Journal of Experimental Botany, Vol. 54, No. 381, pp. 445-450,
January 2, 2003
© 2003 Oxford University Press
Mutant alleles at the rugosus loci in pea affect seed moisture sorption isotherms and the relations between seed longevity and moisture content
Received 11 June 2002; Accepted 24 September 2002
1 Department of Agriculture, The University of Reading, Earley Gate, PO Box 237, RG6 6AR, UK
2 School of Plant Sciences, The University of Reading, Reading RG6 6AS, UK
3 John Innes Institute, Colney Lane, Norwich NR4 7UH, UK
4To whom correspondence should be addressed. Fax: +44 (0)118 378 8297. E-mail: r.h.ellis{at}reading.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Pea (Pisum sativum L.) mutant near-isogenic lines (RRrbrb, rrRbRb, rrrbrb) with lower starch but higher lipid contents, brought about by lesions in the starch biosynthetic pathway, had seed moisture sorption isotherms displaced below that of the wild type (RRRbRb). The negative logarithmic relationship between seed longevity and seed storage moisture content (%, f.wt basis), determined in hermetic storage at 65 °C, also differed: longevity in the mutant near-isogenic lines was poorer and less sensitive to moisture content than in the wild type (i.e. CW was lower). The low-moisture-content limit (mc) to this relation also differed, being lower in the mutant near-isogenic lines (5.4–5.9%) than in the wild type (6.1%). In contrast, all four near-isogenic lines showed no difference (P >0.25) in the negative semi-logarithmic relationship between equilibrium relative humidity (ERH) and seed longevity. It is concluded that the effect of these alleles at the r and rb loci on seed longevity was largely indirect; a consequence of their effect on seed composition and hence on moisture sorption isotherms. However, this explanation could not be invoked at moisture contents below mc where differences in longevity remained substantial (RRRbRb double that of rrrbrb). Hence, these mutant alleles affected seed longevity directly at very low moisture contents.
Key words: Pisum sativum L., seed longevity, seed moisture content, isotherms, gene effects.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Gene effects on seed longevity have long been of interest (Lindstrom, 1942), but remain difficult to discern because pre-harvest (and post-harvest) environmental effects on seed longevity are substantial and make the identification of genetic effects problematic.
Mutants within pea (Pisum sativum L.) which induce wrinkling of the seed as a result of differences in seed composition have been identified and characterized for variation at the rugosus loci (Wang and Hedley, 1991; Wang et al., 1998). Studies with near-isogenic lines differing only in genes at the r and rb loci (Hedley et al., 1986; Wang and Hedley, 1991) have shown that compared with the wild type (RRRbRb) both the single mutants (RRrbrb or rrRbRb) and the double mutant (rrrbrb) have a lower starch content (Wang et al., 1990) and a higher oil content (Jones et al., 1990). Compositional differences are brought about by mutations in the starch biosynthetic pathway: the rr embryos lack one isoform of starch-branching enzyme (Smith, 1988), whilst the rbrb embryos have lower ADP glucose pyrophosphorylase activity (Hylton and Smith, 1992) and lack the large subunit of this enzyme (Martin and Smith, 1995).
Seed hygroscopic properties are a major determinant of relations between longevity and environment (Cromarty et al., 1982; Roberts and Ellis, 1989) and thus contribute towards differences in seed longevity among contrasting species (Ellis et al., 1989). Relations between seed moisture content and equilibrium relative humidity (ERH) at any one temperature, described as isotherms, are affected by the chemical composition of the seed (Pixton, 1967; Vertucci and Leopold, 1987a). Three zones exist within isotherms which relate to the three types of water in seeds: water bound at strong sites, at weak sites, and free water (Vertucci and Leopold, 1984, 1987a, b). Roberts and Ellis (1989) related the three types of water present within seeds to effects on seed longevity; the central zone extends from about 10–85% ERH within which there is a negative logarithmic relationship between seed moisture content and longevity. The slope of this relationship is CW in the viability equation:
where v is probit percentage viability after p d in storage at m% moisture content and t °C, Ki is a constant specific to the seed lot, and KE CW, CH, and CQ are species constants (Ellis and Roberts, 1980a, b).
At one constant temperature, the relation between longevity (
, standard deviation of seed deaths in time (d)) and seed moisture content becomes:
log10 = K–CW log10m(2)
where K=KE–CHt–CQt2. Absolute longevity (KE, or K at one temperature) and CW are known to vary among species (Ellis et al., 1988, 1989), but are believed to be constant within a species (Ellis et al., 1989). Although 10K is the extrapolated value of at 1% moisture content at the specified temperature, there is a lower limit to this negative logarithmic relation; below this critical moisture content (mc), a further reduction in moisture content does not improve longevity. Differences in mc among species can be considerable (Ellis et al., 1988, 1989), but these variant moisture content values provide similar equilibrium relative humidities. Similarly, there is a negative semi-logarithmic relationship between seed longevity and equilibrium relative humidity (Roberts and Ellis, 1989) and the slope of this relationship is similar among contrasting species (Ellis et al., 1989). This implies that variation in CW above mc is due to the hygroscopic properties of the seed.
Within the seed viability equation (equation 1) the environmentally-induced differences in seed quality that affect longevity are accounted for by differences in Ki. Hence, any genetic influence on seed longevity can be investigated by determining effects on the constants KE, CW, CH, and CQ. Given that there is good evidence that the relative effect of temperature on seed longevity (i.e. the constants CH and CQ) does not differ among contrasting species (Dickie et al., 1990), the above task can be simplified to determining effects on just K and CW at one temperature.
The use of pea near-isogenic lines in seed physiology has many advantages. In particular, it permits the effects of mutant alleles at different loci on seed longevity to be ascertained and quantified without interference from other genetic differences. Also, these near-isogenic lines enable variation among isotherms to be compared with variation in the sensitivity of longevity to moisture within a single, uniformly-produced species. Accordingly, the present study tests the hypotheses that variation at the rugosus loci in pea affects (a) seed isotherms, (b) the negative logarithmic relationship between seed longevity and moisture content, (c) the low-moisture-content limit to this relationship, and (d) the negative semi-logarithmic relationship between seed longevity and equilibrium relative humidity.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Seed of each of four near-isogenic lines of pea (Pisum sativum L.) backcrossed six times to the original parent (Hedley et al., 1994) differing only at genes at the r and rb loci were randomly selected from an original seed lot. All seeds had been produced in the same field under similar conditions in the same season and subsequently stored at about 4 °C with about 12% moisture content.
Starch content of seeds was determined in a three-stage process. First, 0.2±0.01 g of pea flour, taken from a 10 g ground sample, was treated with 15 ml of amylase solution according to Macrae and Armstrong (1968), converting starch to water-soluble carbohydrates. This was maintained at 55 °C for 1.5 h with regular mixing, and then filtered. Second, a 2.5 ml aliquot of the filtrate of total water-soluble carbohydrates was treated with 7.5 ml 0.1 M sulphuric acid and maintained at 70 °C for 30 min to produce a solution of reducing sugars (Smith et al., 1964; Smith and Groteluschen, 1966). Finally, these reducing sugars were assayed by the neucuproin method, measured at 460 nm (Fuller, 1967), from which starch content was determined. Lipid contents of seeds were determined by the Soxflo method (Brown and Mueller-Harvey, 1999).
A 220 g subsample of seeds of each isoline was obtained randomly from a 10 kg bulk by division and then further divided into two 10±0.5 g replicates for each relative humidity. These weighed samples were placed in desiccators above a range of saturated salt solutions at a constant temperature of 20 °C to determine isotherms. Saturated salt solutions (equilibrium relative humidities at 20 °C shown in parentheses) comprised ZnCl2 (5.5%), NaOH (7.5%), LiCl (13%), CH3COOK (25%), MgCl2 (32.8%), K2CO3 (43%), NaBr (57%), NaNO2 (65%), NaCl (76%), KCl (85%), and KNO3 (93%) (Winston and Bates, 1960; Weast et al., 1986). Seeds were removed weekly, reweighed and returned to the desiccators. When seeds reached constant weight (i.e. in equilibrium with the saturated salt solution), their moisture content was determined using the high-constant-temperature-oven method. The seeds were ground and two replicates of 4–5 g were dried at 130–133 °C for 2 h, after which they were cooled in a desiccator and reweighed to determine loss in weight on drying (ISTA, 1999a, b).
Seed moisture contents initially varied from 11.6–13.2% (Table 1). Therefore, the sorption isotherms represent desorption below and absorption above these values. Cubic models were fitted to the data using GENSTAT to quantify the isotherms. Differences among near-isogenic lines were determined by comparison of regressions.
|
From a second 1.75 kg sub-lot of each near-isogenic line, 250 g of seed was adjusted to each moisture content, within a range between 4% and 15% (f.wt basis), using silica gel or humidification above water at 20 °C. The duration of this period of adjustment depended upon target moisture content, the greatest being 70 d to 4%. Once at the desired moisture content, ten subsamples each of 100 seeds were hermetically sealed in laminated aluminium foil packets (Moore and Buckle, St Helens, UK; three layers comprising, 55 g m2 polyethylene beneath 7 g m2 aluminium, covered with 17 g m2 polyester), and stored in an incubator at 65 °C. This high temperature was necessary in order that seed survival curves could be obtained at low moisture contents within a comparatively short period.
Samples were removed from storage at intervals of 30 min to 14 d (depending upon seed moisture content). The seeds were then humidified above water at 20 °C for 20–24 h to avoid imbibition damage (Ellis and Roberts, 1982; Ellis et al., 1985). Two replicates of 50 humidified seeds were then tested for the ability to germinate by placing between paper towels (Kimberly-Clark) moistened with deionized water. These tests were maintained at 20 °C for up to 8 d as prescribed by the ISTA (1999a, b). However, this period was sometimes prolonged to up to 28 d in order to allow sufficient time for seedling development to occur. This was common among seeds which had been stored at low moisture contents where hardseededness was detected. Seeds were examined 5 d into the test and, where encountered, the testa of hard seeds was scarified using sand paper. Seedlings were evaluated for normal development and the results reported are for normal germination (ISTA, 1999a, b). Seed survival curves were fitted by probit analysis using GENSTAT.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Table 1 shows the reduced starch content and greater lipid contents of the seeds of the mutant near-isogenic lines used in the present experiments. The double mutant (rrrbrb) showed the greatest reduction in starch content compared with the wild type (RRRbRb), whilst the single mutant at the r locus (rrRbRb) provided the highest lipid content.
The isotherms were described well by cubic models (Table 2) and differed significantly among the four near-isogenic lines (F12,28=18.99; P <0.005). Isotherms of the three mutant near-isogenic lines were displaced below the wild type, except at relative humidities >85% where the isotherms crossed (Fig. 1). The double mutant (rrrbrb) and the single r mutant (rrRbRb) were displaced the furthest below the wild type and did not differ significantly from each other (F4,14=0.38; P >0.25), while that for the sole rb mutant (RRrbrb) was above (F4,25=3.66; P <0.025) this pair.
|
|
All near-isogenic lines showed high initial normal germination (Table 1) before experimental storage at 65 °C. The seed survival curves conformed to negative cumulative distributions, as described by equation (1). Differences in Ki among near-isogenic lines were detected (Table 3). However, even though the seeds of each near-isogenic line were produced in the same field in the same season, estimates of Ki are particularly sensitive to minor differences in seed production environment. Accordingly, more detailed investigation would be required before concluding that the differences in Ki (Table 3) are wholly genetic in origin.
|
It is valid, however, to compare the response of longevity (
) to seed storage moisture content (m) within equation (2) among the near-isogenic lines as pre- and post-harvest conditions do not affect the relationship between and m. Over most of the range of moisture contents studied, this negative logarithmic relationship (Fig. 2A) was significant (P <0.001). The slope, CW, varied significantly between the three mutant near-isogenic lines and the wild type (F1,16=6.20; P <0.025). However, no differences in CW were detected (F2,9=0.92; P >0.25) among the three mutant near-isogenic lines. Thus one slope was applied to all three mutant near-isogenic lines (Fig. 2A; Table 3). Whereas the estimates of at c. 15% moisture content were broadly similar among all four near-isogenic lines, differences among the estimates of K were detected within the three mutant lines (F2,11=6.73; P <0.025), as well as between this cohort and the wild type (F1,17=28.27; P <0.001).
|
Differences in the low-moisture-content limit (mc) to these relationships occurred between near-isogenic lines: mc was greatest in the wild type (6.1%) and lowest in RRrbrb (5.4%) (Table 3). Below mc, there was no significant effect of moisture content on longevity (P >0.25). However, significant differences (>2-fold) in longevity were detected below mc among the four near-isogenic lines (F3,4=36.00; P <0.005). The wild type provided the greatest longevity and the double mutant the poorest longevity at these moisture contents (Fig. 2A).
No difference (F3,19=1.29; P >0.25) was detected among the four near-isogenic lines in the negative semi-logarithmic relationship between longevity and equilibrium relative humidity (Fig. 2B). Observations at 5.5% relative humidity were excluded from this analysis because longevity was no greater than that at 7.5% relative humidity (i.e. 7.5% relative humidity was the lower limit to the negative semi-logarithmic relation between longevity and equilibrium relative humidity). Thus, whereas the analysis in Fig. 2A suggests that mc was in equilibrium with 9.5–13.8% RH at 20 °C (Table 3), the direct analysis in Fig. 2B indicates that seed longevity continued to increase with desiccation down to moisture contents in equilibrium with 7.5% RH at 20 °C.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The common relationship between longevity and ERH for all four near-isogenic lines above 7.5% relative humidity (Fig. 2B) suggests that the differences in longevity among near-isogenic lines (Fig. 2A) are a consequence of the differences in isotherms (Fig. 1). The isotherms result from the different seed compositions (Table 1) which, in turn, are a consequence of the variation at the rugosus loci. Lipid contents reported here for the wild type and the single mutants are consistent with previously published data (Jones et al., 1995). However, the lipid content determined for the double mutant is lower (2.8%) than previously reported (5.3%) (Jones et al., 1995). Further analyses showed consistency in these lower values. Thus the three mutant near-isogenic lines have greater lipid contents than the wild type with reduced longevity, at most moisture contents, and a reduced sensitivity of seed longevity to moisture content (Fig. 2A). Lipid content greatly affects the relationship between seed moisture content and equilibrium relative humidity (Cromarty et al., 1982). The effects of lipid content (Table 1) on the isotherms (Fig. 1) are clear. Generally, the greater the lipid content of a seed, the lower the moisture content at any one RH. Thus, the isotherms of the mutant near-isogenic lines are displaced below that of the wild type up to 78% RH. This equates to moisture contents up to about 15%, over which range their longevity is also shorter than the wild type (Fig. 2A). However, at 78% RH, where the isotherms for the mutant alleles cross that of the wild type (Fig. 1), longevity is least variant among all four near-isogenic lines (Fig. 2A).
The estimate of CW for the wild type (5.17, s.e. 0.18) is within the range of previous estimates for pea of 4.8–5.0 (Ellis and Roberts, 1982) and 5.39 (Ellis et al., 1989). Similarly, the estimate of mc for the wild type (6.1%) is very similar to that of 6.2% determined previously (Ellis et al., 1989). However, CW and mc are lower for the rugosus loci mutants (Table 3). Hence, although there is much evidence to support the original proposal that the seed viability constants KE, CW, CH, and CQ are invariant for a species (Ellis and Roberts, 1980a), the current results show that CW and K (and so by implication KE) do vary within pea as a result of variation at the rugosus loci. By contrast, all four near-isogenic lines showed a common negative semi-logarithmic relationship between longevity and equilibrium relative humidity (Fig. 2B). The slope of this relationship is equivalent to an approximate doubling of longevity for a reduction in equilibrium relative humidity of 10%. This is very similar to estimates for seeds of other species (Ellis et al., 1989).
Below mc, a further reduction in seed moisture content had no effect on longevity (Fig. 2A). This is consistent with earlier investigations with pea (Ellis et al., 1989). It has previously been suggested that the different estimates of mc for contrasting species are all equivalent to equilibrium relative humilities of about 10–12% at 20 °C (Ellis et al., 1989). While the estimates for these four near-isogenic lines of pea vary slightly more than this from 9.5% to 13.8% (Table 3), Fig. 2B demonstrates longevity improving down to moisture contents in equilibrium with about 7.5% relative humidity at 20 °C. Mean estimates of longevity at the two moisture contents below mc differed between near-isogenic lines, providing further evidence of the effect of variation at the rugosus loci on longevity. Whereas most of the differences in longevity caused by this genetic variation can be explained by compositional and so hygroscopic differences (compare Fig. 2A with Fig. 2B), this explanation cannot be invoked at the lowest moisture content studied (because it is within the zone where variation in moisture content has no further effect on longevity). Hence, the variation in below mc would appear to be a direct consequence of variation at the rugosus loci.
The findings here highlight the need for genetic variation to progress current knowledge of seed biology. Hitherto, the mutants used here have been used to understand seed development more fully (Hedley and Ambrose, 1980) and to gain a preliminary understanding of the genetic control of oligosaccharide synthesis (Jones et al., 1999). These mutants have been used here to further the knowledge of seed survival, and to demonstrate that single gene mutations can affect seed composition, which has significant effects on both the isotherms and the longevity of seed mutants.
|
Acknowledgements |
|---|
We thank the University of Reading Research Endowment Trust for a research studentship to TWL, Dr TD Hong for helpful discussions and the Analytical Laboratory for analytical support.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Brown RH, Mueller-Harvey I. 1999. Evaluation of the novel Soxflo technique for rapid extraction of crude fat in foods and animal feeds. Journal of AOAC International 88, 1369–1374.
Cromarty AS, Ellis RH, Roberts E.H. 1982. The design of seed storage facilities for genetic conservation. Rome: International Board for Plant Genetic Resources.
Dickie JB, Ellis RH, Kraak HL, Ryder K, Tompsett PB. 1990. Temperature and seed storage longevity. Annals of Botany 65, 197–204.
Ellis RH, Hong TD, Roberts EH. 1985. Handbook of seed technology for genebanks. Vol. 1. Principles and methodology. Rome: International Board for Plant Genetic Resources.
Ellis RH, Hong TD, Roberts EH. 1988. A low-moisture-content limit to logarithmic relations between seed moisture and longevity. Annals of Botany 61, 405–408.
Ellis RH, Hong TD, Roberts EH. 1989. A comparison of the low-moisture-content limit to the logarithmic relation between seed moisture and longevity in twelve species. Annals of Botany 63, 601–611.
Ellis RH, Roberts EH. 1980a. Improved equations for the prediction of seed longevity. Annals of Botany 45, 13–30.
Ellis RH, Roberts EH. 1980b. The influence of temperature and moisture on seed viability period in barley (Hordeum distichum L.) Annals of Botany 45, 31–37.
Ellis RH, Roberts EH. 1982. Desiccation, rehydration, germination, imbibition injury and longevity of pea seeds (Pisum sativum L.). Seed Science and Technology 10, 501–508.
Fuller KW. 1967. Automated determination of sugars. In: Automation in analytical chemistry. European Technicon Symposium Paris, France 1966 II.
Hedley CL, Ambrose MJ. 1980. An analysis of seed development in Pisum sativum. Annals of Botany 46, 89–105.
Hedley CL, Lloyd JR, Ambrose MJ, Wang TL. 1994. An analysis of seed development in Pisum sativum. XVII. The effect of the rb locus alone and in combination with r on the growth and development of the seed. Annals of Botany 74, 365–371.
Hedley CL, Smith CM, Ambrose MJ, Cook SK, Wang TL. 1986. An analysis of seed development in Pisum sativum. II. The effect of the r-locus on the growth and development of the seed. Annals of Botany 58, 371–379.
Hylton C, Smith AM. 1992. The rb mutation of peas causes structural and regulatory changes in ADP glucose pyrophosphorylase from developing embryos. Plant Physiology 99, 1626–1634.
ISTA (International Seed Testing Association). 1999a. International rules for seed testing. Rules 1999. Seed Science and Technology 27, Supplement.
ISTA (International Seed Testing Association). 1999b. International rules for seed testing. Annexes 1999. Seed Science and Technology 27, Supplement.
Jones DA, Arthur AE, Adams HA, Coxon DT, Wang TL, Hedley CL. 1990. An analysis of seed development in Pisum sativum. IX. Genetic analysis of lipid content. Plant Breeding 104, 144–151.[CrossRef]
Jones DA, Barber LM, Arthur AE, Hedley CL. 1995. An analysis of seed development in Pisum sativum L. XVI. Assessing variation for fatty-acid content by use of a non-destructive technique for single-seed analysis. Plant Breeding 114, 81–83.[CrossRef]
Jones DA, DuPont MS, Ambrose MJ, Frias J, Hedley CL. 1999. The discovery of compositional variation for the raffinose family of oligosaccharides in pea seeds. Seed Science Research 9, 305–310.
Lindstrom EW. 1942. Inheritance of seed longevity in maize inbreds and hybrids. Genetics 27, 154.
Macrae JC, Armstrong DS. 1968. Enzyme method for determination of á-linked glucose polymers in biological materials. Journal of the Science of Food and Agriculture 19, 578–581.[CrossRef]
Martin C, Smith AM. 1995. Starch biosynthesis. The Plant Cell 7, 971–985.[CrossRef][Web of Science][Medline]
Pixton SW. 1967. Moisture content—its significance and measurement in stored products. Journal of Stored Product Research 3, 35–47.
Roberts EH, Ellis RH. 1989. Water and seed survival. Annals of Botany 63, 39–52.
Smith AM. 1988. Major differences in isoforms of starch branching enzyme between developing embryos of round- and wrinkled-seeded peas (Pisum sativum L.). Planta 175, 270–279.[CrossRef]
Smith D, Groteluschen RD. 1966. Carbohydrates in grasses. I. Crop Science 6, 263–266.
Smith D, Paulson GM, Raguse CA. 1964. Extraction of total available carbohydrates from grass and legume tissue. Plant Physiology 39, 960–962.
Vertucci CW, Leopold AC. 1984. Bound water in soybean seed and its relation to respiration and imbibitional damage. Plant Physiology 75, 114–117.
Vertucci CW, Leopold AC. 1987a. Water binding in legume seeds. Plant Physiology 85, 224–231.
Vertucci CW, Leopold AC. 1987b. The relationship between water binding and desiccation tolerance in tissues. Plant Physiology 85, 232–238.
Wang TL, Bogracheva TY, Hedley CL. 1998. Starch: as simple as A, B, C? Journal of Experimental Botany 49, 481–502.
Wang TL, Hadavizideh A, Harwood A, Welham TJ, Harwood WA, Faulks R, Hedley CL. 1990. An analysis of seed development in Pisum sativum. XIII. The chemical induction of storage product mutants. Plant Breeding 105, 311–320.[CrossRef]
Wang TL, Hedley CL. 1991. Seed development in peas: knowing your three ‘r’s’ (or four, or five). Seed Science Research 1, 3–14.
Weast RC, Astle MJ, Beyer WH. 1986. CRC handbook of chemistry and physics, 67th edn. Boca Raton, Florida: CRC Press.
Winston PW, Bates DM. 1960. Saturated salt solutions for the control of humidity in biological research. Ecology 41, 232–237.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Ballesteros and C. Walters Water properties in fern spores: sorption characteristics relating to water affinity, glassy states, and storage stability J. Exp. Bot., March 1, 2007; 58(5): 1185 - 1196. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||