Biochhnica et Biophysica Acta, 1130 ( I q92) 151 - 159 151 c~,; 1992 Elsevier Science Publishers B.V. All rights reserved 01~7-4838/92/505.11{1

BBAPRO 34151)

3-Oxoacyl-[ACP] reductase from oilseed rape ( Brassica napus)

Phi l ip S. S h e l d o n ", R o y G . O . K e k w i c k :', Co l in G. Smi th b, C h r i s t o p h e r S i d e b o t t o m h a n d A n t o n i R. S labas b.I

Department of Bu~'hemi.~tO', Unircrsio" of Birmingham, Edgbaston. Birmingham ( UK) and h Biosciences Dirision, Unih'~'er Research. ('ohvorth lhm.~,. Sharnbr~u~k. Bedford (UK)

(Received 18 September 1991 )

Key words: 3-Oxoacvl-lACPi reducta.~e: Oilseed rape; Fatty acid synthesis: Nt~lulation gent: (B. napus)

3-Oxoacyl-[ACP] reductasc ( E . C . I . l . l . l l ~ J , alternatively known as /3-kctoacyl-[ACP] rcductase), a component of fatty acid synthetase has been pu r i f ed frt~m seeds of rape by ammonium sulphate fractionation, Procion Red H-E3B chromatography, FPLC gel filtration and high pcrfi~rmance hydroxyapatite chromatography. The purified enzyme appears on SDS-PAGE as a number of 20-31) kDa eoml~ments and has a strong tendcncy to exist in a dimeric form, particularly when dithiothreitol is not present to reduce disulphide bonds. Cleveland mapping and cross-reactivity with antiserum raised against avocado 3-oxoacyl-[ACP] r e d u c t a ~ both indicate that the multiple coml~ments have similar primary structures. On gel filtration the enzyme appears to have a molecular mass of 120 kDa suggesting that the native structure is tetrameric. The enzyme has a strong preference for the acetoacetyl ester of acyl carr ier protein (K m = 3 / z M ) twer the corresponding esters of the model substrates N-acetyl cysteamine (Km = 35 raM) and CoA (Km = 261 tzM). It is inactivated by dilution but this can be partly prevented by the inclusion of NADPH. Using an antiserum prepared against avocado 3-oxoacyl-[ACP] reductase, the enzyme has been visualised inside thc plastids of rape embryo and leaf tissues by immunoelectron microcopy. Amino acid sequencing of two peptides prepared by digestion of the purified enzyme with trypsin showed s t r ing similarities with 3-oxoacyl-[ACP] reductasc from avocado pear and the Nod G gen t product from Rhizobium meliloti.

Introduction

in plants de novo fatty acid synthesis is iocaliscd within the plastids and is catalysed by a system analo- gous to that of E. coli, in which all the components are carried on separate polypeptide chains (reviewed in Ref. I). 3-Oxoaeyl-[ACP] reductase (E.C. 1.1.1.100) catalyses the first reduction step in the fatty acid syn- thesis cycle. It has been purified to homogeneity from spinach leaf [2] and avocado pears [3] and has a l ~ been characterised in safflower seed [4], Euglena [5], and barley leaf [6].

In oil rich tissues such as some fruits and seeds, triacylglycerol is the predominant end product made from newly synthesised fatty acids, in order to meet

this demand, the levels of enzymes involved in the fatty acid synthesis pathway become elevated specifically in these tissues [1,7-9]. In sccds there is evidence that acyl carrier protein can exist in a different form from that of its counterpart in leaf [10]. It is therefore conceivable that this will also be the case for the other components of the syntheta~.

In this paper we describe the complete purification of 3-oxoacyl-[ACP] reductase for the first time from any seed tissue, its catalytic properties and subcellular Iocalisation by immunoclectron microscopy.

Materials and Methods

n Present address: Department of Biological Sciences. University of Durham, Science Buildings South Road. Durham, Dil l 3LE, UK.

Abbreviations: ACP, acyl carrier protein.

Correspondence: P.A. Sheldon. Present address: Department of Biochemistry and Molecular Biology, University of Leeds. Leeds LS2 9JT, UK.

Materials The biotinylated donkey anti-rabbit igG and

horseradish peroxidase-streptavidin complex were from Amersham International Pie. (Rickmansworth, UK). Staphylococcus V8 protease was from Boehringer Ltd. (Mannheim, Germany). The Aquapore RP-300 column was from Anachem (Luton, UK). The sources of other materials are described in Ref. 3.

152

l,ape seed material The collection and storage of seed used for enzyme

purification is described in Ref. 11.

Enzyme and protei~ assays These were as described in Ref. 3 except that in the standard enzyme assay S-acetoacetyl-N-acetylcystea- mine was replaced by 100 ttM acetoacetyl CoA. A unit of activity is defined as the amount of enzyme required to catalyse the oxidation of 1 nmol/min in the stan- dard assay.

P~trification of 3-oxoacyi-[ACP] reductase The initial operations were carried out at 0-4°C.

Rapeseed (100 g) was homogenised with 100 mM sodium phosphate pH 7.4, 4 mM EDTA, l mM D'UI" (400 ml) using either a Waring blender or a polytron (Kinemetica, Lucerne) at maximum power. The ho- mogenate was centrifuged at I00000 × g for 40 min. The precipitate and floating lipid layer were discarded. To the liquid supernatant solid ammonium sulphate (176 g/ l ) was added to 30% saturation. Following centrifugation (15000×g , 20 min), solid ammonium sulphate (351 g/ I ) was added to the supernatant to 80% saturation. After centrifugation (150O0×g, 20 min), the precipitate was dissolved in a small volume of buffer and dialysed against 3 changes of 5 I 25 mM sodium phosphate, pH 6.8, 0.5 mM EDTA, 10 mM 2-mercaptoethanol at room temperature over a period of about 18 h. Precipitated material was removed by centrifugation. The subsequent chromatographic steps were carried out at room temperature.

The extract was applied to a Procion Red H-E3b columm (2.2 × 21 cm) previously equilibrated in 25 mM sodium phosphate pH 6.8, 0.5 mM EDTA, 1 mM D'lq'. The column was washed overnight at a flow rate of -- 1 ml/min and subsequently eluted with a linear gradient of the above buffer leading to 2 M NaCI, 25 mM sodium phosphate pH 6.8, 1 mM DTT over 200 ml (Fig. la). The fraction volume was 6. 3 ml. The pooled active fractions were concentrated with a stirred pres- sure cell (Amicon) using a PMI0 membrane.

Following centrifugation (12000 × g, 10 min) to re- move any insoluble material, 200 #.1 extract was in- jected onto a Superose 12 column (Pharmacia 5 /5 series, V t = 25 ml), previously equilibrated in 100 mM sodium phosphate pH 6.8, 0. 5 mM EDTA, 1 mM DTI" at a flow rate of 0.5 ml/min (fig. lb). The fraction volume was 0.32 ml. The gel filtration step was re- peated until all the PM10 concentrate had been pro- cessed.

The final purification step was hydroxylapatite HPLC using a BioRad HTP column (Fig. lc). The active fractions from Superose 12 chromatography were injected directly onto the column, previously equili-

brated in buffer A (100 mM sodium phosphate pH 6.8, I mM DTT, 10 /xM CaCI 2) at a flow rate of 0.5 ml/min. When the A22 o had dropped to a virtual baseline level, the column was eluted with a linear gradient of 0-100% buffer B (1 M sodium phosphate, pH 6.8, 1 mM DTI', 10/xM CaCI 2) over 100 ml. The fraction volume was !.0 mi.

SDS polyacrylamide gel electrophoresis, Western blotting and fluorography

SDS polyacrylamide gel electrophoresis was carried out as described in [12]. Silver staining of gels was by the method described in Ref. 13. lmmunostaining was carried out as described in Ref. 14.

FPLC and HPLC systems These are described in Ref. 14.

Reductice alkylation of cysteine residues This was carried out essentially as described in Ref.

15 except that 2% SDS was used instead of guani- d inium/HCl as denaturant. Instead of dialysis, the reaction mixture was desalted by reverse phase HPLC using an Aquapore RP-300 column as described in the legend to Fig. 3.

Amino acid analysis and sequencing Protein hydrolysis and amino acid sequencing were

carried out as previously described [14]. The amino acid analysis was kindly carried out by Dr. Alistair Aitken (MRC la0oratories, Mill Hill) using the Picotag method with an Applied Biosystems Model 420 Anal- yser.

hnmunoelectron microscopy This was carried out as described in Ref. 16.

Results

Enzyme stability A 35-80% ammonium sulphate fraction prepared as

described in Materials and Methods and dialysed against potassium phosphate pH 7.0 (20 mM or 100 mM), 0.5 mM EDTA. 1 mM DTT, was incubated for 24 h at either 4°C or 27°C and then assayed for enzyme activity. After incubation at 4°C in 20 mM potassium phosphate the enzyme activity was reduced to 28.3% of its starting level. Higher levels of activity were obtained by incubation under the following conditions: 4°C in 100 mM potassium phosphate (47.9% of starting level), 27°C in 20 mM potassium phosphate (53.8% of starting level) and 27°C in 100 mM potassium phosphate (87.0% of starting level). The enzyme activity is therefore best maintained by higher ionic strength and moderate rather than low temperature. Similar properties were observed in the enzyme from avocado pears [3].

153

Enzyme purification The enzyme was purified 346-fold with a yield of

17% by ammonium sulphate fractionation followed by the 3 consecutive chlomatographic steps which arc shown in Fig. 1 and summarised in Table 1. in order to preserve activity, chromatography was carried out at room temperature using buffers of as high ionic strength as possible. In the final step, the enzyme was bound strongly to the hydroxylapatite matrix, the activ- ity peak eluting at - 470 mM ,sodium phosphate con- currently with a well resolved peak at 220 nm. Since the purified protein was contained at dilute concentra- tion in a very high ionic strength buffer, it was impossi- ble to quantif3/directly because preparation for amino acid analysis almost certainly resulted in significant sample losses.

Analysis of the purified enzyme by polyacrylamide gel electrophoresis in the presence of sodium dodecyi sul- phate

On analysis by SDS-PAGE under reducing condi- tions, the purified enzyme appears as a number of components of molecular mass 20-30 kDa (Fig. 2A, lane 1). A l ~ present is a small amount of a 66 kDa component. In the absence of D'IT the 66 kDa compo- nent predominates, suggesting that it contains non-re- duced disulphide bonds (Fig. 2A, lane 2). The multiple components of the rapeseed end. me are cross-reactive with antiserum raised against avocado 3-oxoacyl-[ACP] reductase which is a single 28 kDa polypeptide (Fig. 2B).

Cleceland mapping of the components of purified 3- oxoacyl-[ACP] reductase

The relationship of the polypeptides visualised on SDS-PAGE analysis of purified "~-oxoacyl-[ACP] reduc- tase was investigated by Cleveland mapping [17]. After hydroxylapatite HPLC, the component polypeptides were separated in two steps: (a) following denaturation with 2% SDS, the hydroxyapatite-eluted enzyme poly-

A=,o

1.5

1.0

0.5

0.0

A " ,,..s 3000

. " " [Activity] / (U/ml)

..- 1ooo

to - " fo so ,o ° Fraction No. (6.3ml)

t= ":: 1

o.o ~ - - - - - - ~ - - - ~ - - - ~ - - ~ - - - ~ - - ~ o .t'! ",.. ..... 1

0 10 20 30 40 50 60 70 80 90 Fraclion No. (0.32ml)

2.0 6000 C ; " :"

1.5 :: .. .. :. 4 0 ~ " " lAclivity]

1.0 . ' . ' i (U/ml)

o.0 = / ~ ' ........ " "" .... 0 0 10 20 30 40 f,o 60 70

Fraction No. (t.0ml)

Fig. I. Purification of 3-oxoacyl-lACP] reductase (a) Procion red H-E3B Sepharos¢ ch roma tography of the dialysed 3 0 - 8 0 % ammo- n ium su lpha te fraction. T h e pass (850 ml, A2~ ~ = 1.07, no detectable activity) is not shown. Elut ion with a linear salt gradient over 200 ml as descr ibed in the text c o m m e n c e d simultaneously with the start of fraction collection. (b) F P L C gel filtration on Superose 12. (c) hydrox3,apatite HPLC. The shaded areas represent the pooled frac-

tions. [] or ( . . . ) , A2~,; I , [Activity].

T A B L E I

Ptvification of rape seed 3-oxoacyl-[ACP/ reductase

T h e quant i ty o f s tar t ing mater ia l used was 100 g.

Fract ion Volume A 2~ Total (ml) protein

(rag)

Total Specific Purification Yield

activity activity ( ~ ) (U ) ( U / m g )

Superna tan t 455 - 440 ~

4 5 - 8 0 % (NH4)2SO 4 49 - 392 a Dialysate 64 243

Procion red H-E3b Sepharose 70 0.56 27 b

Supcrose 12 FPLC 12.6 0.42 5.3 b Hydroxyapati te H P L C I 1.0 0.020 0.22 b

173 000 393 ;' I 100 147000 376 ~ 0.96 85 I 110O0 456 ;' 1.2 64

75 200 2 790 b 7. I 43 38 8{X) 7340 h 18.7 22 29900 136000 b 346 17

a Protein de te rmined by Bradford method. b Protein de te rmined by a s suming a i% solution has an A2,u) of I0.0.

154

A

1 2 k D a

l

• 66

~ 4 5

24

B

1 2 kDa

. ~ - - ' - 66

i": i ":. . . . . • - - 4 4

- - 2 4 . 5

Fig. 2. Analysis of purified 3-oxoacyl-[ACP] reductase by SDS-poly- acrylamide gel clectrophoresis. (A) a 12% gel stained with Coomassie Blue R250. Lane !, 20 mM D'vr present; lane 2, no D'vr present. (B) crossreactivity of rape seed 3-oxoacyl-[ACP] reductase against antiserum raised against the enzyme from avocado pear. Western blot of a 10% SDS-polyacrylamid¢ gel. Lane 1, purified rape seed

enzyme; lane 2, purified avocado enzyme.

peptides were partially resolved by reverse-phase chro- matography on Aquapore RP-300 (Fig. 3); (b) each set of components was subdivided further by SDS-PAGE. After staining with Coomassie Blue R250 and cutting out the gel slices [17], five components were resolved (Fig. 4 lanes 1-51. The tendency of the protein to form as dimers during SDS-PAGE is clearly observed. The

mobilities of the dimeric species correlate with the change in mobilities of the monomers. The dimeric bands must have formed from the monomeric bands at some time between fixing of the first gel and eicc- trophoresis of the second. The components were mapped using Staphylococcus aureus V8 protease (Fig. 4 lanes 7-111. The five polypeptides produced similar digestion patterns, two bands being common to all the polypeptides. This indicates that after the hydroxyapa- pite chromatography purification step, the multiple components detected on SDS-PAGE have related pri- mary structure. Either they may have arisen from a single initial component by proteolysis in the seed or during the purification procedure; alternatively they may each be encoded by distinct mRNAs. The inclu- sion of 2 mM PMSF in the initial homogenization mix

0.0

' 11- 11

3 _ 11~ 11

6 ~ 4 ~ 1 1 ~ - - I ~ I

IiIi111111 V l

2.0 4.0 Elufi~ Volume (~)

6.0

I 2 3 4 5 6 kDa I

66

I l l l ram' 29 - - -~24

Fig. 3. Partial separation of rape seed 3-oxoacyl-[ACPJ reductase components by reverse phase semi-microbore HPLC on Aquapore RP-300. To hydroxyapatite-eluted 3-oxoacyl-[ACP] reductase was added solid SDS (to 2% w/v) and D'VI" to 20 raM. Following equilibration of the column in 0.1% TFA (v/v) in 40% (v/v) (pro- pan-2-ol/acetonitrile 2/1) (aq), the protein containing mixture was injected. The protein wa~ eluted with a linear gradient to 0.1% TFA (v/v) in 60% (v/v) (propan-2-ol/acetonitrile 2/1) (aq) over 6 ml at flow rate of 100/zl/min (A). Fractions labelled 2-6 were collected separately, lyophili~d and analysed by SDS-PAGE (st!vet-stained 13% gel) (B). Lane 1, total hydroxyapatile eluted enzyme, dialysed against 2(1 mM N-ethyl morpholine/acetate pH 8.0 and lyophilised.

TABLE II

Amino acid composition of rape seed 3-oxoa('yl-lACP] re~htctas,, am/ comparison with the enzyme from .~'pitutch h'af and at'o('aeh) mesocurp

Amino acid Residues/subunit

Rape seed Avocado Spinach leaf (28 kDa) mesocarp (24.2 kDa)

(28 k D a l

Asx 21 26 23 Thr I0 16 15 Ser q 15 13 GIx Z~ 32 13 Pro 9 8 16 Gly 311 28 30 Ala 311 36 23 Val 26 28 23 Met 8 I 2 lie 211 23 14 Leu 21 19 2(I Tyr 9 2 6 Phe 14 4 19 His 3 I} h Lys 18 21) 15 Arg I i 9 3 Cys 3 3 2

had no effect on the appearance of the purified en- zyme when analysed by SDS-PAGE.

Native molecular mass and subunit structure in the Superose 12 gel filtration step used in the

purification, 3-oxoacyl-[ACP] reductase eluted at a po- sition corresponding to a native molecular mass of 120 kDa. The avocado enzyme elutes at a position corre- sponding to a native molecular mass of 130 kDa and

155

100

8o e~

40 . . . . . . t ~ .

20 • •

0 . . . . . . . . . . . . .

10 20 3 o 4 o 51] "lime afl~" diluli~'~ (mini

Fig. 5. Inactivation of rape seed 3-oxoacyl-[ACP] reductase by dilu- lion. Hydrox3'apatite-eluted enzyme was diluted I : IIX) into Ill mM MOPS pll 7.11. I mM DTT. I mg/ml BSA. +varying concentrations of NADPlt . a . NADPll absent: I I . It) #M NADPH; El. 10i)

g M NADPit .

consists of a singlc 28 kDa polypcptide leading to the conclusion that it is tetramcric [3]. Despite the hetero- geneity of its polypeptide composition, the rapeseed enzyme is also likely to be tetrameric. The species containing disulphide bonds (Fig. 2B) could either exist in vivo or alternatively they have arisen during sample preparation.

Total amhto acid analysis Following hydrolysis, the amino acid composition of

rapeseed 3-oxoacyl-[ACP] reductase was determined, it is shown in Table II and compared with those previ- ously determined for the enzymc from spinach leaf [2] and avocado pear [3]. There are some significant differ- ences, particularly in the content of methionine, tyro- sine and phenylalanine. Although the tryptophan con-

1 2 3 4 5

F

z

- - . . , j

6 7 8 9

{

10 "1

& ~ . - •

' f .

Fig. 4. Cleveland mapping of rape seed 3-oxoacyl-[ACP] reductase polypeptides. The individual polypeptides hands were cut out of a similar Coomassie Blue R250-stained gel (10rA) to Ihal shown in Fig. 3b and rerun on 15~/; polyacr~lamide gels [20]. When the dye froni had reached the stacker/separating gel interface, electrophoresis was suspended for 30 rain in order to allow digestkm to take place. Polypeptides were visualised by silver-staining. Lanes 1-5, the separated undigested 3-oxoacyl-[ACP] reductase polypeptides; lane 6, Staphyloc¢u'cus V8 protease (50 ng): lanes

7-11, the separated 3-oxoacyl-[ACP] reductase polypeptides, each partially digested with 50 ng Staphylococcus V8 prolease.

156

tent was not determined, it seems likely that the E2~4) of the rape seed enzyme is much greater than that of the avocado enzyme because of the difference in tyrosine content.

Catalytic properties of rapeseed 3-oxoao,l-[ACP] reduc- tase

The enzyme is inactivated by dilution but is partially protected by the inclusion of NADPH (Fig. 5). Similar properties were observed in the enzymes from avocado pear and E. coil [3,18]. With the model substrate acetoacetyl coenzyme A concentration maintained at 100/LM, the enzyme has an apparent K m of 23 ~tM for NADPH (Fig. 6A). This value compares with 25 ,~M for ~.hc enzyme from spinach leaf [4], 21 /~M from E. coil (using saturating levels of acetoacetyi-[ACP] as thioester substrate) [18] and 5.6 p.M for the avocado enzyme (with 5 mM acetoacetyI-N-acetylcysteamine) [31.

The Kms for three thioester substrates were deter- mined. In order to measure activity it was necessary to use enzyme solutions at different dilutions for the various thioester substrates. Since this also would have resulted in a variable amount of inactivation as de- scribed above, it was not possible to obtain meaningful comparitive values of V~. When the concentration of NADPH is maintained at 100 p,M, the KmS for the acetoacetyl esters of N-acetylcysteamine and ace- toacetyl coenzyme A are 35 mM and 261 /,LM (Fig. 6B,C) respectively, and are comparable with those measured in avocado and spinach [2,3]. The rapeseed enzyme has an apparent K= of 2.6 /~M for ace- toacetyi-[ACP] compared with 3.7/z M for the spinach leaf enzyme and 7.9 #M for the avocado enzyme (Fig. 6D). Thus, as would be expected, the enzyme has a higher affinity for its natural substrate.

Immunoelectron microscopic localization in seed and leaf material

The antiserum raised against avocado 3-oxoacyl- [ACP] reductase [3] was used for immunoelectron mi- croscopy of embryo and leaf tissue from rape (Fig. 7A,B). In both cases the enzyme appears to be Io- calised within the plastids supporting previous evi- dence that fatty acid synthesis takes place in these organelles [19-23]. There does not appear to be any preferential distribution between the stroma and thyl- akoid membranes.

& lS0

t2S

t00

s(~v) 7s

5O

2S

0

-25 0

I m

5 I() ll6 20 SN (rain t ' x 10~

B 1GO

s (mu) as

0

-2S

2.0 s'0 ¢o 7'.o io do ~ (rain t ' x 10~

C 12S

1.00

0.7S

s(mu) 0.5o

02S

0.20

-02S

SN (n~n t ' • 10')

11 40

3O

2O s0a~

10

0

-10

Fig. 6. Kinetics of rape seed 3-oxoacyl-[ACP] reductase. (A)[NADPH! varied, [AcAcCoA] maintained at 100 p.M. (B) [AcAcNAC] varied, [NADPH] maintained at I(X) P.M. (C) [AcAcCoA] varied. [NADPH] maintained at 100 p.M. (D) [AcAcACP] varied, NADPH maintained

at IO0 p.M.

Amino acid sequencing Following digestion with trypsin and separation by

reverse phase HPLC, amino acid sequencing of two tryptic peptides of rapeseed 3-oxoacyl-[ACP] reductase showed similarities with sequences from avocado meso-

carp 3-oxoacyl-[ACP] reductase [3] and the Nod G gene product from the nitrogen fixing bacterium Rhi- zobium meliloti [24] (Fig. 8). Amino-terminal sequenc-

157

!

o b •

',~, ~ 3 ~ - - " " v -

Fig. 7. Immunogold localisation of 3-oxoacyl-[ACP] reductase: (A) rape embryo, pl, plastids: ob. oil body: cw cell wall: nuc, nucleus. Bar = I p.m: (B) rape leaf. pl, plastids: cyt, cytoplasm: or, cell wall: mit; milochondrion. Bar --- 1 p.m.

158

Rapeseed 3-oxoacyl-[ACP] reolu~ tryptic peplide 20

Avocado mesoearp 3.oxoacyl-[ACP] reduclase ~ I:Hllptide 26

Nod G (P, hizol;)mm melilob) 9ene product, residues 73-92

D A V D Y W G Q I OV I A N N A G I T

? A V D A W G Q V D V L I N N A G I T R

R A E A D L E G V O I L V N N A G I T G

Rapeseed 3-oxoacyl.[ACP] recluctase tp/plic pepllde 6

Rapeseed 3-oxoacyl-[ACP] re(luctase tp/plic peplide 11

Avocado mesoca~ 3-oxoecy1-[ACP] reductase t r~ ic peplide 9

I MMK

E A D V D A M M K

e^ovE ,,(

Fig. 8. Amino acid sequencing of rape seed 3-oxoacyl-[ACP] reductase tryptic peptides and compari~m with .sequences from the avocado mesocarp enzyme and the Ntv, l G gene product of Rhizobium meliloti.

ing of the purified enzyme preparation suggested the presence of at least 3 termini.

Discussion

In rape seed, 3-oxoacyl-[ACP] reductase activity is induced at the same time as many of the other compo- nents of fatty acid synthetase [1] when storage lipid synthesis is at a maximum. It is possible that the syntheses of the separate components are under com- mon developmental control. The enzyme level de- creases during the desiccation stage of development. It is therefore critical when purifying 3-oxoacyl-[ACP] reductase to use seed harvested at the correct develop- mental state.

The enzyme was purified by ammonium sulphate precipitation, Procion Red H-E3b chromatography, Su- perose 12 FPLC and hydroxylapatite HPLC. In the final step the activity elution profile correlated exactly with a well resolved peak at 220 nm. This would be expected if the enzyme was pure after this step. How- ever on analysis by SDS-PAGE, the preparation con- sisted of a number of components of molecular mass 20-30 kDa. Cleveland mapping and immunoblotting using antiserum raised against the avocado enzyme indicated that all these components have similar pri- mary structure and might therefore be derived from a single parent polypeptide. It may be significant that another component of fatty acid synthetase, enoyl- [ACP] reductase is also purified from rape seed as two closely running polypeptides of M r 34.8 kDa and 33.6 kDa. The 33.6 kDa species may have arisen by cleavage of a 6 amino acid peptide from the N-terminus of the 34.8 kDa species [11,14]. Recent evidence using im- munological procedures has demonstrated this to be the case [25]. Proteolytic degradation has also been encountered during the purification of acetyl CoA car- boxylase from rape seed [26].

In common with 3-oxoacyl-[ACP] reductases from avocado mesocarp and E. coil, the rape seed enzyme is inactivated by dilution and is protected by NADPH. It

seems that binding of NADPH converts it to a more stable conformation.

The enzyme has a marked preference for thioesters of acyl carrier protein over these of N-acetylcystea- mine and coenzyme A. This distinguishes it from other enzymes catalysing similar reactions which could also be present in plant tissues but which are probably involved in other pathways [27-29].

]mmunoelectron microscopy indicates that the en- zyme is located within the plastids of rape embryo and leaf tissue. Unlike ACP which may be associated with the thylakoid membrane [16], 3-oxoacyl-[ACP] reduc- tase does not appear unevenly distributed between the thylakoids and the stroma. It may be significant that ACP is used not only as a component of fatty acid synthetase but also as a cofactor in the synthesis of galactolipids in thylakoid membranes [30].

The Rhizobium meliloti Nod G gene product which possesses a similar sequence at residues 73-92 with the rape seed 3-oxoacyl-[ACP] reductase tryptic peptide 20 (Fig. 8) has previously been implicated to play a role in a process similar to fatty acid synthesis. The rape seed tryptic peptides 6 and 11 differ in length, yet share the conserved sequence MMK. However, residues preced- ing this motif are different. This suggests that the enzyme preparation either may contain a number of gene products, or alternatively a single gen~ product could contain an internal duplication.

Acknowledgements

P.S.S. wishes to thank SERC and Unilever Research for support by a CASE studentship.

References

I Slabas, A.R., Hellyer, A., Sidebottom. C.M., Bambridge, H., Cottingham, I.R., Kessel, R., Smith. C.G.. Sheldon, P., Kekwick, R.G.O,, de Silva, J., Lucas. C., Windhurst, J., James, C.M., Hughes. S.G. and Safford, R. (1988) in NATO Advanced Study Institute, Series A, Vol. 140: Plant Molecular Biology (Wenstein, D. and Chua, N., eds), pp. 265-277. Plenum Press, New York.

2 Shimakata. T. and Stumpf, P.K. (1982) Arch. Biochem. Biophys. 218. 77-91.

3 Sheldon. P.S., Kekwick, R.G.O.. Smith, C.G.. Sidebottom, C.M. and Slabas. A.R. (1990) Biochem. J. 271,713-720.

4 Shimakata, T. and Stumpf. P.K. (1982) Arch. Biochem. Biophys. 217. 144-148.

5 ttendren. R.W. and Bloch, K. (1980) J. Biol. Chem. 255. 1504- 1508,

6 H¢j, P.B. and Mikkesen, J.D. (1982) Carisberg Res. Commun. 47. 119-141.

70hlrogge, J.B. and Kuo. T-M. (It~.14) Plant Physiol. 74, 622-625. 8 Simcox, P.D., Garland, W., De Luca, V., Canvin. D.T. and

Dennis, D.T. (1979) Can. J. Bot. 57, 11}08-1014. 9 Turnham, E. and North¢ote. D.lt. {1983) Biochem. J. 21~ 223-

220. 10 Saflrord R, Windhurst, J.H.C.. Luca.s, C., De Silva, J., James,

C.M., Hellyer, A., Smith. C.G., Slabas, A.R. and Hughes. S.G. (1988) Eur. J. Biochem. 174. 7,87-295.

!1 Slabas. A.R., Sidebottom. C.M., Hellyer. A.. Kessel. R.M.J. and Tombs, M.P. (1986) Biochem. Biophys. Aeta 877. 271-280.

12 Laemmli. U.K. (1971) Nature 227. 680-685. 13 Morissey. J.H. (1981) Anal. Biochem. 117. 307-310. 14 Cotlingham, I.R.. Austin. A.. Sidebottom, C. and Slabas, A.R.

(1988) Biochim. Biophys. Acta 954. 201-207. 15 John~n, D.M.A.. Gagnon. J. and Reid, K.B.M. (1980) Biochem.

J. 187, 863-874.

159

16 Slabas. A.R. and Smith. C.G. (1988) Planta 175. 145-152. 17 Cleveland, D.W. (1983) Methods Enzymol. 96. 222-229. 18 Sehulz, H. and Wakil, S.J. ~1971) J. Biol. Chem. 246. 1895-1901. 19 Ohlrogge. J.B.. Kuhn, D.N. and Stumpf, P.K. (1979) Proc. Natl.

Aead. Sci. USA 76, 1194-1198. 20 Weaire. P.J. and Kekwick. R.G.O. (1975) Biochem. J. 146, 439-

445. 21 Zikley, B.F. and Canvin, D.T. (1972) Can. J. Bot. 50, 323-326. 22 Nakamura, Y. and Yamamada, M. (1979) Plant Sci, Lett. 14,

291-295. 23 Vick, B. and Beevers, H, (1978)Plant Physiol. 62, 173-178. 24 Debell~. F. and Sharma, S.B. (1986) Nucleic Acids Res. 14.

7453-7472. 25 Slabas. A.R.. Cottingham. I.R., Austin, A.. Hellyer, A.. Safford,

R. and Smith, C.G. (1990) Biochim. Biophys. Acta 1039, 181-188. 26 Slabas, A.R. and Hellyer, A. (1985) Plant Sci. 39, 177-182. 27 Fukui. T.. Ito. M., Saito, T. and Komita, K. (1987) Biochim.

Biophys. Acta 917, 365-371. 28 Ernst-Fonberg. M.L. (1986) Plant Physiol. 82, 978-984. 29 Caughey. I. and Kekwick, R.G.O. (1982) Eur. J. Biochem. 123,

553-561. 30 Douce, R., Dome, A.J.. Block, M.A. and Joyard. J. (1984) in

Chloroplast Biogenesis (R.J. Ellis, ed.) pp. 193-224. Cambridge Univ. Press. Cambridge.