as scions differing by one allele under a 20-h photoperiod.
|
|
A similar situation occurred in the time of flowering of the scion (Table 2). Both grafted and ungrafted scions reacted similarly to the daylength, with a difference of a longer time until flowering for grafted isolines. The longer time needed for flowering of grafted scions was probably because of the recovery time needed for the scion after grafting. The F test was significant only for the scion; the genotype of the stock had no influence on the flowering time of the scion. The relative differences in flowering times of the scions were the same regardless of which stock was grafted to the scion; that is, no significant interaction was observed between scion and stock for flowering time. Therefore, the times of flowering for each scion were averaged across all five isogenic lines used as stock. Only the E1 and E3 alleles significantly delayed flowering of the scion by 12 and 27 d, respectively (Table 2).
View this table:
[in this window]
[in a new window]
|
Table 2. Time to bloom from grafting until first flower opened of scions averaged across five different isogenic soybean lines used as stocks differing by one allele under a 20-h photoperiod.
|
|
Varying the number of leaves maintained on the scion did not give strong evidence for translocation of flowering hormones between branches in Y-shaped plants of Harosoy near-isogenic lines (Table 3). Self-grafted plants flowered 9 and 11 d later than ungrafted controls, for the e3 and E3 alleles respectively. The flowering of the early maturing (e3) scion was slightly (but not significantly) delayed when the scion had one or two leaves. Similarly, slightly (but not significantly) earlier flowering of the late-maturing (E3) scion occurred when grafted on the early stock, but only when the scion contained one or two leaves. Increasing the number of leaves maintained on the scion caused a reverse reaction. The reaction of the stock to scion with different numbers of leaves was not conclusive (Table 3). In both cases, the stock flowered at a similar time to the grafted control when more than two leaves were present. These results are in agreement with the research by Shanmugasundaram et al. (1979), where the number of leaves of the receptor branch determined the action of flowering stimulus from the other branch.
View this table:
[in this window]
[in a new window]
|
Table 3. Time to bloom from grafting until first flower open of the stock and scion of early (e1e2e3e5) and late-maturing (e1e2E3e5) isogenic lines of ‘Harosoy’ soybean grafted on each other with different numbers of leaves maintained on a scion. The stock retained all its leaves in all treatments. Plants were grown under a 20-h photoperiod.
|
|
The location of the first node to flower also exhibited only slight effects of grafting (Table 4). The stock and scion started flowering on similar nodes, except when the scion of the E3 isoline was grafted on the e3 stock. In this case, flowering occurred on higher nodes when more than four leaves were present on a scion.
View this table:
[in this window]
[in a new window]
|
Table 4. Number of the first flowering node of the stock and scion of early (e1e2e3e5) and late-maturing (e1e2E3e5) isogenic lines of ‘Harosoy’ soybean grafted on each other with different numbers of leaves maintained on a scion. The stock retained all its leaves in all treatments. Plants were grown under a 20-h photoperiod.
|
|
The lack of stock x scion interaction for time to flowering is an important finding that will simplify future experiments. The two factors operate independently and it may not be necessary to test a scion with more than one stock or a stock with more than one scion.
Overall, the effect of grafting on flowering was minimal in our experiments. A possible explanation is that the flower-inducing substance(s) produced by one branch could have been successfully countered by the inhibitors from the other branch. Similar results were observed by Shanmugasundaram et al. (1979), where the number of leaves of the receptor branch determined the action of the flowering stimulus from the other branch. The receptor branch had to have no more than three trifoliolate leaves to be able to react to the flowering promoter. In our research, only the branches which were defoliated showed some promotion (for the late-maturing isoline) or delay (for the early maturing isoline). Another possibility is that there was not enough strength of any of the dominant alleles, operating singly in these pairs of isogenic lines, to produce a sufficient amount of flowering promoter to overcome the inhibitors. The results of promoting or inhibiting the flowering time in isogenic lines were less definite than those with cultivars of widely differing maturity (Kiihl et al., 1977). The theory that grafting of near-isogenic lines involves less translocation of signal than grafts between genetically distinct cultivars should be investigated more fully in the future.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank Dr. James D. Metzger for helpful suggestions.
|
NOTES
|
|---|
Salaries and research support provided by state and federal funds appropriated to the Ohio Agric. Res. and Dev. Ctr., The Ohio State Univ. This report is Journal Article no. HCS 01-43.
Received for publication June 13, 2002.
|
REFERENCES
|
|---|
- Asian Vegetable Research and Development Center. 1977. Decapitation technique to screen for photoperiod insensitivity in soybean, Glycine max (L.) Merr. Soybean Genet. Newsl. 4:13.
- Bernard, R.L. 1971. Two genes for time of flowering and maturity in soybeans. Crop Sci. 11:242–244.[Abstract/Free Full Text]
- Bernard, R.L., G.A. Juvik, and R.L. Nelson. 1987. USDA Soybean germplasm collection inventory. Vol. I. INTSOY ser. No. 30.
- Bernard, R.L., R.L. Nelson, and C.R. Cremeens. 1991. USDA soybean genetic collection: Isoline collection. Soybean Genet. Newsl. 18:27–57.
- Bezdicek, D.F., B.H. Magee, and J.A. Schillinger. 1972. Improved reciprocal grafting technique for soybean (Glycine max L.). Agron. J. 64:558.[Abstract/Free Full Text]
- Borthwick, H.A., and M.W. Parker. 1938. Photoperiodic perception in Biloxi soybeans. Bot. Gaz. (Chicago) 100:374–387.
- Borthwick, H.A., and M.W. Parker. 1939. Photoperiodic responses of several varieties of soybean. Bot. Gaz. (Chicago) 101:341–365.
- Buzzell, R.I. 1971. Inheritance of soybean flowering response to fluorescent-day length conditions. Can. J. Genet. Cytol. 13:703–707.[Web of Science]
- Buzzell, R.I., and H.D. Voldeng. 1980. Inheritance of insensitivity to long day length. Soybean Genet. Newsl. 7:26–29.
- Byth, D.E. 1968. Comparative photoperiodic responses for several soybean varieties of tropical and temperate origins. Aust. J. Agric. Res. 19:879–890.
- Cober, E.R., J.W. Tanner, and H.D. Voldeng. 1996a. Genetic control of photoperiod response in early maturing near-isogenic soybean lines. Crop Sci. 36:601–605.[Abstract/Free Full Text]
- Cober, E.R., J.W. Tanner, and H.D. Voldeng. 1996b. Soybean photoperiod-sensitive loci respond differently to light quality. Crop Sci. 36:606–610.[Abstract/Free Full Text]
- Criswell, J.G., and D.J. Hume. 1972. Variation in sensitivity to photoperiod among early maturing soybean strains. Crop Sci. 12:657–660.[Abstract/Free Full Text]
- Fehr, W.R., and C.E. Caviness. 1977. Stages of soybean development. Spec. Rep. No. 80, Cooperative Extension Services, Iowa State University, Ames, Iowa.
- Garner, W.W., and A.H. Allard. 1920. Effect of the relative length of day and night and other factors of the environment on growth and reproduction in plants. J. Agric. Res. (Washington, DC) 18:553–606.
- Gent, P.N.M. 1982a. Effect of defoliation and depodding on long distance translocation and yield in Y-shaped soybean plants. Crop Sci. 22:245–250.
- Gent, P.N.M. 1982b. Effect of defoliation and depodding on 14CO2 assimilation and photosynthate distribution in Y-shaped soybean plants. Crop Sci. 22:860–867.[Abstract/Free Full Text]
- Kiihl, R.A.S., E.E. Hartwig, and T.C. Kilen. 1977. Grafting as a tool in soybean breeding. Crop Sci. 17:181–182.[Abstract/Free Full Text]
- Kiyosawa, S., and K. Kiyosawa. 1962. A study of varietal differences in flowering habits of soybean plants as followed by grafting experiments. Plant Cell Physiol. 3:263–273.[Abstract/Free Full Text]
- McBlain, B.A., and R.L. Bernard. 1987. A new gene affecting the time of flowering and maturity in soybean. J. Hered. 78:160–162.[Abstract/Free Full Text]
- Saindon, G., W.D. Beversdorf, and H.D. Voldeng. 1989. Adjustment of soybean phenology using the E4 locus. Crop Sci. 29:1361–1365.[Abstract/Free Full Text]
- Shanmugasundaram, S., W. Chao-Chin, and T.S. Toung. 1979. Photoperiodic response of flowering in two-branched soybean plants. Bot. Gaz. (Chicago) 140(4):4141–4147.
- Thomas, J.F., and C.D. Raper, Jr. 1976. Photoperiod control of seed filling for soybean. Crop Sci. 16:667–672.[Abstract/Free Full Text]
- Zhang, G., and W. Du. 1999. The effect of daylength on the growth of soybean and the creation of wide-adaptation germplasm. [Online]. [6 p.] Available at: http://www.soygenetics.org/articles/sgn1999001.html [verified 10 Apr. 2003]. Soybean Genet. Newsl. 26 (February).
TOP
ABSTRACT
INTRODUCTION
