ARCHIVES

OF BIOCHEMISTRY

AND BIOPHYSICS

Vol. 297, No. 1, August 15, pp. 92-100, 1992

The Influence of Endogenous Acyl-Acyl Carrier Protein Concentrations on Fatty Acid Compositions of Chloroplast Glycerolipids Grattan Roughan*,’

and Tomoaki

Matsuot

*DSIR Fruit and Trees, Mt Albert Research Centre, Private Bag, Auckland, New Zealand and TDepartment of Horticulture, Kagoshima University, Kagoshima, 880 Japan

Received October 24, 1991, and in revised form April

15, 1992

The concentrations of long-chain acyl-acyl carrier proteins (acyl-ACP) occurring during fatty acid synthesis from [ 1-14C]acetate within chloroplasts isolated from spinach, pea, and amaranthus leaves were manipulated by making minor changes to a basal incubation medium containing sn-glycerol 3-phosphate (G3P). Pools of and palmitoyl-ACP were compared oleoyl- , stearoyl-, with those of the corresponding 1-acyl glycerol 3-phosphates to determine how endogenous acyl-ACP concentrations affected the fatty acid compositions of chloroplast glycerolipids. .The 1-acyl G3P synthesized by isolated chloroplasts contained more palmitate than would be expected for the precursor of thylakoid phosphatidylglycerol in the different plant species. However, treatments which increased ratios of oleoyl- to palmitoylACP by about 50% increased synthesis of sn-1-oleoyl G3P to the extent anticipated from known fatty acid compositions of the different phosphatidylglycerols. Since stearate constituted 70-73% of the acyl-ACP and 4851% of the I-acyl-G3P pool of spinach and pea chloroplasts incubated in the presence of cyanide, it is transferred to G3P much more efficiently in situ than would be predicted from competition studies using mixtures of acyl donors and purified acyltransferases. Increasing concentrations of G3P in incubation media from 0.1 to 2 mM had relatively little effect on the amounts and proportions of acyl-ACPs but forced the synthesis of palmitoyl-G3P and, ultimately, disaturated glycerolipid. It is concluded that the chloroplast G3P acyltransferases are primarily responsible for determining the fatty acid compositions of procaryotic glycerolipids in plants, but that acyl-ACP concentrations may play a more important role than would be anticipated from the kinetics of the purified enzyme. However, those kinetics may be quite complex; allosteric effecters may influence the affinities of the enzyme for oleoyl-ACP and for G3P. o ms2Academic Press,

92

Inc.

It is by now widely held that the fatty acid compositions of procaryotic types of plant lipids are controlled first by the acyl specificities of the glycerol 3-phosphate (G3P)2 acyltransferase and the 1-acyl G3P acyltransferase within chloroplasts, and second, by a somewhat selective desaturation of oleate and palmitate depending upon the nature of the glycerolipid into which the fatty acids finally become incorporated (1). Palmitoyl-, stearoyl-, and oleoylacyl carrier protein (ACP) are the acyl donors within isolated chloroplasts (2,3), and the G3P acyltransferase isolated from pea and spinach preferentially transferred oleate to G3P (4) when presented with a mixture of donors. The selectivity of the reaction was deemed sufficient to account for oleate dominating the fatty acids at the sn1 position of, e.g., phosphatidylglycerol (PG) in those plant species. On the other hand, the sn-2 position of procaryotic plant glycerolipids contains Cl6 fatty acids alone. In a number of chilling-sensitive plants, a lack of a similar selectivity of the G3P acyltransferase reaction for oleoyl-ACP (5) is thought to account for the presence of “disaturated” molecular species of PG (6, 7). These high-melting, disaturated lipids initiate thermotropic phase separations within thylakoids at physiological temperatures (8) and may thus be responsible for conferring chilling sensitivity upon such plants. Although the fatty acid composition of the sn-1 position of, e.g., PG from a given plant species may be strictly controlled in uiuo (7), there are indications that glycerolipids synthesized by isolated chloroplasts are not as rigorously regimented. For example, in spite of the high selectivity of the purified G3P acyltransferase toward oleoylACP (4), spinach chloroplasts may synthesize significant 1 To whom correspondence should be addressed. ’ Abbreviations used: ACP, acyl carrier protein; chl, chlorophyll; CoA, coenzyme A; DAG, sn-1,2-diacylglycero1; FAME, fatty acid methyl esters; G3P, sn-glycerol3-phosphate; LPA, lyso-phosphatidate; PA, phosphatidate; PG, sn+phosphatidylglycerol; UFA, unesterified fatty acids. 0003-9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

ACYL-ACYL

CARRIER

PROTEIN

AND

amounts of disaturated glycerolipids in the presence of increasing concentrations of exogenous G3P (9,lO). Also, apart from proportions of long-chain acyl-ACPs in illuminated spinach leaves (ll), nothing is currently known of the concentrations of the different acyl donors within chloroplasts in vivo nor how these might affect the fatty acid compositions of glycerolipids. It seemed opportune therefore, to determine whether the acyl specificity of the chloroplast G3P acyl transferase reaction might be manipulated in isolated, intact chloroplasts; a system which is a compromise between that which is truly in vivo and that which is truly in vitro. We have used conditions known to effect either acyl-ACP concentrations during fatty acid synthesis or the proportions of lipid classes synthesized from acetate, and have measured how endogenous concentrations of acyl-ACPs influence the fatty acid composition of 1-acyl G3P (lyso-phosphatidate, LPA). METHODS

AND MATERIALS

Sodium [ l-‘4C]acetate, 54 Ci/mol, was from Amersham, coenzyme A (CoA) and sn-glycerol 3-phosphate were from Sigma and Percoll was from Pharmacia. Spinach (Spinacia olearacea), pea (Pisum satiuun), and amaranthus (Amaranthus lividus) plants were grown under a controlled environment as before (12, 13), and about 15 g of rapidly expanding leaves was harvested for chloroplast isolation using the highEDTA method (14) for amaranthus, and low ionic strength media (15) for spinach and pea. Spinach and pea chloroplasts were initially pelleted through a cushion of 40% (v/v) Percoll(16) in low ionic strength buffer, then washed. All chloroplasts were incubated with illumination in 0.25 ml of basal buffer (12, 13) containing 0.22 mM [1-i4C]acetate (3 pCi) plus the additions shown in the legends to tables and figures. Potassium cyanide and sodium sulfite stock solutions were prepared immediately before use. Reactions were stopped at the times indicated by adding 0.25 ml of propan-2-01 containing 5% (v/v) glacial acetic acid, and an acyl-ACP fraction was recovered by coprecipitation with ammonium sulfate following the addition of methanol/chloroform and was analyzed as in (3). Lipids were recovered in the supernatant as described (3). From separate aliquots of the lipid extract, unesterified fatty acids (UFA), 1,2-diacylglycerols (DAG), and total polar lipids were separated on thin layers of 5% (w/w) boric acid in silica gel G, while LPA, PG, and phosphatidate (PA) were separated on thin layers of 5% NaHSOl (w/w) in silica gel G (17). The former were detected using iodine vapor, and the latter were localized by autoradiography, prior to scintillation counting. To analyze fatty acid labeling within lipid classes, about one-half of the total sample was concentrated, mixed with reference LPA and PA, and then separated semipreparatively on thin layers of 0.5% (w/w) EDTA in silica gel H developed with chloroform/methanol/ammonia (13/7/l, v/v). LPA and PA migrated as compact bands with I?, 0.05 and 0.14 [see also Ref. (IS)], respectively, and were well separated from other labeled lipids. Following detection with iodine vapor, the LPA band was recovered from the chromatogram and treated directly with 0.5 ml of 0.5 M NaOCHa in methanol while PA was eluted from the adsorbent into chloroform/methanol/O.25 M HC1(5/5/4, v/v). One-half of the eluted PA was transmethylated using NaOCH, and the other half was treated with diazomethane to form dimethylphosphatidate, which was then resolved into molecular species on thin layers of 5% (w/w) AgNOa in silica gel G developed with 5% (v/v) acetone in chloroform. Separated bands were detected by autoradiography. UFA, along with other nonpolar lipids, were eluted from the adsorbent using acetone/petroleum ether/water (5/5/4, v/v) containing about 1% (v/v) glacial acetic acid, and were treated with diazomethane. Fatty acid methyl esters (FAME) prepared by the above procedures were separated into saturated and monoenoic fractions on thin layers of 5% AgNOB in silica gel G. The monoenoics

CHLOROPLAST

LIPID

93

SYNTHESIS

were deemed to be oleate alone and were transferred directly to scintillation fluid, while the saturated FAME were eluted from the adsorbent for separation into stearate and palmitate by reverse-phase TLC (3) prior to scintillation counting.

RESULTS AND DISCUSSION Stability

of Acyl-ACP

Concentrations

A reaction time of 10 min was considered appropriate for accumulating sufficient label within LPA to allow reliable measurement of constituent fatty acids, and for ensuring that the composition of labeled fatty acids within PA was not unduly biased by the presence of endogenous 1-acyl G3P at the beginning of the incubation (19). However, it had already been shown (3) that acyl-ACP concentrations within amaranthus chloroplasts could change significantly with time of incubation, and this would complicate interpretation of the results. Therefore, acetate incorporations into total lipids and into acyl-ACPs were followed over a lo-min incubation of chloroplasts in basal media with and without G3P. The results (Fig. 1) indicated that, while fatty acid synthesis was essentially linear, acyl-ACP concentrations were reasonably constant from 4 to 10 min, at least in the presence of G3P. Therefore, it seemed likely that after incubating chloroplasts for 10 min the fatty acid compositions of the different lipid classes would reflect the composition of the acylACP pool measured at that time. It should be noted that 0.4 mM G3P had no effect on the concentrations of acyl-

loof----lA

“0 2 4 6 8 10 TIME (mid

0 2 4 6 8 10 TIME (mid

0 2 4 6 8 10 TIME (mln)

FIG. 1. Time-courses of acyl-ACP concentrations (A-C), and of lipid synthesis from acetate (D-F) by spinach (A, D), pea (B, E), and amaranthus (C!, F) chloroplasts. Isolated chloroplasts were incubated in basal media containing 0.22 mM [1-i4C]acetate (3 pCi) and in both the absence (open symbols) and the presence (closed symbols) of 0.4 mM glycerol 3phosphate. Here, and in Figs. 2 and 3, each data point is a mean value from two separate chloroplast preparations, and acyl-ACPs and lipids were analyzed as in (3). In A-C A, oleoyl-ACP; 0, stearoyl-ACP; and 0, palmitoyl-ACP and in D-F 0, unesterified fatty acids and 0, glycerides; chl, chlorophyll.

94

ROUGHAN

AND TABLE

MATSUO I

Effect of Various Additions to Incubation Media upon the Incorporation of Acetate into the Different Lipid Classes of Isolated Chloroplasts Additions Plant

Lipid

Nil

KCN

Na2S03

MgClz

(nmol acetate incorporated/mg Spinach

Pea

Amaranthus

UFA DAG LPA PA PG Other

CoA

Triton

chl)

69” 96 10 18 10 18

54 68 16 13 21 18

79 84 11 12 7 13

49 50 10 20 11 15

86 114 12 23 12 21

76 139 17 33 4 24

Total

221

190

206

155

268

293

UFA DAG LPA PA PG Other

136 7 8 36 2 11

75 4 10 16 1 9

60 3 3 10 1 3

119 6 9 29 1 14

166 8 9 31 2 15

142 10 12 41 1 22

Total

199

115

80

178

231

228

UFA DAG LPA PA PG Other

158 16 10 41 14 13

56 5 3 7 6 11

96 10 8 19 6 14

121 12 7 30 10 12

177 21 11 48 11 10

159 25 15 50 5 8

Total

252

88

153

192

278

262

a Values are means from two chloroplast preparations. Reactions (0.25 ml) contained 0.22 mM [1-r4C]acetate (3 PCi), 0.4 mM sn-glycerol 3phosphate, and the additions shown and were started by addition of chloroplasts equivalent to about 50 pg chlorophyll. After incubation in the light for 10 min at 25”C, reactions were stopped by addition of 0.25 ml 5% (v/v) acetic acid in propan-2-01. Lipids were recovered from the washings following coprecipitation of the acyl-ACP (3). KCN, 0.25 mM; Na,SOs, 2 mM; MgClr, 2 mM; CoA, 0.5 mM; Triton X-100, 0.13 mM. chl, chlorophyll.

ACPs in pea chloroplasts and affected only stearoyl-ACP in amaranthus chloroplasts in these particular experiments (Fig. 1). It appears likely that the ability of added G3P to depress acyl-ACP concentrations in pea chloroplasts, as reported previously (3), was negated in the present work by higher rates of acyl-ACP synthesis. The Effects of Treatments Concentrations

to Modulate Acyl-ACP

In the first set of ex(i) On lipid classes synthesized. periments, chloroplasts were incubated with G3P and in the presence of compounds known to affect acyl-ACP concentrations in isolated spinach chloroplasts. Both CoA and Triton X-100 increase (2), while MgCl, decreases (Sol1 and Roughan, unpublished results), acyl-ACP concentrations. In addition, since low concentrations of cyanide inhibit stearate desaturation without affecting total fatty acid synthesis (20), and low concentrations of sulfite

inhibit incorporation of fatty acids into spinach chloroplast glycerides without affecting total fatty acid synthesis (Roughan, unpublished result), both cyanide and sulfite should also alter acyl-ACP concentrations in isolated chloroplasts. The effect of these additions on lipid synthesis by chloroplasts isolated from spinach, pea, and amaranthus are shown in Table I. In all cases fatty acid synthesis was stimulated by both CoA and Triton X-100 and depressed by Mg. However, the effects of cyanide and sulfite were markedly species specific at the concentrations used. Cyanide (0.25 mM) inhibited fatty acid synthesis in amaranthus chloroplasts by 65%, in pea chloroplasts by 43%, and in spinach chloroplasts by 14%, while sulfite inhibited fatty acid synthesis in pea chloroplasts by 60%, in amaranthus chloroplasts by 39%, and in spinach chloroplasts by only 7%. In general, the additions had the expected effects upon the proportions of lipid classes synthesized, although those effects were small. In addition, PG synthesis in spinach chloroplasts was consis-

ACYL-ACYL

CARRIER

PROTEIN

AND

tently stimulated by cyanide and inhibited by Triton X100, and the ratio of LPA to PA synthesized, particularly by spinach chloroplasts, was highest in the presence of cyanide and sulfite. (ii) On acyl-ACP concentratiom. The most dramatic effect resulting from the different treatments was the accumulation of stearoyl-ACP in spinach and pea chloroplasts in the presence of cyanide (Fig. 2). This accumulation was greatest at the highest rates of de nouo fatty acid synthesis and is a consequence of high rates of stearoyl-ACP synthesis coupled with low rates of stearoylACP desaturation. The increased acyl-ACP pool size was presumably responsible for the increased ratios of LPA to PA in spinach and pea chloroplasts treated with cyanide (Table I). The stearoyl-ACP concentration in spinach was doubled compared with the control in the presence of sulfite. More subtle differences were the increases in total acyl-ACP in the presence of CoA and Triton X-100, and increases in the proportion of oleoyl- to palmitoyl-ACP in the presence of Triton X-100 and sulfite, in spite of an expectation that a sulfite-induced anaerobiosis might decrease oleoyl-ACP concentrations. (iii) On LPA fatty acids. The fatty acid compositions of the LPAs synthesized by the isolated chloroplasts were expected to be 90, 80, and 50 mol% oleate for spinach, pea, and amaranthus chloroplasts, respectively, as judged from the known fatty acid compositions of their PGs (7, 17). However, the amount of oleate esterified was almost invariably slightly less than anticipated (Fig. 3). Only in the presence of Triton X-100, and to a lesser extent sulfite, were LPA fatty acid compositions in close agreement with expectations from in uiuo data. In both spinach and pea chloroplasts, relatively small increases in the oleate to palmitate ratio within the acyl-ACP pool were associated with significant increases in the oleate content of LPA.

ADDITIONS

TO BASAL

MEDIUM

FIG. 2. Effects of various additions to incubation media on fatty acid compositions of the acyl-ACP and lyso-phosphatidate pools during lipid synthesis from [l-WJacetate by chloroplasts isolated from spinach (S), pea (P), and amaranthus (A). Bars to the left, in the middle, and to the right of each grouping represent oleate, stearate, and palmitate, respectively. Incubation conditions were as for Table I. CONT, no additions to basal medium; CN, 0.25 mM KCN; S03, 2 mM Na,SO,; Mg, 2 mM MgC&; CoA, 0.5 mM; TRIT, 0.13 mM Triton X-100; chl, chlorophyll.

CHLOROPLAST

LIPID

SYNTHESIS

GLYCEROL

3-PHOSPHATE

(mM)

FIG. 3. Effect of increasing concentrations of exogenous glycerol 3phosphate on the fatty acid compositions of acyl-ACP and lyso-phosphatidate pools during lipid synthesis from [l-“C]acetate by chloroplasts isolated from spinach (S), pea (P), and amaranthus (A). Bars to the left, in the middle, and to the right of each grouping represent oleate, stearate, and palmitate, respectively. Incubation conditions were as for Table IV; chl, chlorophyll.

Therefore, although there was a strong preference for oleate in the G3P acyltransferase reaction in spinach and pea, small changes in the concentrations of the two acyl donors could influence the fatty acid compositions of chloroplast lipids. Amaranthus LPA contained oleate and palmitate in almost the same proportions as those occurring within the acyl-ACP pool, so it may be assumed that there is little or no discrimination between the two by the amaranthus G3P acyltransferase [see also Ref. (5)]. The extent to which stearate was incorporated into LPA of spinach and pea in this study was unexpected. Competition experiments in vitro indicate that the G3P acyltransferase of pea and spinach chloroplasts does not discriminate significantly between palmitoyl- and stearoyl-ACP, but the spinach enzyme in particular discriminates strongly against stearate when supplied with a mixture of stearoyl- and oleoyl-ACP (4). This is consistent with the measured K,,, of the enzymes for the different acyl donors and for G3P (4). However, when all three acyl donors, as acyl-CoAs, are mixed in equal proportions and presented to the purified acyltransferases, stearate is completely excluded from the reaction (spinach), or transferred weakly (pea) to G3P (21). In addition, the G3P concentration used here is expected to be saturating with respect to oleate transfer to G3P, but only about 5% of that required to saturate stearate transfer (4). However, the reaction within the intact organelle appeared to be much less selective. A relatively small increase in the proportion of stearoyl-ACP in the acyl donor pool resulted in a significant increase in the stearate content of LPA in spinach and pea chloroplasts (Fig. 2, sulfite treatment), and an impressive transfer of stearate to G3P occurred when stearoyl-ACP dominated the acyl-ACP pool (Fig. 2, cyanide treatment). Since the LPA pool size increased a little under the same conditions (Table I, Fig. 2), stearate incorporation into LPA was not simply by default, but

96

ROUGHAN

AND

was an active process. It is difficult to decide from these data whether actual concentrations or relative concentrations are important in determining the acyl specificity of the acyltransferase reaction in situ (i.e., K,,,, V,,,,,, or selectivity). For instance, compared with the controls, Triton X-100 treatment virtually doubled oleoyl- and stearoyl-ACP concentrations in chloroplasts while changing the proportions of the acyl donors much less dramatically. Yet stearate and palmitate were even more rigorously excluded from spinach and pea LPA in the presence of Triton X-100 compared with controls, thus suggesting that the actual concentration of oleoyl-ACP was more important than the proportional mix of acyl donors. Unless Triton has some effect other than that of increasing acyl-ACP concentrations within spinach chloroplasts, then at some concentration between 40 and 60 pmol/mg chlorophyll (2-4 PM), oleoyl-ACP must saturate the acyltransferase in situ virtually precluding binding of palmitoyl- and stearoyl-ACP to the enzyme (Fig. 2). (iv) On PA molecular species. Fatty acid compositions and molecular species of PA were analyzed to check on the validity of the results for LPA, and to determine whether there had been a selection of LPA species for PA synthesis. Since palmitate alone may be esterified at the sn-2 position of chloroplast PA (4), there was in spinach and amaranthus good correspondence between the fatty acid compositions of LPA and PA (Table II). In pea chloroplasts, on the other hand, a high proportion of stearate within LPA was not consistently reflected by a compleTABLE

MATSUO

mentary high stearate content within PA. In three separate chloroplast isolations LPA from cyanide-treated chloroplasts averaged 17% palmitate, 50% stearate, and 33% oleate while the corresponding PA averaged 60% palmitate, 20% stearate, and 20% oleate. Therefore, unlike the spinach enzyme, the pea 1-acyl G3P acyltransferase appeared to discriminate slightly against 1-stearoyl G3P. In other treatments, the fatty acid composition of PA was a more reliable indicator of the fatty acid composition of its precursor LPA. However, invariably there was slightly more disaturated PA measured by AgN03 TLC than would be expected from fatty acid analyses alone (Table II). The close correspondence between the fatty acid compositions of LPA and PA tended to confirm that the LPA isolated for these analyses was predominantly, if not entirely, 1-acyl G3P. (v) On the composition of the UFA. A comparison of the labeled fatty acids esterified to LPA with those being released as UFA gave an indication of the competition between the acyltransferase and thioesterase reactions when presented in situ with the same pool of acyl-ACP (Table III). The lower ratio of oleate to palmitate in UFA of spinach compared to pea chloroplasts is characteristic for chloroplasts from 16:3 plants (12), and probably reflects the greater capacity of such chloroplasts to synthesize glycerolipids (19). Particularly in those cases where stearoyl-ACP comprised a high proportion of the acylACP pool (cyanide treatment), the thioesterases of both spinach and pea discriminated more strongly than the II

Relationship between Fatty Acid Compositions of LPA and PA and Molecular Species of PA Synthesized by Isolated Chloroplasts Amaranthus

Spinach Cont

CN

SOB

Mg

CoA

Trit

% of total FAME

Cont

CN

SOB

Mg

CoA

Trit

radioactivity

LPA Oleate Stearate Palmitate

82 5 13

42 49 9

84 12 4

81 5 14

86 4 10

91 3 6

33 12 55

25 19 56

49 8 43

32 9 59

30 13 57

45 12 43

PA Oleate Stearate Palmitate

42 2 56

25 28 47

45 6 49

38 2 60

44 2 54

47 2 51

12 5 83

15 11 73

17 6 77

14 5 81

12 7 81

22 7 71

as disaturated

molecular

68 72 75

70 76 80

55 56 61

% of PA radioactivity From LPA FAME From PA FAME by TLC

18 16 20

58 50 54

16 10 18

19 24 27

14 12 18

9 6 12

67 76 80

75 68 75

species 51 66 70

Note. LPA and PA were isolated by TLC. FAME were prepared from LPA and PA by transmethylation in NaOHC3 and were analyzed by a combination of reverse-phase and AgNO, TLC. LPA data are extracted from Fig. 2. PA was also converted to dimethyl-PA using diazomethane and was separated into saturated and monoenoic molecular species by AgN03 TLC. Cont, control; CN, KCN; SOS, Na,SOs; Mg, MgCl,; concentrations as for Table I.

ACYL-ACYL

CARRIER

PROTEIN

AND TABLE

CHLOROPLAST

LIPID

97

SYNTHESIS

III

Comparison between Compositions of Fatty Acids from LPA and Unesterified Fatty Acids Synthesized by Spinach and Pea Chloroplasts in the Presence of Various Additions to Reaction Media Cont Fatty acid

LPA

CN UFA

LPA

so3 UFA

LPA

CoA

Mg UFA

% of radioactivity

LPA

UFA

Trit

LPA

UFA

LPA

UFA

in 18:l + 180 + 16:0

Spinach Oleate Stearate Palmitate

82 5 13

67 5 28

50 40 10

52 20 28

84 12 4

74 4 22

81 5 14

58 3 39

86 4 10

69 4 27

91 3 6

88 6 6

Pea Oleate Stearate Palmitate

73 6 21

88 5 7

34 50 16

62 29 9

62 21 17

84 6 10

78 7 15

89 4 7

77 7 16

89 5 6

89 4 7

92 5 3

Note. FAME prepared from LPA and UFA following isolation All values are means from two separate chloroplast preparations.

by TLC were separated using a combination of AgNOa and reverse-phase TLC. LPA data are extracted from Fig. 2 and treatments were as for Table II.

acyltransferases against stearoyl-ACP (Table III, Fig. 2). However, while spinach acyltransferase discriminated more than the thioesterase against palmitoyl-ACP, the opposite was the case for the pea enzyme (Table III). Both reactions showed a strong preference for oleoyl-ACP, particularly in the presence of Triton X-100 which also elevated oleoyl-ACP concentration relative to that of palmitoyl-ACP. The stronger discrimination of the thioesterases against stearate may also be inferred from a comparison of the fatty acids within the LPA and UFA pools of amaranthus chloroplasts (13), and probably explains the slight decrease in the proportion of unesterified fatty acids to glycerides synthesized by spinach chloroplasts in the presence of cyanide [Table I and Ref. (20)]. The stimulation by Triton X-100 of glycerolipid synthesis within chloroplasts from 16:3 plants, in both the presence and the absence of exogenous G3P, is accompanied by a decrease in the proportion of unesterified palmitic acid [Table III; Refs. (12) and (22)]. This suggests that the membrane-bound 1-acyl G3P acyltransferase competes more successfully than the thioesterase for endogenous palmitoyl-ACP. The Effects of Increasing

G3P Concentrations

This second set of (i) On chloroplast lipid synthesis. experiments was carried out to attempt to force the synthesis of disaturated molecular species of chloroplast glycerolipids by using relatively high concentrations of G3P. The well-documented (12, 22) decrease in UFA accumulation and increase in glyceride synthesis by isolated chloroplasts with increasing concentrations of exogenous G3P was also observed in the present study (Table IV). The data highlight yet again the relatively weak capacity of chloroplasts from l&3 plants, e.g., pea and amaranthus, to synthesize glycerolipids. Increasing G3P concentrations

had only minor effects on rates of fatty acid synthesis de novo but had a marked effect upon the proportion of LPA to PA synthesized, particularly in spinach. In both spinach and pea the synthesis of LPA was stimulated more than that of PA, and it seems possible that acylation of LPA had become rate-limiting at the higher G3P concentrations. Decreases in the size (ii) On acyl-ACP concentrations. of acyl-ACP pools in response to added G3P (Fig. 3) were lower than those reported previously (2, 3) probably for the reasons given above. Palmitate and oleate were depressed in spinach, palmitate and stearate in amaranthus, and there was little effect on pea. (iii) On LPA fatty acids. In all cases, the proportion of oleate within LPA was slightly lowered by increasing G3P concentrations (Fig. 3). This was balanced by increasing proportions of palmitate and stearate in spinach, of palmitate in pea, and of stearate in amaranthus. These changes were taking place even when there was little or no change in acyl-ACP concentrations and must be assumed to be due to G3P concentration per se rather than an effect of G3P on acyl-ACP concentrations. Values recorded for O-O.1 mM G3P are of limited reliability because of the small amounts of LPA accumulated. (iv) On PA molecular species. The small increases in the proportions of saturated fatty acids within LPA translated to significant increases in the amounts of disaturated PA synthesized at the higher G3P concentrations (Table V). As in the previous experiment, there was reasonably good correspondence between the fatty acid compositions of LPA and PA (results not shown), particularly since there would be a disproportionate incorporation of palmitate into PA at the lower G3P concentrations (19). However, the agreement between molecular species calculated from fatty acid compositions

98

ROUGHAN

AND

TABLE

MATSUO

IV

Effect of Increasing Concentrations of Exogenous sn-Glycerol3-Phosphate on the Different Lipid Classes Synthesized from Acetate by Isolated Chloroplasts Glycerol

Plant

Lipid

0

0.1

J-phosphate 0.2

added (mM) 0.4

nmol acetate incorporated/mg Spinach

Pea

Amaranthus

1.0

2.0

chl

UFA DAG LPA PA PG Other

152’ 34 3 7 6 17

108 72 6 15 9 19

97 84 9 21 10 20

79 87 12 22 11 21

70 94 16 22 12 20

69 89 20 22 14 18

Total

219

229

241

232

234

232

UFA DAG LPA PA PG Other

222 4 1 5 19

214 5 3 13 15

199 7 5 19 1 13

184 7 7 23 2 12

170 9 13 31 2 15

153 9 17 37 2 14

Total

256

250

244

235

240

232

UFA DAG LPA PA PG Other

188 6 2 11 4 11

161 9 5 26 8 15

155 12 7 32 11 13

149 13 8 34 11 15

153 13 10 38 13 15

163 12 12 43 16 12

Total

222

224

230

230

237

258

DValues are means from two chloroplast preparations. -, 70% of that synthesized by chloroplasts from amaranthus, was disaturated. Even at “physiological” concentrations of G3P, estimated to be between 0.2 and 0.4 mM in illuminated leaves (23,24), the isolated chloroplasts tend to synthesize a greater proportion of disaturated glycerolipids than are found in vivo or than would be expected from the selectivities shown by enzyme preparations. This discrepancy cannot be attributed entirely to a disproportionate incorporation of labeled palmitate into PA resulting from the presence of unlabeled LPA in the freshly isolated chloroplasts (19) since, in the presence of Triton X-100, the molecular species synthesized agreed with expectations derived from known fatty acid compositions of plant glycerolipids.

CONCLUSIONS The aim of the present work was to determine whether the fatty acid compositions of chloroplast lipids were controlled solely by the specificity of the G3P acyltransferase reaction as determined in vitro, or whether other factors, such as endogenous acyl-ACP and G3P concentrations, might also be involved. The results indicate that, while the acyltransferase plays the dominant role in specifying which fatty acid is esterified to G3P, the acyl compositions of glycerolipids synthesized by isolated chloroplasts may, nevertheless, be manipulated more readily than would be anticipated from either the acyl specificity or the selectivity of the soluble acyltransferases (4, 5) or from the relatively invariant fatty acid composition of PG within a plant species (7). As discussed previously (5), the kinetics of the reaction catalyzed by the G3P acyltransferases purified from spinach and pea chloroplasts are complex. Reaction rates at

ACYL-ACYL TABLE

CARRIER

PROTEIN

AND

V

G3P (mM) 0.2

Disaturated Spinach From LPA FAME From PA FAME From TLC Pea From LPA FAME From PA FAME From TLC Amaranthus From LPA FAME From PA FAME From TLC

0.4

1.0

2.0

PA (% of PA radioactivity)

15 8 13

12 6 11

12 10 15

18 20 20

27 30 31

24 21 28

18 17 29

16 19 28

21 26 36

27 32 43

65 62

68 66

69 68

73 70

13

74

16

78

74 72 79

Note. Molecular species composition were either calculated from the fatty acid compositions of LPA and PA or measured directly by AgNOa TLC. Values for zero added G3P (not shown) were quite anomalous and are presumed to be most strongly affected by the presence of unlabeled l-acyl G3P in the isolated chloroplats (19).

saturating substrate concentrations were 8- to lo-fold greater with palmitoyland stearoyl-ACP than with oleoyl-ACP (4) and yet, when all three substrates were supplied simultaneously, the product was 90% oleoyl-G3P (4). It was therefore suggested (5) that in pea and spinach chloroplasts, oleoyl-ACP might become a more active acyl donor in the presence of palmitoylor stearoyl-ACP through an allosteric mechanism. In addition, chloroplasts from the chilling-sensitive plant, Cucurbitu moschatu, contain three isoforms of G3P acyltransferase with about equal apparent activities (25). Although the kinetic parameters of one of the isoforms (ATl) predict that oleoylACP would be the least preferred substrate at saturating substrate concentrations, the enzyme preferentially transfers oleate from mixtures of oleoyl-ACP with palmitoyl- or stearoyl-ACP (25). However, this preference of AT1 for oleoyl-ACP is significantly reduced when the reaction is carried out at pH 8 (as in the stroma of illuminated chloroplasts) rather than pH 7.4, and the other two isoforms exhibit little selectivity toward oleoyl-ACP. At least in the case of ATl, it could again be argued that oleoyl-ACP becomes a more active acyl donor when palmitoyl- or stearoyl-ACP is also present in the reaction mixture. In other work (26) on the chilling-sensitive cucurbits, cantaloupe and cucumber, both containing high proportions of disaturated PG, the chloroplast G3P acyltransferase was found to be as selective as the pea enzyme in favor of oleoyl-ACP even though the G3P concentration was very high (8 mM), and the K,,, for oleoyl-ACP (25

LIPID

SYNTHESIS

99

was 2- to 3-fold higher than that for palmitoyl-ACP (26). Therefore, it was concluded (26) that a selection of LPA molecular species for PA synthesis, and of PA molecular species for PG synthesis, was more important than any acyl preferences of the chloroplast G3P acyltransferase for determining the proportion of disaturated PG occurring in the thylakoids of those plant species. Such a selectivity in favor of saturated molecular species of precursor glycerolipids was, however, not evident either in more detailed studies of PG synthesis by chloroplasts from the chilling-sensitive Amaranthus liuidus (13) or in the present work. In other circumstances, the preference of the reaction for oleoyl-ACP may be a consequence of endogenous G3P concentrations. Whereas the K,,, for G3P in acyl transfer from oleoyl-ACP by the purified spinach enzyme was 31 PM, that for acyl transfer from stearoyl- and palmitoylACP was over 3 mM (4). Therefore, stromal concentrations of G3P, estimated to be 0.1-0.2 mM for spinach chloroplasts (24), are saturating with respect to oleate transfer but only about 1.5-3% of those required to saturate palmitate and stearate transfer. Given that significant synthesis of palmitoyl- and stearoyl-G3P did occur at “physiological” concentrations of G3P in the present study, it seems possible that an allosteric mechanism involving the binding of oleoyl-ACP greatly increases the reactivity of G3P in acyltransferase with respect to acyl transfer from palmitoyland stearoyl-ACP. Unfortunately, the Km for G3P in palmitate transfer by the purified enzyme has not been measured in the presence of oleoyl-ACP. One report (24), however, has suggested K, (G3P) of 0.6 and 0.3 mM for oleate and palmitate transfer, respectively, from mixed endogenous acyl donors in spinach chloroplast extracts. An additional complication in attempting to interpret the results is that, even though it may be possible to calculate a G3P concentration for the cytoplasm or stroma, such a concentration may not accurately reflect that in the immediate environment of the enzyme in situ. One of the difficulties in relating these results to physiological reality is the uncertainty as to how accurately the acyl-ACP concentrations of isolated chloroplasts reflect those occurring in Go. For instance, immunoblot analyses have indicated that stearoyl-ACP is the major long-chain ester of ACP in illuminated spinach leaves (11). However, it is a minor component of the long-chain acyl-ACP pool in isolated spinach chloroplasts actively synthesizing fatty acids from [1-14C]acetate (D. PostBeittenmiller, personal communication). In general, highest concentrations of long-chain acyl-ACPs within isolated chloroplasts occur when rates of fatty acid synthesis are highest, equivalent to about 2 pmol acetate incorporated per hour per milligram chlorophyll, and about 50% higher than the best estimates for rates of fatty acid synthesis in uiuo (27, 28). Also, since the concentration of an individual acyl-ACP at any given time represents PM?)

The Effect of Increasing Concentrations of G3P on the Proportion of Disaturated Molecular Species of PA Synthesized by Isolated Chloroplasts

0.1

CHLOROPLAST

ROUGHAN

100

the difference between its rates of synthesis and utilization, then lower rates of fatty acid synthesis de rwvo might be expected to result in lower concentrations of palmitoylrelative to oleoyl-ACP. Although this was not verified experimentally with isolated chloroplasts, it still seems possible that lower average rates of fatty acid synthesis in vivo could result in higher ratios of oleoyl- to palmitoylACP and, therefore, a more selective transfer in vivo than in vitro of oleate to G3P by chloroplasts. For example, only in those chloroplast preparations having a particularly active fatty acid synthetase did stearoyl-ACP accumulate in cyanide-treated chloroplasts to the extent shown in Fig. 1. This was presumably a result of a relatively slow turnover of stearoyl-ACP in the presence of KCN, and suggests that desaturation rather than acyl transfer is the predominant fate of stearoyl-ACP in vivo. Turnover of stearoyl-ACP in pea chloroplasts incubated in the presence of KCN but without added G3P was about a fourth (tljz = 1 min in the dark) that of palmitoyl- and oleoyl-ACP in the same preparation (result not shown).

Part of this work was performed under the auspices of the International Joint Research Project of the Japanese Society for the Promotion of Science. Tomoaki Matsuo thanks the JSPS for supporting his visit to New Zealand.

REFERENCES 1. Roughan, P. G., and Slack, C. R. (1984) Trends Biochem. Sci. 9,

383-386. 2. Soll, J., and Roughan, P. G. (1982) FEBS Lett. 146, 189-192. P. G., and Nishida,

MATSUO 5. Cronan, J. E., Jr., and Roughan, P. G. (1987) Plant Physiol.

63,

676-680. 6. Murata, N., Sato, N., Takahashi, N., and Hamazaki, Y. (1982) Plant Cell Physiol. 23, 1071-1079. 7. Roughan, P. G. (1985) Plant Physiol. 77, 740-746. 8. Murata, N., and Yamaya, J. (1984) Plant Physiol. 74,1016-1024. 9. Roughan, P. G., Holland, R., and Slack, C. R. (1979) Biochem. J.

184,193-202. 10. McKee, J. W. A., and Hawke, J. C. (1979) Arch. B&hem.

Biophys.

197,322-332. 11. Post-Beittenmiller, D., Jaworski, J. G., and Ohlrogge, J. B. (1991) J. Biol. Chem. 266, 1858-1865. 12. Gardiner, S. E., Heinz, E., and Roughan, P. G. (1984) Plant Physiol.

74,890-896. 13. Roughan, P. G. (1986) Biochem. Biophys. Acta 878,371-379. 14. Jenkins, C. L. D., and Russ, V. J. (1984) Plant Sci. Lett. 35, 19-

24. 15. Cerovic, Z. G., and Plesnicar, M. (1984) Biochem. J. 223,543-545. 16. Mills, W. R., and Joy, K. W. (1980) Planta 148, 75-83. 17. Roughan, P. G. (1986) Plant Sci. 43,57-62. 18. Vick, B., and Beevers, H. (1978) Plant Physiol. 62, 173-178. 19. Gardiner, S. E., Roughan, P. G., and Browse, J. (1984) Biochem. J.

224,637-642. 20. Roughan, P. G., and Beevers, H. (1981) Plant Physiol. 67,926-929.

ACKNOWLEDGMENT

3. Roughan,

AND

I. (1990) Arch. Biochem. Biophys.

276,38-46. 4. Frentzen, M., Heinz, E., McKoen, Eur. J. Biochem. 129, 629-636.

T. A., and Stumpf, P. K. (1983)

21. Bertrams, M., and Heinz, E. (1981) Plant Physiol. 68,653-657. 22. Roughan, P. G., Slack, C. R., and Holland, R. (1976) Biochem. J.

158,593-601. 23. Gardiner, S. E., Roughan, P. G., and Slack, C. R. (1982) Plant Physiol. 70,1316-1320. 24. Sauer, A., and Heise, K-P. (1984) 2. Naturforsch. 39c, 593-599. 25. Frentzen, M., Nishida, I., and Murata, N. (1987) Plant Cell Physiol. 28, 1195-1201. 26. Thomas, S. R., Sanchez, J., and Mudd, J. B. (1987) in The Metabolism Structure and Function of Plant Lipids (Stumpf, P. K., Mudd, J. B., and Nes, W. D., Eds.), pp. 283-291, Plenum, New York. 27. Browse, J., Roughan, P. G., and Slack, C. R. (1981) Biochem. J.

196,347-354. 28. Murphy, D. J., and Leech, R. M. (1981) Plant Physiol. 68, 762765.

The influence of endogenous acyl-acyl carrier protein concentrations on fatty acid compositions of chloroplast glycerolipids.

The concentrations of long-chain acyl-acyl carrier proteins (acyl-ACP) occurring during fatty acid synthesis from [1-14C]acetate within chloroplasts i...
1MB Sizes 0 Downloads 0 Views