Molecular Species of PC and PE Formed During Castor Oil Biosynthesis Jiann-Tsyh Lin*, Jennifer M. Chen, Pei Chen, Lucy P. Liao, and Thomas A. McKeon USDA, ARS, Western Regional Research Center, Albany, California 94710

ABSTRACT: As part of a program to elucidate castor oil biosynthesis, we have identified 36 molecular species of PC and 35 molecular species of PE isolated from castor microsomes after incubations with [14C]-labeled FA. The six [14C]FA studied were ricinoleate, stearate, oleate, linoleate, linolenate, and palmitate, which were the only FA identified in castor microsomal incubations. The incorporation of each of the six FA into PC was better than that into PE. The [14C]FA were incorporated almost exclusively into the sn-2 position of both PC and PE. The incorporation of [14C]stearate and [14C]palmitate into 2-acylPC was slower compared to the other four [14C]FA. The incorporation does not show any selectivity for the various lysoPC molecular species. The level of incorporation of [14C]FA in PC was in the order of: oleate > linolenate > palmitate > linoleate > stearate > ricinoleate, and in PE: linoleate > linolenate > oleate > palmitate > stearate > ricinoleate. In general, at the sn-1 position of both PC and PE, linoleate was the most abundant FA, palmitate was the next, and oleate, linolenate, stearate, and ricinoleate were minor FA. The activities of oleoyl-12-hydroxylase, oleoyl-12-desaturase seem unaffected by the FA at the sn-1 position of 2-oleoyl-PC. The FA in the sn-1 position of PC does not significantly affect the activity of phospholipase A2, whereas ricinoleate is preferentially removed from the sn-2 position of PC. The results show that (i) [14C]oleate is most actively incorporated to form 2-oleoyl-PC, the immediate substrate of oleoyl-12-hydroxylase; (ii) 2-ricinoleoyl-PC is formed mostly by the hydroxylation of 2-oleoyl-PC, not from the incorporation of ricinoleate into 2-ricinoleoyl-PC; and (iii) 2-oleoyl-PE is less actively formed than 2-oleoyl-PC. Paper no. L9031 in Lipids 37, 991–995 (October 2002).

Ricinoleate has many industrial uses. Its only commercial source is castor oil, in which ricinoleate constitutes 90% of the FA (1). Since castor bean contains the toxin ricin as well as potent allergens, it is hazardous to grow, harvest, and process. It would be desirable to produce ricinoleate instead in a transgenic oilseed lacking these toxic components. To develop a transgenic plant capable of producing a high level of ricinoleate in its seed oil, one must understand the biosynthesis of castor oil. The identified enzymatic steps on the path*To whom correspondence should be sent at USDA, 800 Buchanan St., Albany, CA 94710. E-mail: [email protected] Abbreviations: 2-oleoyl-PC, 1-acyl-2-oleoyl-sn-glycero-3-phosphocholine; 2-oleoyl-PE, 1-acyl-2-oleoyl-sn-glycero-3-phosphoethanolamine; 2-ricinoleoyl-PC, 1-acyl-2-ricinoleoyl-sn-glycero-3-phosphocholine; RT, retention time. Copyright © 2002 by AOCS Press

way regulating the production of ricinoleate-rich castor oil can be used to identify the corresponding genes to be cloned and expressed in other oilseed plants. We have previously reported on the biosynthesis of castor oil and identified the key enzymatic steps on the pathway that drive the ricinoleate into castor oil (2,3). The main pathway is: 2-oleoyl-PC (1-acyl-2-oleoyl-sn-glycero-3-phosphocholine) ⇒ 2-ricinoleoyl-PC (1-acyl-2-ricinoleoyl-sn-glycero-3phosphocholine) ⇒ ricinoleate → ricinoleoyl-CoA → ricinoleoyl-lysoPA → diricinoleoyl-PA → 1,2-diricinoleoyl-snglycerol ⇒ triricinolein. The heavy arrows (⇒) indicate the key enzymatic steps. In castor microsomal incubation, oleoyl-CoA is rapidly incorporated into 2-oleoyl-PC by acyl-CoA:lysoPC acyltransferase (EC 2.3.1.23) (4,5). The 2-oleoyl-PC formed is then hydroxylated to 2-ricinoleoyl-PC by oleoyl-12-hydroxylase and desaturated to 2-linoleoylPC by oleoyl-12-desaturase (EC 1.3.1.35) (2). 2-RicinoleoylPC is then hydrolyzed by phospholipase A2 (EC 3.1.1.4) to release ricinoleate for the biosynthesis of TAG containing ricinoleate (2). 2-Oleoyl-PE (1-acyl-2-oleoyl-sn-glycero-3phosphoethanolamine) cannot be hydroxylated to 2-ricinoleoyl-PE, but can first be converted to 2-oleoyl-PC and then hydroxylated (3). The key enzyme, oleoyl-12-hydroxylase, has been characterized (5,6). We would like to extend the understanding of this pathway to the molecular species of lipid classes. We have recently reported the molecular species of acylglycerols produced in microsomal incubations after incorporating six [14C] FA in castor microsomal incubations (7). The incorporations of these FA into TAG are in the order of: ricinoleate > oleate > linoleate > linolenate > stearate > palmitate. In this work, we also describe the incorporation of six [14C]FA, representing the endogenous FA, into the molecular species of PC and PE, intermediates in the biosynthesis of castor oil. EXPERIMENTAL PROCEDURES Microsomal incubation. Microsomes from castor bean were prepared as described previously (2). To obtain suitable amounts of label incorporation for analysis, the incubation mixture was scaled up 20-fold in a total volume of 20 mL: sodium phosphate buffer (0.1 M, pH 6.3), CoA-SH (10 µmol), NADH (10 µmol), ATP (10 µmol), MgCl2 (10 µmol), catalase (20,000 units), and microsomal fraction from

991

Lipids, Vol. 37, no. 10 (2002)

992

J.-T. LIN ET AL.

endosperm of immature castor bean (300 µL, 2.76 mg of protein). The [1-14C]FA, ricinoleate (55 Ci/mol; American Radiolabeled Chemicals, Inc., St. Louis, MO), stearate (44 Ci/mol; NEN Life Science Products, Boston, MA), oleate (52 Ci/mol; NEN), linoleate (51 Ci/mol; NEN), linolenate (52 Ci/mol; NEN), and palmitate (56 Ci/mol; NEN) were incubated individually. The [14C]FA (5.0 µCi) in 400 µL ethanol was added last into a screw-capped bottle containing incubation mixture and then immediately mixed. The mixture was then incubated in a shaking water bath for 60 min at 22°C. The incubation was stopped by suspension in 75 mL of chloroform/methanol (1:2, vol/vol). The mixture was again mixed with 12.5 mL of chloroform and 12.5 mL of water. The lower chloroform layer, containing the lipid extract, was dried and fractionated by silica HPLC to separate lipid classes as described below. For time-course studies, 1⁄20 of the incubation components and extraction solvents, and [14C]FA (0.5 µCi) were used. Duplicate incubations were done at various times and the averages of incorporations were used. Determination of the sn-1,2 positions of [14C]FA on PC and PE. The [14C]PC and [14C]PE obtained from the castor microsomal incubations were hydrolyzed with 0.2 mg phospholipase A2 (Naja mossambica mossambica, P4034; Sigma) dissolved in incubation buffer, in a total of 1 mL buffer (0.1 M Tris, pH 8.9). The PC or PE was dissolved in 20 µL of ethanol and added to start the reaction, followed by immediate mixing. The mixture was incubated in a shaking water bath at 25°C overnight and then neutralized with HCl (0.1 M). The total lipid was extracted as described above for the castor microsomal incubation. The free FA and lysoPC (lysoPE) formed were separated by silica HPLC and counted by a flow scintillation analyzer as described below. Under these conditions, the esters at sn-2 of both PC and PE were completely hydrolyzed by phospholipase A2. HPLC. HPLC was carried out on a liquid chromatograph (Waters Associates, Milford, MA), using a UV detector (Waters 2487) at 205 nm and a flow scintillation analyzer (150TR; Packard Instrument Co., Downers Grove, IL) to detect [14C]labeled compounds. Labeled lipids were separated by HPLC and where possible identified by co-chromatography with lipid standards and matching the retention times (RT) on the UV chromatogram and radiochromatogram. The PC and PE standards were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL) and Sigma. The flow rate of HPLC eluents was 1 mL/min. The flow rate of liquid scintillation fluid (Ultima Flo M; Packard Instrument Co.) in the flow scintillation analyzer was 3 mL/min. The maximum scaling value of the graph from the flow scintillation analyzer was set at 10000. (i) Separation of lipid classes. Lipid classes were separated according to Singleton and Stikeleather (8) on a silica column (25 × 0.46 cm, 5 µm, Luna, silica(2); Phenomenex, Torrance, CA) with a linear gradient of 2-propanol/hexane (4:3, vol/vol) to 2-propanol/hexane/water (4:3:0.75, by vol) in 20 min, then isocratically for 20 min. A prepacked silica saturator column (3 × 0.46 cm, 15–25 µm; Phenomenex) was installed between Lipids, Vol. 37, no. 10 (2002)

the pump and injector to saturate the mobile phase with silica before it reached the column. To elute lysoPC more rapidly after phospholipase A2 hydrolysis of PC, a linear gradient of 2-propanol/hexane (4:3, vol/vol) to 2-propanol/hexane/water (4:3:0.85, by vol) in 20 min, then isocratic for 30 min, was used. Eluent B is near saturation with water. In this HPLC system, RT were: 40.0 min, lysoPC (1-oleoyl-lysoPC); 30.6 min, PC (1-palmitoyl-2-oleoyl-PC); and about 3.5 min, FFA. For the separation of lysoPE, the former silica HPLC system (2-propanol/hexane/water, 4:3:0.75 by vol) was used, and the RT were: 22.6 min, lysoPE (1-oleoyl-lysoPE); 17.7 min, PE (1-palmitoyl-2-oleoyl-PE); and about 3.5 min, FFA. (ii) Separation of molecular species of PC. Molecular species of PC were separated in 40 min as previously reported (9), using a C8 column (25 × 0.46 cm, 5 µm, Luna C8; Phenomenex) with a linear gradient of 90% aqueous methanol to 100% methanol, both containing 0.1% of concentrated NH4OH. NH4OH was used as silanol-suppressing agent. (iii) Separation of molecular species of PE. Molecular species of PE were separated as previously reported (10), using a C8 column (Luna C8; Phenomenex) with a linear gradient of 88% aqueous methanol to 100% methanol containing 0.1% of concentrated NH4OH in 40 min. RESULTS AND DISCUSSION Separation of lipid classes. The total lipids obtained from the incubation of each [14C]FA were separated into lipid classes by silica HPLC as shown earlier (Fig. 1; Ref. 7). The lipid classes separated were AG (and FFA), PE, PC (without ricinoleoyl chain), and ricinoleoyl-PC in the order of elution. All of the six [14C]FA used were incorporated into PC and PE. The levels of [14C]PC were higher than those of [14C]PE and [14C]AG (7). Time course studies, of up to 120 min, of microsomal incubations of the six [14C]FA individually showed that in general the level incorporated by PE, PC (without ricinoleoyl chain), and ricinoleoyl-PC increased initially and then equilibrated at 45 to 60 min when the rate of incorporation from FA became equal to the rate of hydrolysis. Since ricinoleate is actively removed from the mixture of 2-acyl-PC containing different FA at sn-2 (2), the radioactivity levels at 15 min (initial rate) were used to compare the rates of incorporation. The incorporation of [14C]stearate and [14C]palmitate into 2-acylPC was slower than of the other four [14C]FA. Usually these two FA are strongly excluded from the sn-2 position of phospholipids. In comparing the amount of label incorporated at the equilibrium stages, PC and PE incorporating [14C]-labeled oleate, linoleate, and linolenate were at higher levels than those with ricinoleate, stearate, and palmitate. The low incorporation of label by 2-ricinoleoyl-PC at equilibrium stage may be attributed to a futile cycle of incorporation and hydrolysis. With the exception of ricinoleate, which was most actively incorporated into TAG (7), the incorporation of FA (FA preference) is similar to that observed for the incorporations of the six [14C]FA into TAG. This is also consistent with

MOLECULAR SPECIES OF PC IN CASTOR OIL BIOSYNTHESIS

earlier studies in which [14C]oleoyl-CoA and [14C]ricinoleoyl-CoA were incorporated into [14C]PC in castor microsomal incubation (4) and when [14C]oleoyl-CoA, [14C]linoleoyl-CoA, and [14C]palmitoyl-CoA were incorporated into [14C]PC in soybean microsomal incubation (11). The incorporation of label from [14C]oleate, [14C]linoleate, and [14C]linolenate into PE was higher than that from [14C]ricinoleate, [14C]stearate, and [14C]palmitate. PE had not been previously reported in microsomal incubations of [14C]FA and [14C]acyl-CoA. Determination of the ratio of [14C]FA on sn-1,2 positions of [14C]PC and [14C]PE. The [14C]PC and [14C]PE fractions collected after silica HPLC from the 60-min incubations were subjected to phospholipase A2 hydrolysis to determine the ratio of [14C]FA on the sn-1 and -2 positions. The percentage of label at sn-1 in the total [14C]PC (sn-1 and -2 combined) from each [14C]FA were: 0.3% (1-[14C]ricinoleoyl-lysoPC, RT 47.7 min), 3.5% (1-[14C]stearoyl-lysoPC, 38.7 min), 2.3% (1-[14C]oleoyl-lysoPC, 40.0 min), 0.6% (1-[14C]linoleoyllysoPC, 40.8 min), 0.3% (1-[14C]linolenoyl-lysoPC, 41.2 min), and 7.3% (1-[14C]palmitoyl-lysoPC, 40.3 min). In earlier studies, in a soybean microsomal incubation, 95% of the [14C]PC incorporating [14C]oleoyl-CoA was labeled at the sn-2 position (11), and in a castor microsomal incubation of [14C]oleoyl-CoA, 84.4% of the [14C]PC incorporated label at the sn-2 position (4). The percentage of the label at sn-1 in the total [14C]PE (sn-1 and -2 combined) from each [14C]FA were: 2% (1-[14C]stearoyl-lysoPE, RT 22.3 min) and 1.8% (1-[14C]linoleoyl-lysoPE, 22.7 min). The levels of 1-[14C]ricinoleoyl-lysoPE, 1-[14C]oleoyl-lysoPE (22.6 min), 1[14C]linolenoyl-lysoPC and 1-[14C]palmitoyl-lysoPC were not detectable. For both PC and PE, the [14C]FA are incorporated predominantly at the sn-2 position. Identification of the molecular species of [14C]PC. The radiochromatogram separating molecular species of PC from the incubation of [14C]oleate on a C8 column (9) is shown as Figure 1. Peaks #4, 5, and 6 were identified by co-chromatography with the standards 1-palmitoyl-2-oleoyl-PC, 1,2-dioleoyl-PC, and 1-stearoyl-2-oleoyl-PC. Since the standards for peaks #1, 2, and 3 were not available, they were designated as 1-ricinoleoyl-2-oleoyl-PC, 1-linolenoyl-2-oleoylPC, and 1-linoleoyl-2-oleoyl-PC, based on relative RT and elution characteristics of the molecular species of PC (9). We have calculated the relative RT of the molecular species of PC from the six [14C]FA incubations based on the RT (32.15 min) of 1-palmitoyl-2-oleoyl-PC in our previous report (9) by co-chromatography with this PC standard (Lin, J.T., and McKeon, T.A., unpublished data). The calculated RT of the six 2-ricinoleoyl-PC in Figure 1 from the [14C]ricinoleate incubation are 9.4 min (1,2-ricinoleoyl-PC), 14.6 min (1-linolenoyl-2-ricinoleoyl-PC), 16.9 min (1-linoleoyl-2ricinoleoyl-PC), 18.5 min (1-palmitoyl-2-ricinoleoyl-PC), 19.6 min (1-oleoyl-2-ricinoleoyl-PC), and 22.7 min (1stearoyl-2-ricinoleoyl-PC). 1,2-Diricinoleoyl-PC (9.4 min) was not detected. It was probably hydrolyzed by phospholipases immediately after formation. The RT of peaks #7–10

993

FIG. 1. The C8 HPLC radiochromatogram separating the molecular species of PC incorporated from [14C]oleate in castor microsomal incubation. (1) 1-Ricinoleoyl-2-oleoyl-PC; (2) 1-linolenoyl-2-oleoyl-PC; (3) 1-linoleoyl-2-oleoyl-PC; (4) 1-palmitoyl-2-oleoyl-PC; (5) 1,2-dioleoylPC; (6) 1-stearoyl-2-oleoyl-PC; (7) 1-linolenoyl-2-ricinoleoyl-PC; (8) 1-linoleoyl-2-ricinoleoyl-PC; (9) 1-palmitoyl-2-ricinoleoyl-PC; (10) 1-stearoyl-2-ricinoleoyl-PC; (11) 1-linolenoyl-2-linoleoyl-PC; (12) 1palmitoyl-2-linoleoyl-PC.

are the same as the calculated RT. Labeled 1-oleoyl-2-ricinoleoyl-PC is a minor component in peak #1 (19.7 min, mostly 1-ricinoleoyl-2-oleoyl-PC), because the levels of label in 2ricinoleoyl-PC were lower than those of 2-oleoyl-PC and the ratio of the six labeled 2-ricinoleoyl-PC to each other were similar to those of the six 2-oleoyl-PC. Similarly, the calculated RT of 2-linoleoyl-PC in Figure 1 from the incubation with [14C]linoleate were used to designate the peaks #11 (1linolenoyl-2-linoleoyl-PC) and #12 (1-palmitoyl-2-linoleoylPC). The other four 2-linoleoyl-PC co-eluted with other [14C]PC as minor components. The label incorporated in 2linoleoyl-PC was lower than in 2-ricinoleoyl-PC. The molecular species of PC incorporating the other five [14C]FA were identified and designated the same as those from [14C]oleate, except the [14C]FA were unchanged in the incubations. The incorporations of [14C]oleoyl-CoA, [14C]oleate, and 14 [ C]linoleate into the molecular species of PC in the microsomes prepared from soybean (11), potato tuber (12), pea leaf (13), and developing sunflower seeds (14) have been reported. The labeled PC in these systems contained two of the labeled and unlabeled oleate, linoleate, linolenate, stearate, and palmitate (no ricinoleate) at both sn-1 and sn-2 positions. Radioactivity levels of the molecular species of [14C]PC. Table 1 shows the level of [14C]-label in the molecular species of [14C]PC incorporating each [14C]FA during a 60-min castor microsomal incubation. The [14C]PC are primarily derived from the action of acyl-CoA:lysoPC acyltransferase with endogenous lysoPC as acyl acceptor. The lysoPC are endogenous to the microsomes and contain any of the six different FA in the sn-1 position, and the [14C]FA at sn-2 are from [14C]acyl-CoA. In general, the levels of label of [14C]PC with the FA at the sn-1 position were in the order of linoleate > palmitate > oleate > linolenate, stearate, and ricinoleate, with linoleate and palmitate as the major sn-1 component of PC. Lipids, Vol. 37, no. 10 (2002)

994

J.-T. LIN ET AL. TABLE 1 Level of Incorporation of Radiolabeled FA into Molecular Species of PC in Castor Microsomal Incubations (60 min) FA (nmol)a Ricinoleate (91) Stearate (114) Oleate (96) Ricinoleatec Linoleatec Linoleate (98) Linolenate (96) Palmitate (89)

RFa-PCb

LnFa-PC

LFa-PC

nmol PFa-PC

OFa-PC

1.3 2.0 3.0 ND 0.1d 1.1 2.6 3.6

1.4 1.5 4.8 0.2 0.2 2.0 8.4 2.4

7.6 9.1 26.8 1.5 1.1d 17.2 24.2 14.6

5.3 4.9 12.7 0.9 0.5 6.8 13.9 9.5

2.2 2.1 12.7 0.9d 0.5d 3.8 5.6 3.5

SFa-PC

Total

1.6 1.2 4.0 0.4 0.2d 2.3 4.7 1.8

19.4 20.8 64.0 3.9 2.6 33.2 59.4 35.4

a For each FA incubation 5.0 µCi was used. The amounts (nmol) of FA used are different because the specific radioactivities of FA are different. b Fa is the FA shown in the FA column on the Table at the sn-2 position of PC. c 2-Ricinoleoyl-PC and 2-linoleoyl-PC in the incubation of [14C]-oleate. d Estimated from the ratio of the levels of molecular species of 2-oleoyl-PC. ND, not detected; R, ricinoleate; Ln, linolenate; L, linoleate; P, palmitate; O, oleate, S, stearate.

For example, in the incorporation of [14C]ricinoleate into PC as shown in Table 1, the levels of label were 1-linoleoyl-2-ricinoleoyl-PC (7.6) > 1-palmitoyl-2-ricinoleoyl-PC (5.3) > 1oleoyl-2-ricinoleoyl-PC (2.2). This order, reflects the relative levels of the six molecular species of lysoPC in castor microsomes. In previous reports of microsomal incubations of [14C]oleoyl-CoA from soybean (11), potato tuber (12), and pea leaf (13), the major PC molecular species also contained linoleate and palmitate at the sn-1 position. The ratios of the six PC incorporating each of the six [14C]FA were similar and thus the activity of the enzyme seems unaffected by the difference of FA at sn-1 position on lysoPC. The total [14C]-label (Table 1) from each [14C]FA was incorporated in the order: oleate > linolenate > palmitate > linoleate > stearate > ricinoleate. The acyl-CoA:lysoPC acyltransferase activity also may be affected by which FA more readily forms acyl-CoA, since acyl-CoA synthetase may be specific for certain FA. In soybean microsomes, acyl-CoA:lysoPC acyltransferase was highly specific for oleoyl-CoA and linoleoyl-CoA over palmitoyl-CoA and did not show any selectivity for the various lysoPC molecular species (11). In the incubation of [14C]oleate (Table 1), the proportion of label incorporated into the six 2[14C]oleoyl-PC with different FA at sn-1 was similar to the proportion of the six 2-[14C]ricinoleoyl-PC and that of 2-

[14C]linoleoyl-PC from the 2-[14C]oleoyl-PC. Thus, the activities of both oleoyl-12-hydroxylase and oleoyl-12-desaturase seem not to be affected by the different FA at sn-1 of 2-oleoylPC, the immediate substrate of these two enzymes. The same conclusion was also drawn for the activity of oleoyl-12-desaturase in the microsomes of potato tuber (12) and pea leaf (13). However, the effect on oleoyl-12-hydroxylase activity has not previously been reported. Oleoyl-12-hydroxylase from Ricinus communis L. is an oleoyl-12-desaturase homolog (15). Time course for production of the molecular species of [14C]PC. The levels of labeled molecular species of PC at various incubation times were also determined. The ratios of the levels of six molecular species of PC containing the same sn-2 FA from the incubations of the six [14C]FA at various times were similar to the ratio in Table 1. It seems that the activity of phospholipase A2 in castor microsomes, one of the key enzymes in castor oil biosynthesis, is not significantly affected by the difference of these FA at the sn-1 position of PC. However, ricinoleate is preferentially removed from the sn-2 position of PC by phospholipase A2 (2). Identification of the molecular species of [14C]PE. Molecular species of PE were identified and designated as those of PC. The RT of minor [14C]PE from [14C]oleate incubation cannot be matched with our calculated RT of 2-ricinoleoyl-PE

TABLE 2 Level of Incorporation of Radiolabeled FA into Molecular Species of PE in Castor Microsomal Incubations (60 min) Fatty acid (nmol)a Ricinoleate (91) Stearate (114) Oleate (96) Linoleate (98) Linolenate (96) Palmitate (89) a

RFa-PEb

LnFa-PE

LFa-PE

nmol PFa-PE

OFa-PE

SFa-PE

Total

0.01 0.08 0.35 0.32 0.09 0.21

0.01 0.34 0.58 0.46 0.50 0.26

0.28 1.48 4.65 6.26 3.17 2.06

0.22 0.07 3.21 3.94 3.01 0.27

0.02 0.17 0.41 0.56 0.54 0.30

ND 0.02 0.20 0.43 0.20 0.04

0.54 2.16 9.40 11.97 7.51 3.14

See Table 1 for description of experimental conditions. Fa is the FA shown in the FA column on the Table at the sn-2 position of PE. For abbreviations see Table 1.

b

Lipids, Vol. 37, no. 10 (2002)

MOLECULAR SPECIES OF PC IN CASTOR OIL BIOSYNTHESIS

and 2-linoleoyl-PE. They may be the molecular species of other lipid classes, which elute near PE on the silica HPLC including PI, PA, and PS (see Fig. 1, Ref. 2). That 2-[14C]ricinoleoyl-PE and 2-[14C]linoleoyl-PE were not detected agrees with our earlier conclusion (3) that 2-oleoyl-PE is not a substrate for oleoyl-12-hydroxylase and oleoyl-12-desaturase. Although 2-oleoyl-PE is not a substrate under these isolated conditions, we cannot yet rule it out as a substrate in vivo. Incorporation of label into molecular species of [14C]PE. The radiolabeling of molecular species of PE incorporating the six [14C]FA in castor microsomal incubations (60 min) was lower than those of PC as shown in Table 2. Except for the PE products incorporating [14C]stearate and [14C]palmitate, PE with linoleate and palmitate at sn-1 position were the major PE, as for PC. Unlike PC, the levels of molecular species of PE incorporating [14C]ricinoleate were much lower than the other five [14C]FA. The labeling of PE containing only saturated FA was also very low. The total incorporation of label was in the order: linoleate > oleate > linolenate > palmitate > stearate > ricinoleate. 2-Oleoyl-PE seems less important than 2-oleoyl-PC as an intermediate in the biosynthesis of castor oil, not only because it has a lower radioactivity level but also because it cannot be hydroxylated to 2-ricinoleoyl-PE (3). Both 2-ricinoleoyl-PC and 2-ricinoleoyl-PE can be used as the substrates of phospholipid:DAG acyltransferase for the biosynthesis of triricinolein (16). In conclusion, we have identified and quantified 36 molecular species of PC and 35 molecular species of PE in castor microsomal incubations of six [14C]FA individually. Both PC and PE are intermediates in the biosynthetic pathway of castor oil (3). The schematic overview of the plant lipid biosynthetic pathway and its compartmentation have been reported (17). The results here show that (i) 2-oleoyl-PC is actively formed as the immediate substrate of oleoyl-12-hydroxylase, comparing the six endogenous FA; (ii) 2-ricinoleoyl-PC is mostly formed by the hydroxylation of 2-oleoyl-PC, not from the incorporation of ricinoleate into 2-ricinoleoyl-PC; and (iii) 2-oleoyl-PE is not actively formed for the biosynthesis of castor oil. REFERENCES 1. Achaya, K.T., Craig, B.M., and Youngs, C.G. (1964) The Component Fatty Acids and Glycerides of Castor Oil, J. Am. Oil Chem. Soc. 41, 783–784. 2. Lin, J.T., Woodruff, C.L., Lagouche, O.J., McKeon, T.A., Stafford, A.E., Goodrich-Tanrikulu, M., Singleton, J.A., and Haney, C.A. (1998) Biosynthesis of Triacylglycerols Containing Ricinoleate in Castor Microsomes Using 1-Acyl-2-oleoylsn-glycero-3-phosphocholine as the Substrate of Oleoyl-12hydroxylase, Lipids 33, 59–69. 3. Lin, J.T., Lew, K.M., Chen, J.M., Iwasaki, Y., and McKeon, T.A. (2000) Metabolism of 1-Acyl-2-oleoyl-sn-glycero-3-phos-

995

phoethanolamine in Castor Oil Biosynthesis, Lipids 35, 481–486. 4. Bafor, M., Smith, M.A., Jonsson, L., Stobart, K., and Stymne, S. (1991) Ricinoleic Acid Biosynthesis and Triacylglycerol Assembly in Microsomal Preparations from Developing Castorbean (Ricinus communis) Endosperm, Biochem. J. 280, 507–514. 5. Moreau, R., and Stumpf, P.K. (1981) Recent Studies of the Enzymatic Synthesis of Ricinoleic Acid by Developing Castor Beans, Plant Physiol. 67, 672–676. 6. Lin, J.T., McKeon, T.A., Goodrich-Tanrikulu, M., and Stafford, A.E. (1996) Characterization of Oleoyl-12-hydroxylase in Castor Microsomes Using the Putative Substrate, 1-Acyl-2-oleoylsn-glycero-3-phosphocholine, Lipids 31, 571–577. 7. Lin, J.T., Chen, J.M., Liao, L.P., and McKeon, T.A. (2002) Molecular Species of Acylglycerols Incorporating Radiolabeled Fatty Acids from Castor (Ricinus communis L.) Microsomal Incubations, J. Agric. Food Chem. 50, 5077–5081. 8. Singleton, J.A., and Stikeleather, L.F. (1995) High-Performance Liquid Chromatography Analysis of Peanut Phospholipids. II. Effect of Postharvest Stress on Phospholipid Composition, J. Am. Oil Chem. Soc. 72, 485–488. 9. Lin, J.T., McKeon, T.A., Woodruff, C.L., and Singleton, J.A. (1998) Separation of Synthetic Phosphatidylcholine Molecular Species by High-Performance Liquid Chromatography on a C8 Column, J. Chromatogr. A 824, 169–174. 10. Lin, J.T., Lew, K.M., Chen, J.M., and McKeon, T.A. (2000) Separation of the Molecular Species of Intact Phosphatidylethanolamines and Their N-Monomethyl and N,N-Dimethyl Derivatives by High-Performance Liquid Chromatography, J. Chromatogr. A 891, 349–353. 11. Demandre, C., Bahl, J., Serghini, H., Alpha, M.J., and Mazliak, P. (1994) Phosphatidylcholine Molecular Species Formed by Lysophosphatidylcholine Acyltransferase from Soya Bean Microsomes, Phytochemistry 35, 1171–1175. 12. Demandre, C., Tremolieres, A., Justin, A.M., and Mazliak, P. (1986) Oleate Desaturation in Six Phosphatidylcholine Molecular Species from Potato Tuber Microsomes, Biochim. Biophys. Acta 877, 380–386. 13. Serghini-Caid, H., Demandre, C., Justin, A.-M., and Mazliak, P. (1988) Oleoyl-phosphatidylcholine Molecular Species Desaturated in Pea Leaf Microsomes–Possible Substrates of Oleate-desaturase in Other Green Leaves, Plant Sci. 54, 93–101. 14. Triki, S., Demandre, C., and Mazliak, P. (1999) Biosynthesis of Triacylglycerols by Developing Sunflower Seed Microsomes, Phytochemistry 52, 55–62. 15. van de Loo, F.J., Broun, P., Turner, S., and Somerville, C. (1995) An Oleate 12-Hydroxylase from Ricinus communis L. Is a Fatty Acyl Desaturase Homolog, Proc. Natl. Acad. Sci. USA 92, 6743–6747. 16. Dahlqvist, A., Stahl, U., Lenman, M., Banas, A., Lee, M., Sandager, L., Ronne, H., and Stymne, S. (2000) Phospholipid:Diacylglycerol Acyltransferase: An Enzyme That Catalyzes the Acyl-CoA-Independent Formation of Triacylglycerol in Yeast and Plants, Proc. Natl. Acad. Sci. USA 97, 6487–6492. 17. Budziszewski, G.J., Croft, K.P.C., and Hildebrand, D.F. (1996) Uses of Biotechnology in Modifying Plant Lipids, Lipids 31, 557–569. [Received March 21, 2002, and in revised form October 22, 2002; revision accepted October 23, 2002]

Lipids, Vol. 37, no. 10 (2002)