Planta (1990)180:420-428

pl~(_lt

~

9 Springer-Verlag1990

Nature of the effect of the r locus on the lipid content of embryos of peas (Pisum sativum L.) Mary Bettey and Alison M. Smith John Innes Institute and Institute of Plant ScienceResearch, ColneyLane, Norwich NR4 7UH, UK

Abstract. The aim of this work was to discover why pea (Pisum sativum L.) embryos recessive at the r locus (rr) have a higher lipid content than embryos dominant at this locus (RR). The r locus is a gene encoding starchbranching enzyme, rr embryos have a much lower activity of this enzyme than R R embryos, and hence a reduced rate of starch synthesis. The higher lipid content of rr embryos must be a consequence of this. We suggest that neither differences in the availability of substrate for lipid synthesis as a consequence of different rates of starch synthesis, nor differences in the capacity of the pathway for malonyl-CoA synthesis, account for the different lipid contents of R R and rr embryos. Lipid contents of the two sorts of embryo first diverge at a much later stage in development than divergence in starch content. Amounts of pyruvate and acetate, and activities of enzymes that convert triose phosphate to malonyl CoA are the same in the two sorts of embryo. Most of the lipid in developing embryos is polar, structural lipid, and polar lipid accounts for a large proportion of the difference in lipid content between the two sorts of embryo. This difference in structural-lipid content reflects considerable structural differences between the two sorts of embryo and is presumably the consequence of differences in rates of lipid turnover. Key words: Embryo (development, lipid content) - Lipid content (embryo) - Pisum (embryo lipid content, r locus)

Introduction The aim of this work was to discover how the r (rugosus) locus affects the lipid content of developing pea embryos. The r locus has profound effects on the morpholAbbreviations: DW = dry weight; FW = fresh weight; FAME= fat-

ty-acid methylesters

ogy and storage-product composition of pea seeds. At maturity, seeds containing embryos that are homozygous dominant (RR) or heterozygous (Rr) are round, whereas those containing embryos that are homozygous recessive (rr) are wrinkled. Embryos of wrinkled-seeded peas (referred to as wrinkled embryos) contain less starch than round embryos (30% as opposed to 55% of the final dry weight), and their starch is less branched (35% as opposed to 60%-70% amylopectin; Kellenbarger et al. 1951 ; Kooistra 1962). Wrinkled seeds also have a higher sugar content (10% as opposed to 6% final dry weight, Kooistra 1962), and less legumin (a storage globulin) than round seeds (Davies 1980). During development wrinkled embryos have a higher osmotic pressure, water content, and abscisic acid content than round embryos (Hedley et al. 1986a; Wang et al. 1987). Seed lipid content is affected by the r locus. Round embryos have a lipid content of 2%-3% of the final dry weight, whereas in wrinkled embryos it is 4%-5% (Coxon and Davies 1982). Recent work has shown that a gene encoding an isoform of starch-branching enzyme is located at the r locus. In rr lines of peas an insertion into this gene prevents the production of this isoform (Bhattacharyya et al., in press). As a result wrinkled embryos have reduced activity of starch-branching enzyme. All of the effects of the r locus are highly likely to be consequences of this. Some of these effects are readily explicable. A lower activity of the enzyme could lead to less branched starch, a lower overall starch content, accumulation of sucrose, and hence a higher osmotic pressure (Smith 1988; Bhattacharyya et al., in press). This in turn leads to greater water uptake during development, and a larger embryo volume in the mutant embryo. On drying out, the larger loss of volume of the mutant embryo leads to wrinkling of the testa (Wang et al. 1987; Kooistra 1962). It is less obvious how the lack of an isoform of starch-branching enzyme leads to an increased lipid content in wrinkled embryos. The difference in lipid con-

M. Bettey and Alison M. Smith: The r locus and the lipid content of pea embryos

tents between mature round and wrinkled seeds appears to be a consequence of different amounts of both neutral and polar lipids (Coxon and Davies 1982). The nature of the allele at the r locus could affect the lipid content of embryos either through a direct effect on the capacity for lipid synthesis, or through an indirect effect on, for example, the rate of lipid turnover. We suggest two ways in which direct effects might be brought about. First there might be an effect on the availability of substrates for fatty-acid synthesis. Most of the metabolites on the pathway of starch synthesis from sucrose accumulate in developing wrinkled embryos to higher levels than in developing round embryos (Edwards and ap Rees 1986; Edwards et al. 1988). This is probably a consequence of the reduced rate of starch synthesis in wrinkled embryos (Edwards etal. 1988; Smith 1988), and may mean that they have more carbon from sucrose available for synthesis of lipid than round embryos. Second there might be some effect on the maximum catalytic activities of important regulatory enzymes of lipid synthesis. Carboxylation of acetyl CoA has frequently been proposed as the key regulatory step in lipid synthesis. Activity of the enzyme in oil seeds is comparable with net rates of lipid synthesis, and peaks when maximum rates of lipid synthesis are achieved (Turnham and Northcote 1983; Simcox etal. 1979; Charles et al. 1986). Differences in the activity of this enzyme between round and wrinkled embryos might lead to a difference in the rate of lipid synthesis. To discover which of these possible effects of the r locus is responsible for the difference in lipid content between round and wrinkled embryos, we made the following measurements. 1) Lipid content of embryos during development. 2) Amounts of putative substrates of acetyl-CoA synthesis in round and wrinkled embryos during development. Pyruvate and acetate could both be immediate precursors of fatty-acid synthesis. 3) Maximum catalytic activities of glycolytic enzymes and acetyl-CoA carboxylase, which convert triose phosphate to malonyl CoA. Because of the difficulty of making reliable estimates of the activity of plastidial pyruvate-dehydrogenase complex (Denyer and Smith 1988) we have not attempted to measure this enzyme. 4) Amounts of neutral and polar lipid and of fatty acids in round and wrinkled embryos during development.

Material and methods

Plant material and growth of plants The BC1/7 lines of pea (Pisum sativum L.; John Innes germplasm collection) derived from JI 430 (Hedley et al. 1986b), which are essentially isogenic except at the r locus, were used. Plants were grown in a greenhouse according to Smith (1988). Measurements of metabolites, enzymes and total lipid content measured on a fresh-weight (FW) basis were made on embryos from a single batch of plants. Measurements of lipid classes and of total lipid content measured on a dry-weight (DW) basis were made on separate batches grown under the same conditions. Side shoots and basal

421

branches were removed leaving one main stem per plant. When three nodes had flowered the apex was removed. Buds were removed to allow only one flower at each of the three nodes. For enzyme and metabolite assays one seed from each end of the pod was rejected; for lipid assays seeds were used provided that they fell within the correct size class. Embryos were dissected from the testas and weighed before use. Wheat (Triticum aestivum L. cv Maris Huntsman) was grown according to Boffey et al. (1980).

Measurement of total lipid content Lipid content was determined gravimetrically by the method of Denyer (1987) which is a modification of the method of Coxon and Wright (1985). Measurements were made both on fresh peas and on dried pea flour produced by freeze-drying peas and then milling them in a miniature ball-mill. Non-lipid contaminants were removed using the method of Folch et al. (1957) with two washes in HCl:saturated aqueous NaC1 (1:1, v/v), and the third wash with distilled water. A blank was used in which the whole process of extraction and washing was performed in the absence of tissue. This gave a small but significant residue which was taken into account in the calculations. Determination of total lipid as fatty-acid methyl esters (FAME) was done by a modification of the method of Morrison et al. (1980) (Coxon and Wright 1985).

Metabolite measurements Embryos were placed on a polythene sheet then freeze-clamped between two aluminium blocks cooled to the temperature of liquid nitrogen (ap Rees et al. 1977), within 10 rain of removal from the plant and 2 rain of removal from the pod. The freeze-clamped material was added to 1 ml of 1 M perchloric acid, 1 mM ethylendiaminetetraacetic acid (EDTA), agitated with a glass rod and left on ice for 30 rain, then centrifuged (27000.g, 10 min, 4 ~ C) and the supernatant decanted. The pellet was washed twice by resuspension and centrifugation in 0.5 ml of 1 M perchloric acid, 1 mM EDTA. The supernatants were combined and brought to between pH 7.0 and pH 8.0 with 5 M potassium carbonate. Potassium perchlorate was sedimented by centrifugation (5000.g, 10min). The pellet from this was washed by resuspension and centrifugation in neutralized perchlorate. The supernatants from these two centrifugations were combined and used in metabolite assays. Extracts were kept at 2 ~ C throughout extraction. Assays were carried out spectrophotometrically at 25 ~ C and 340 nm in a final volume of 1 ml.

Acetate: according to Beutler (1984). The calculation of acetate content was according to Bergmeyer (1983), using the method for preceeding indicator reactions.

Other metabolites: according to Lowry and Passonneau (1972). For the triose-phosphate assay an excess of triose-phosphate isomerase was added, and the assay for dihydroxyacetonephosphate used.

Measurement of enzyme activities Acetyl-CoA carboxylase (EC 6.4.1.2) (Nikolau et al. 1981). Embryos (0.5-1 g FW of tissue) were homogenized using a pestle and mortar followed by an all-glass homogenizer in 2 ml of 100 m M N,N-bis(2-hydroxyethyl)glycine (Bicine; pH 7.7); 5 mM MgC12; 100 mM KC1; 5 mM dithiothreitol (DTT); 3 mM MnCI2; 0.2%

422

M. Bettey and Alison M. Smith: The r locus and the lipid content of pea embryos

(v/v) Triton X-100; and 2 mg.m1-1 polyvinylpolypyrrolidone, at 4 ~ C. The extract was centrifuged (27000-g, 10 min, 4 ~ C), desalted on a Sephadex G25 (coarse) column (10 cm long, 0.6 cm i.d.) equilibrated with extraction medium, and stored on ice. The assay contained, in a final volume of 300 Ixl: 50 mM Bicine (pH 7.7); 2.5 mM MgC12; 50 mM KC1; 2.5 mM D T T ; 1.5 mM MnC12; 2 mg.m1-1 bovine serum albumin (BSA); 3 3 m M NaH14CO3 (3.7GBqmol-1; Amersham plc, Amersham, Bucks., UK); 0.6 mM ATP; 0.1 mM acetyl CoA; 50 pl extract. The assay was performed as described by Nikolau et al. (1981), except that incubation was at 25 ~ C for 30 min.

14

A

12'

10 84

z:

8

.'9- 6

Other enzymes. Embryos (0.5-1.5 g FW of tissue) were homogenized as above in 2-3 ml of 100 mM 3-(N-morpholino)propanesulfonic acid (Mops; pH 7.2); 5 mM DTT; 10% (v/v) ethanediol; and 5 mM MgC12 at 4 ~ C. The extract was centrifuged (27000-g, 20 rain, 4 ~ C) and the supernatant stored on ice. Assays were carried out spectrophotometrically at 25~ and 340 nm in a final volume of 1 ml. Optimized assays for developing embryos of this line of wrinkled peas were used (Denyer and Smith 1988). This optimization was checked for round embryos and any differences are given below. Phosphoglycerate kinase (EC 2.7.3.2): pH 8.5 for round-pea extracts. Phosphoglycerate mutase (EC 2.7.5.3): pH 7.8 and 12 mM 3-phosphoglycerate for round-pea extracts. Enolase (EC 4.2.1.11): pH 8.2 for round-pea extracts.

o. m

E

4 84

50-1()0 150-200250-300350-400450"500550-'600 650:700 Fresh weight embryo

(rag)

Separation and measurement of neutral and polar lipids Silica columns were pasteur pipettes plugged with degreased cotton wool, containing about 0.9 g CC4 silica (Mallinckrodt Inc., Paris, Ky., USA). Embryos were weighed, freeze-dried, and milled in a miniature ball-mill. Lipid was extracted as described above to give a washed lipid fraction. A 4-ml sample was used for gravimetric determination of lipid. A second was applied to the silica column for lipid-class separation using a modification of the method of Rouser et al. (1967), as follows. The sample was evaporated under nitrogen, then taken up in 0.5 ml chloroform and loaded onto the column which was previously primed with 1 ml of chloroform. Lipid classes were eluted as follows: 10 ml chloroform to elute neutral lipids; 35 ml acetone followed by 15 ml methanol to elute polar lipids (glycolipids and phospholipids). The fractions from the column were collected and evaporated under a vacuum at 25 ~ C. The residue from each fraction was taken up in 1 ml of the appropriate solvent (chloroform for the neutral-lipid fraction, methanol for the polar-lipid fraction). Samples were placed in preweighed vials, allowed to evaporate overnight, and reweighed.

Results Lipid contents of developing embryos

We checked that lipid was not lost during extraction and washing procedures. Recovery through killing and extraction procedures of purified tripalmitin (similar in amount to the lipid expected in the tissue) was 92%. Prior to the drying-out stage of development, lipid content expressed on an FW basis increased with increasing embryo weight (Fig. 1 A). There were no statistically significant differences between round and wrinkled embryos of the same FW. There were, however, differences in lipid content between round and wrinkled embryos when data were expressed on a DW or embryo basis. Fig. 1 B, C shows data from Fig. 1 A replotted on

3= ,..~ 30. t "~20 O~ E 10"

4'0

1~;o

8b

Dry weight embryo

16o

200

16o

200

(mg)

C

To6 E

4

I

io

do

&o

Dry weight embryo (rag)

Fig. 1 A--C. Lipid contents of developing pea embryos. A gravimetric measurements made on an FW basis. Points are means of three measurements each made on a separate sample of pea embryos. Standard errors are indicated by bars. B, C Values from A replotted using measurements of FW: DW ratios for embryos of these lines of peas from Hedley et al. (1986b). o-----o, round embryos; o--o, wrinkled embryos; *, means for round and wrinkled are significantly different at the 5% level but not at the 1% level, when assessed using a t-test

M. Bettey and Alison M. Smith: The r locus and the lipid content of pea embryos

wrinkled embryos always contained more lipid than round embryos. This was confirmed by direct measurements of lipid content on a DW basis made on two further independently grown batches of peas by gravimetric and FAME methods (Table 1). Lipid contents were the same in round and wrinkled embryos at 40 mg DW but 50%-80% higher in wrinkled than round embryos at 80-90 and 140 mg DW. The FAME measurements of crude plant lipids are always lower than gravimetric measurements because the FAME method measures fatty acids only, whereas the gravimetric method measures total lipids (Coxon and Wright 1985). Detailed comparisons of lipid contents of round and wrinkled embryos during the drying-out stage of development were difficult because changes in FW: DW ratios were different in the two sorts of embryo. Gravimetric measurements made at the start of, during, and after this stage indicated that lipid contents increased slightly on an embryo basis but did not alter on a DW basis (data not shown). A large difference in lipid content between round and wrinkled embryos was maintained through this stage of development. Wrinkled embryos contained 70%-110~ more lipid than round embryos.

Table 1. Lipid contents of developing round (RR) and wrinkled (rr) pea embryos, measured by gravimetric and F A M E methods. Values are the means of measurements made on three separate samples of pea embryos (gravimetric measurements) or pea flour (FAME measurements), and are expressed as mean • SE Method

Lipid content (mg' (g D W ) - l)

Genotype

Embryo DW (mg)

Gravimetric FAME

RR rr RR rr

40

80-90

140

25.9• 29.8• 19.0• 19.5•

27.4• 45.6• 19.5• 28.4•

22.6• 41.3• 19.0• 28.6•

423

these bases, using measurements of the F W : D W ratios of embryos of these lines of peas from Hedley et al. (1986b). Lipid content increased in both sorts of embryo during development on an embryo basis, but changed relatively little on a DW basis. Lipid contents were the same in young round and wrinkled embryos, but they diverged at about 40 mg embryo DW. After this point,

~ 160 ,2d

320' "o

o ~ 240

.o o= ~

160

~-~ 80 ~50-160

200-250

350-400

500-550

Embryo fresh wt

.

.

.

.

.

50-100 200-250 350-400 500-550 650-700 Embryo fresh wt (mg)

(mg)

t

i160 I

~12 0-~

~"-120

7o

to)

80.

~ 5o >

v

650-760

40

40.

50-100 200-250 350-400 500-550 650-700 Embryo fresh wt

(mg)

Fig. 2. Metabolite contents of developing pea embryos. Triose phosphate, 3-phospboglyceric acid, pyruvate, and acetate were measured in extracts of developing embryos of round- ( o - - o ) and wrinkled- ( e - - e ) seeded peas. Extraction and assay were as described in Material and methods. Each point is a mean of at least four measurements each made on a separate extract. Bars

50-'100 200"250 350-'400 500"550 650~700 Embryo fresh wt (mg) indicate standard error. At each stage of development studied, the difference between the means for round and wrinkled peas was assessed using a t-test. * indicates that the means are significantly different at the 5% level, but not at the 1% level, o indicates that means are significantly different at the 1% level

424

M. Betteyand AlisonM. Smith: The r locus and the lipid content of pea embryos

Amounts of metabolites in developing embryos We checked that our measurements are likely to be accurate reflections of the steady-state amounts of metabolites in the tissue in vivo. Duplicate samples of cotyledons were prepared, freeze-clamped (ap Rees et al. 1977), and a measured amount of purified metabolite similar to the amount found in the tissue was added to one of the samples before killing and extraction. Recovery of added metabolite was calculated from the amounts in the two extracts. Recoveries of metabolites were within 15% of that expected, except for acetate where both the endogenous amount and the recovery were quite variable. Recoveries of acetate were 111 + 4% in round-pea extracts and 151 +19% in extracts from wrinkled peas (mean _+SE, n = 3). This may be the result of the difficulty of obtaining good replicate samples in recovery experiments. Amounts of pyruvate and acetate were not significantly different between round and wrinkled embryos at all but the latest stages of development (Fig. 2). Triose phosphate was consistently higher, and 3-phosphoglyceric acid consistently lower, in wrinkled than in round embryos. For both metabolites, differences between embryos were statistically significant at two stages of development.

Activities of enzymes in developing embryos Our measurements of enzyme activity are likely to be accurate reflections of the maximum catalytic activities in the tissue since the concentration of each component and the pH of the assay were optmised to give the maximum rate. For all enzymes the activity measured was dependent upon the presence of the appropriate substrates and cofactors. In a mixed sample where two tissues (e.g. round and wrinkled embryo) had been coextracted, the recovery of activity was close to that predicted from measurements on the separate tissues. (i) For phosphoglycerate kinase, phosphoglycerate mutase and enolase, recoveries between 99 and 115% were obtained with a mixed extract of round and wrinkled embryos. (ii) For pyruvate kinase this type of mixed extract appeared to have stimulated activity, as two separate experiments gave recoveries of 151 and 135%. As a further check a measured amount of purified pyruvate kinase, similar to the amount expected in the tissue was added to one of two duplicate samples before extraction. The recovery was 106% for round embryos, and 115% for wrinkled. (iii) For acetyl-CoA carboxylase the recovery obtained when round and wrinkled embryos were coextracted was 127%. Coextraction of a 7-d-old wheat leaf and a wrinkled embryo gave a recovery of 101%. The activity obtained from the wheat leaf alone was very similar to that found by Hawke and Leech (1987) for this tissue. We believe that our extraction and assay methods for this enzyme are reliable. The maximum catalytic activities of the enzymes differed very little between round and wrinkled embryos

(Fig. 3). General trends were similar for both lines, and none of the enzymes showed a consistently higher activity throughout development in one type of embryo than in the other. Activity of acetyl-CoA carboxylase was very low compared with that of the other enzymes on the pathway.

Separation of lipid and fatty-acid classes from pea embryos 1) Neutral andpolar lipids. We wished to determine what proportion of the lipid in pea embryos is polar (membranes etc.), and what proportion is neutral (mainly storage lipid). Using a silica column to separate these two classes, 93% of added trilinolein (a neutral lipid) was recovered in the neutral-lipid fraction, and 94% of added phosphatidylcholine (a polar, phospholipid) and 78% of added digalactosyl diglyceride (a glycolipid)) were recovered in the polar-lipid fraction. More than 75% of the pea lipid loaded onto the column was recovered in the two fractions, except for the youngest embryo sizeclass where results were more variable and recovery was greater than 70%. Amounts of both sorts of lipid increased on an embryo basis with increasing embryo weight. Prior to the drying-out stage of development, round and wrinkled embryos contained considerably more polar than neutral lipid (Fig. 4). There was little difference between round and wrinkled embryos at 40 mg, but at 80 and 140 mg embryo DW wrinkled embryos contained more of both neutral and polar lipids. The difference in polar-lipid contents of the two sorts of embryo accounted for about 75% of the difference in total lipid content at 140 mg embryo DW. During the drying-out stage of development, the difference between the contents of neutral and polar lipids became much less marked, so that at maturity embryos contained roughly equal amounts of neutral and polar lipid (Fig. 4). This change during drying out appears to be because the amount of neutral lipid increased considerably at the onset of drying out, whereas the amount of polar lipid changed much less. The difference in total lipid content between the two sorts of embryo could not consistently be attributed primarily to one class of lipid or the other. Thus, the major contribution to the difference was made by polar lipid at 140 mg embryo DW, but by neutral lipid during drying-out. 2) Fatty-acid composition. There was little change in the fatty-acid composition of embryos between 40 and 140 mg DW, and the compositions of round and wrinkled embryos were very similar (Table 2). The data indicate that significant changes in composition occurred immediately prior to and-or during the drying-out stage of development. Mature seeds had a higher proportion of 18:0 and 18:1 fatty acids, and a lower proportion of 16:0 fatty acid, than developing embryos up to 140 mg DW.

M. Bettey and Alison M. Smith: The r locus and the lipid content of pea embryos

f

r~

425

I

6-1

E

4O

.__> "~ 3 0 0=

T= "6

=

E

|

20

"k

g' lo o.

--,

,

,

50-100

200-250

,

350-400

,

500-550

,

650-700

Embryo f r e s h w t ( m g )

50-100

200-250

350-400

500-550

650-700

E m b r y o f r e s h wt ( m g ) 6"

IC

5>, .>

m

o E ;z

4

>, ._>

3

"6 e~

.~3E~

== r

2

~'~

Jc o.

50-100 50-100

200-250

350-400

Embryo fresh wt

500-550

650-700

200-250

350-400

500-550

650-700

Embryo f r e s h w t ( m g )

(mg)

2f :=_ 10"

O

E "6

8 .>_

*6

6"

m

4.

~o

8

2,

Nature of the effect of the r locus on the lipid content of embryos of peas (Pisum sativum L.).

The aim of this work was to discover why pea (Pisum sativum L.) embryos recessive at the r locus (rr) have a higher lipid content than embryos dominan...
879KB Sizes 0 Downloads 0 Views