Plant Phi siol. (1968) 43, 1637-1647

Fat Metabolism in Higher Plants XXXVI: Long Chain Fatty AcidSynthesis in Germinating Peas'

Michael J. K. Macey2 and P. K. StumpfDepartment of Biochemistry and Biophysics, University of California, Davis, California 95616

Received May 20, 1968.

A bstract. A low lipid, h.igh starch containing tissue, namely cotyledons of germinatingpea seedlings was examined for its capacity to synthesize fatty acid. Intact tissue slices readilyincorporate acetate-14C into fatty acids from C16 to C,4. Although crude homogenates syn-thesize primarily 16:0 and 18:0 from malonyl CoA, subsequent fractionation into a 10,00Ogpellet, a 105g pellet and supernatant (soluble synthetase) revealed that the 105g pellet readiilysynthesizes Cl. to C28 fatty acidcs whereas the 10,000g and the supernatant synthesize primarilyC16 and C18. All systems require acyJ carrier protein (ACP), TPNH, DPNH if malonylCoA is the substrate and ACP, Mg2, Co,, ATP, TPNH, and DPNH if acetyl CoA is the sub-strate. The cotyledons of germinating pea seedlings appear to have a soluble synthetase and10,000g particles for the synthesis of C6 and C18 'fatty acid, and 105g particles which specificallysynthesize the very long chain fatty acid from ma-lonyl CoA, presumably via malonyl ACP.

Previous studies of fatty acid synthesis in plantshave usually utilized tissues such as castor beanendospermi and avocado mesocarp whliclh store largeamounts of lipid as a food reserve. Though suchtissues have obvious advantages and yield synthetasesystenis of highi activity, they must of necessity bespecialized towards the production of fatty acidstypical of storage lipids. This paper reports on theproperties of fatty acid synthesizing systems fromPisutpn satizumZ, a low li,pid containing seed.

Hawke anid Stumpf (10) working with barleyleaves, showed that both green and etiolated tissuewere capable of synthesizing from acetate fatty acidswith a chain length up to C24 as well as the usualsaturated and unsaturated fatty acids in the C14,1Rrange. Attemiipts to detect elongation of Cl,; andC1, labeled fatty acids in the presence of tissue slicesand unlabeled acetate were unsuccessful. All thesubstrate re-mained in the form supplied olr wasbroken down completely and recovered as CO.).However, decarboxylation studies on the acetate fedtissue synthesized fatty acids indicated that elonga-tion was a component of the long clhain fatty acidsynthesisibey-ond a chain lengtll of Cl1a. Kolattukudy(12-13) working with Brassica leaf tissue obtainedevidence that the C., hydrocarbon nonacosane couldbe produced by chopped leaf tissue direct from eitherpalmitate or steirate, the latter being the more

1 This investigation was supported by the UnitedStates Department of Agriculture through AgriculturalResearch Service, 'Western Utilization Research and De-velopment Division, Albany, California, under Grant No.7990-74 anld National Science Foundation GB 5879X.

2Present address: Department of Botany, Universityof New Southli Wales, Kensington, N.S.W. 2033, Aus-tral ia.

effective substrate. Some label was also found infatty acids tup to C0 wlihel pahlnlitate or stearatewere fed. Hawke and Stumpf (10-11) found thatnearly 10 % of the radioactivity fromipalmitate-1-'1Cappeared as CO., when fed to chopped barley leaves,but Kolattukudy (11) foundl only a fraction of 1 %under very similar conditions.

Hawke and Stumpf (10-ll) reported that thevery long chain fatty acids (chain length >C18 )were located in the particulate cellular materialfollowing biosynthesis from acetate, but cell freesystems capable of synthesizing these componentswere not obtained.

Guckhait et al. (8) have stated that C18 2(2 satu-rated and unsaturated fatty acids are formed by anelongation process in pigeon liver microsomes, withmalonyl CoA, supplied exogenouslv, reacting withthe endogenous fatty acids. Abraham et al. (1)and Lorch et al. (13) have also demonstrated thepresence of microsomal systems.

Yang and Stumpf (20) using avocado mesocarptissue, distinguished between a supernatant and aparticulate 12,000g system. Both systems synthe-sized palmitate and stearate from malonyl CoA, andthe particulate system in addition synthesized tracesof C,8.1. Unlike the supernatant systems found inmost preparations from an'imal tissues, the super-natant system synthesized more stearate than palni-tate (about 6-4). The particulate preparation couldutilize acetyl CoA effectively, while the supernatantsystem preferred malonyl CoA.

Since pea cotvledon tissuie is readily available inlarge quantities and has been shown by Glew (5)to be capable of considerable lipid synthesis fromacetate, it was selected to study the fatty acidssynthetase activity of a low lipid tissue. The em-phasis in this work has been on the production ofvery long chaini fatty acids (>C,).

1637 www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

Materials and Methods

TISSUE SLICfES. (rowtlh Coniditionis of Peas.Peas were growvn for periods up to 8 (lays in per-forated pl)astic trays und(ler a spray of water at roonmtemiiperature. Undler tllese conlition-s nlo sign of bac-terial or fungal contamination was apparent l)roN-idedthe seeds were in a single lay er anid exposed directlvto the spray Only the cotyledons were used for

preparing tisstie slices.Jvlcu1bationi of 7Tissite Slieces. 'or incubation of

tissue slices with acetate-l-W4C. tissule slices were

ctit about nmm thick wA ith a razor blade, and washe(din distilled water unitil the w\ashlinigs were clear.Two gramils (fr xx-t) of thle tissue were iimmersed in4 mil of a solution containiing 200 ,umoles potassiumphospha,tte and 100 xlmloles potassium bicalrhonate atpTJ 7.2. The mixture plus radioactive substrate wvasplace,d in a 50 ml conical flask sealed by a serumncap to which a small polyethylenee culp w-as fitted fortral)pinig CO., at the enid of the Hicuhiatioil period,the latter beinig lihr in-i all exlper-ieiits with tissue

slices.At the endl of thle inicuibation period, 2 nil 10 N

1%..SO, was injected into thle flask throuighi the serunicap, and 0.2 ml Hyani-ine lOX iiito the polyethyleneClup. 14CO.2 estiniatiolis were miade according to themethod described by HAawAke aii(l Stumipf (10).

In order to estimiate the amiount of substrateremaininii-g inl the aiiihienut soluitioii. the tissue wvaswashed withl 100 ml1 11'.,0 aIndI three 0.2 nil ali(quotscotliite(l inl Brav's soluitioil oil the scilntillatiol s-pec-tronleter.

lipid E.r1raction a1(d Th;in Laey- (CIi,romn(atog-r(aphly of f.ipid. Total lipid5 wN-ere extracte(l froiiithe tissuie with 2:1 chlloroforimi/iiietilaiiol, using 10 mlof solvenit for 2 g tissuie. The mixturi-e was shakengently for 12 hr in a stopperedl flask at roomii teml-peratuire, aniid at the ncld of this period the tissuewas straiiied off ainid washed withi a furthier 10 ml

of solveiit. The washii-igs were bulked w\-ith theniain extract, 20 iiil HT-O wAere added to the cliloro-form-methaniol extract ancd the chiloroforml layer wxasseparatedl and driel ov-er anhvdrous Na.,SO,.

Aliquots of total lipid were chromi-iatographed oilorated in scintillatioll vials, a-id counted iii toltuenescintillator to obtain total lipid incorpooration. Thepreseiice of tunchaniged acetate w-as detected by acidi-fying 1 aliqtuot withl glacial acetic acid before evap-oration of the solvent, aud heatiiig in a streaiml ofair at 80c.

Aliquiots of total lipi(d were chiromiiatographed on

0.33 mi-i thiiii layer plates tising the solxent suggeste(dby Freeilan aiid \Vest (4), consisting of ether:ben-zene ethanol :acetic acid, 40:50:2.0 :0.2. This solventseparates tri- anid diglycerides, free acids, the polarcomplex lipids remiaiiiing on the origini. The samesolven,t as tusedI for preparative TLC on I mm

plates to analyze the occulrrelice of dlifferent chainlengthis, in the (lifferelit classes thils distingulished.

No attempt NNas mclade to analYze the nature of thepolar complex lipids.

AMethyl e4ters were I)repared fromii the cru(le lil)iextract by tr-aiis-esterification with BF.,-nmethanolsolution. Methyl esters of fatty acids were analyzedby- thin layer chromiatography on Silicagel ( ulsiiig6:4 hexane-ether as the solvent (it)) alnd also onsilver nitrate-impregnated Silicagel G to separateunsatturate(l fatty acid methyl esterss ( 17). Ninetyto 95 % of the radioactivity w-as associated witleither straiglht chain satturated or unsaturated methylesters, and the rest of the activity was oni or closeto the origin. Standards use(l incluided methyl estersof stearate, palmitate, oleate. and ,B-hvdroxy-laurate.There was no evidence of appreciable radioactivitvappearilig in hvdroxv-acids and radiochlelilcal anial-vsis was perforimed on1 the unpurified methyl esters.

Preparationi of Homlogcnatcs nd C-eli hFra ctioll.s.Honmogenizing wals carried ouit in a pestle and miiortarat 00. The tissuie coukl be hlomlogeniized to a smoothpaste withcout the use of abrasives. Wh'ien lhomiog-eniizationi was carried ouit in a bleni,der (oiiiomnilix)the actix-itv of the lOO,OOOj pellet was severely re-(Iticed. The homogenizing mledlitumii contained 0.3mole suicrose. 0.05 mlole potassiumi pllos_lphate btuffer.an(l 10-4 imole DTT (dith,iothreitol ) per liter, itspH /.2. The ratio of nml miediumni/g tis.lsue uise(d inhomiogenizing was 2:1. Normally ahout 50 g tissuiewas used in a preparationi. The homlogenate wvastheni strained through 4 layers of cheeseclothi toremiiove cell debris. The filtrate wa.s spun at 10oogto reni-ove starch grains audI the larger cell inclu-siolls. The sv\nthetase activity of this fraction wvasnegligille alid it w-as usuiall- (liscar(lc(l. The super-

natant was centrifiuged at 100.00q for 30 mmi toobtain maink mitochondrial material contaminatedwith underdeveloped chloroplasts and chloroplastfragmients. These fragimienits miainly settled at thetol) of the pellet ancd couild be remio\-ed plartially bydisturbing the stirface of the pellet carefully withi apipette. The remainder of the pellet W<as resuis-pen(led in the homogenizing imiediulim and s>pun at20,0010g. The washied pellet w-as resuspended ineither pIhosphate buffer- containing DTT to givebrokeni organelles, (lesiginate(l 10){))q/ "B" fractionin the text or in the homilogenlizinig milediulil, desig-nated 1 0))g A" fraction in the text. \h71en10,000!J '" fractiouis xN ere prepared. thll pellet wassusl)ed(led in 10 ml of ilmediumn wN-itholut sucrose for30 mini the material pelleted as ahove, aln(d thenresuispelided in 2 to 4 ml. \Vhein the 10),(X)Og A"fractions wNere prepared, they were (directly suispendedin 2 to 4 ml of the homllogenlizing mediuim containingsuicrose.

The supernaitant from the 100I(g)fJ pellet wvasspun at 1OO,OOOg oni a -Spi1nco Alodlel L Ultracentri-fuge for 60 min. Under these coniditions it wvasusual for the sedimented material to be in 2 )arts.One part formed a tight pellet oln the bottom of thetube. The other I)art formed a 'fluffy' lavyer justabox-e tlis. The liquidiI above the fluffy layer was

1638

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

MACEY AND STUMPF-LONG CHAIN FATTY ACID SYNTHESIS

pipetted off, and the fluffy layer was transferredinto a large volume of homogenizing medium andre-centrifuged at the sa;me speed. The other layer,which was considerably greener and whklch probablycontained some chloroplast and mitochondrial frag-ments, was sometimes retained where comparisonsbetween it and the fluffy layer were to be made.On a second spinl, Ithe fluffy layer sedimented nor-mally and this was the fraction designated as the100,0(Og pellet in the text. This pellet was bestresuspended using a gentle swirling action aided bya tube shaker set at low speed. After about 5 mina uniform suspenision could be obtained. Usiuallythe addition of 1 to 2 ml 0.05 M phosphate bufferpH 7.2 gave a suspension with a protein content of7 to 10 mg/0.2 ml. The supernatant from the100,000g spini is designated "supernatant" system inthe text.

Proteini Deterininations. Protein was determinedby the biuret reaction (7). BSA was used as astandard in protein determinations.

Inicubations. Cell fractions were incubated inthe following meditum except where stated other-wise-ATP, 2 jumoles; D;PNH, 0.5 ,umole; TPN+,0.2 iumole; glucose-6-P, 4.0 umoles; glucose-6-Pdehydrogenase, 0.5 units; GSH, 8.0 ,mmoles; E. coliACP 0.39 mg protein from crude extract of E. coli.\XThen malonate was the substrate (instead of malonylCoA), the medium also contained 0.2 Avmole CoAand 2 jumoles MnCL. When acetyl CoA was thesubstrate 30 jnmoles bicarbonate wvas included in themixture. The total volume after addition of 0.2 mlof supernatant enzyme or particulate preparationswas 1.07 nml. The amounit of substrate used in eachincubation was usually 140 m,u¢woles but in someinstances was as low as 70 mjumoles, the radioactivitybeing between 75,000 and 150,000 dpm per incubation.

Lipid extractions were carried out by the usualprocedures (10, 11). To anlalyze for fatty acids,0.1 ml of 60 % KOH was added to the reaction tubeto stop the reaction, and the tube was heated to 800for 30 min. The contents were then acidified andextracted wvith chloroform-methanol as describedabove.

Gas Chroinatography and Radiochemitical Anal-ysis. Gas chromatography was carried out on theAerograph Model A-9OP2 fitted with a thermalconductivity detector. Two columns were used-(i) 5' X 0.25" packed with 12 % diethylene glycolsuccinate (DEGS) on Anakrom P (60-70 mesh).This column wvas held at 160° where acids of chaiiilength tup to C1i were being ana,lyzed and wheregood seplaration of C,s:, and 'C,8l1 was desired. Thecoltm-n w-as uised at 1850 where the latter operationwas less critical and where it was desired to quan-titate the radioactivity in components up to a chainlength of C24J. (ii) 5' X 0.25" pa,eked with 17 %S.E. 30 on Chromasorb W (60-70 mesh). Thiscolumn wvas used isothermally at 2700 particularlyfor the detection of radioactivity in very long chainproducts uip to n-C.8, and also for temperature pro-

gramming of the products of degradation (seebelow).

The helium flow rate through both columns wasmaintain,ed at 60 ml. min.

Radiochemical analysis of the methyl esters offatty acids was accomplished by passing the effluentfrom the GLC through a Nuclear Chicago Biospan#4998 unit, at 250°. The signals from the detectorwere integrated using a scaler-recorder system de-scribed by Pearson et al. (10). Fatty acid methylesters were identified by usinig appropriate standardsusing the relative retention values for these com-ponents as listed by Buchfield and Storrs (3). Theidentity of C26 and C.8 components was tentativelyinferred by extrapolation of the semi-log plot ob-tained from the use of the standards and verified bydegradative studies (see below).

Methyl esters for degradative studies were ob-tained by collecting fractions from the effluent ofthe GLC in glass tubes containing glass wool soakedwith methanol.

Degradative Stutdies of Fatty Acids. Long chainfatty acids were broken down to a series of shorterchain length homologues using the technique ofHTarris and James (9).

GLC analysis ot fhe methvl esters of the degradedprodtucts was performed using temperature program-m.ing from 1200 to 2900 onl the S.F. 30 column.N.I.H. mixture F was used as standard up to a chainlength of C94. The 2 peaks beyond this whichploted isothermally at the positions of n-C,6 andn-C28 methyl esters gave the expected number ofextra peaks when degraded (2 anid 4 respectively),thus verifying their tentative identification as longchain fatty acids. In order to detect the presence ofelongation patterns, the ratio peak heighlt of the masstrace :peak height of radioactivity trace was used.Direct comparison between peaks at different pointsin the chromatogram was possible. Under the con-ditions of temperature programming used the stanid-ard deviations of successive peaks did not alter, andpeak area was proportional to peak height.Schmidt decarboxylations were carried out olnthe palmitate and stearate synthesized bv the super-natant system, u1sinig the method of Brady et al. (2).Deterwminationt of Radioactivity. Fatty acids,methyl esters, total lipid and CO, were determinedin toluene containing 0.6 % 2,5-phenyloxazole (PPO)and 0.5 % l,4-bis-2-(5-phenyloxazole L) -benzene(POPOP) witlh a Packard liquild scintillation spec-trometer. Aqueous samples of sulbstrate and ambientsolutionis were counted using Bray's solution. Count-ing efficiency was 72 % in toluene and 63 % inBray's solution.

Chem1icals anid Substrates. ATP, CoA, DPNH,TPN, GSH, and glticose-6-P were obtained fromSigma, and 1-14C labeled substrates from the NewEngland Nuclear Corporation. Malonate-2-14C wasobtained from Nuclear Chicago. Malonate-l-14Cfrom the same source was used to make malonylCoA-1-14C according to a method communicated by

1639

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

P. R. Vagelos, Washington University. Acyl car-rier protein was prepared from Escherichia coli bythe method of Goldnian et al. (6). All radioactivesubstrates were diluted to the specific activitiesindicated in the captions to the figures.

Results

Tissue Slices Stuidied. Lipid incorporation andCO2 evolution from acetate-1-_14C were measuredover an 8-day period from the onset of imbibition(fig 1). After 1 day, the fresh weights of the peasdid not alter appreciably anid the results are recordedon a fresih weight basis. Lipid incorporation wasat a maximum after 3 days soaking, as was CO2evolution. However, the lipid::C0 ratio was high-est after about 1 day's growth and decreased there-after. When 2 day old tissue was analyzed in termsof the fate of fed acetate, 35.' % of the acetate wasincorporated into lipid, 21.4 % was unchanged acetateabsorbed into the tissue, 8.8 % was non-volatile, withsoluble components, 13 % was released as 14CO., andthe remainder was associated with insoluble com-pounds presumably proteins and carbohydrates. Theratio of lipid soluble to water solufble products was4 to 1. Thus, incorporation into lipid is a majorpathAwNay for the metabolism of acetate in the lowlipid, germinating pea seed, and the ability to syn-thesize lipid from acetate is established early in thegerminiation within the cotyledons. Other results,not deta.iled, showed that other embryonic tissues,and shoot tissue also synthesized fatty acids of asimilar range and type but that the lipid:1CO ratioswere of the order of 1:10. The average percentcomposition of the synthesized products in terms ofchain length is reported in table I. Starting with

600.

500-

400cpmJo'

300-

200

100 Y2 3 4 5 6 7 8

daysgermination

FIG. 1. Incorporation of acetate-1-14C into lipid andrelease as '"CO, over 1 to 8 days by tissue slices frompea cotyledons of various ages. Incubation: 5 hrs 300,normal daylight. Substrate: 25O m,moles, 5 /Ac. 0,total lipid. A, 14CO.,.

Table I. Averagc Percent Distributioni of Radioactivityin Fatty Acids Synthesized by Tissue Slices

From 1-5 Days OldAnalysis by chainlength only usinig 5 foot x one-

fourth inch S. E. 30 column at 200 to 2700 (program-med). Incubation conditions desicribed in text.

Component %

C16 16.1Cl Is 34.0C20 3.8C22 18.9C,4 11.5C26 2.6C2R 13.1

300 m,umoles of acetate-1-14C. the one day tissueincorporated 6.4 % acetate into fatty acids, the 2 and3 day tissue 9.4 %. and the 4 day tissue 6.5 %. Theability to make all types of fatty acids did not varymuch over the period of 5 days, except that one daytissue had no capacity to synthesize the C,6: comii-ponent, but still synthesized the normal quantities ofC,g:l typical of other tissue samples. This suggeststhat the C16,: production 'is not directly tied toC181l. An aliquot of the methyl esters from the 3day tissue were analyzed by thin layer chromatog-raphy on silver nitrate impregnated Silicagel G.The content of labeled monounsaturates was esti-mated at 30.1 %, and that from GLC at 27.2 %.

Incorporation of acetate over a period of 7 daysremained essentially similar with respect to the lipidclasses involved, but as the tissue aged there was an

increasingly large incorporation into free acid, whichvaried from 6 % for 1 day old to 25 % for 7 day oldtissue.

Over 95 % of the label in total lipid was re-covered as methyl esters of fatty acids. There wereno components corresponding to hydroxy-acids syn-thesized from acetate in this tissue.

Since the tissue had a pronounced capacity forthe synthesis of very long chain fatty acids, fattyacids of various chain lengths were used in anattempt to detect elongation of such substrates.Table II shows the results of feeding C8 18 radio-active fatty acid substrates in the presence of coldacetate. It can be seen that even though they werereadily broken down to CO2, neither C1, nor C1,fatty acids gave any detectably labeled higher homo-logues, and no substrates except acetate and malonatewere incorporated into very long chain fatty acids.Fattv acids of intermediate chain length, (C, 14)gave varying amounts of products with a chainlengtih up to C18, but higher homologues were notdetected. Such precursors were also reasonablyefficient precursors of unsaturated fatty acids (C]6 1and C,18:1). These results are similar to those ob-tained by Hawke and Stumpf (11). If it is assumedthat the breakdcwn of palmitate and stearate is anormal ,-oxidation process, it is apparent thatbreakdown to acetate followed by re-svnthesis did

1640

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

MACEY AND STU.MPF-LONG CHAIN FATTY ACID SYNTHESIS

Table II. .Metabolic Activity of Several Fatty Acids as SubstratesTwo g tissue slices and 300 nipmol.es of each substrate (2 yc) were used. In the case of fatty acid substrates

250 mjumoles of cold acetate were included. Only fatty acids up to C22 analyzed. Substrates were all 1-labeledexcept malonate which was 2-labeled.

Products (mnignoles)Substrates CO, C8 C10 C12 C14 C16 C16:1 Cl8:0 C18.1 C20 C22C18.0 8.07 ... ... ... ... ... 12.90C16:0 10.03 ... ... ... ... 28.30 ... ...C14 8.04 ... ... ... 8.50 6.04 ... 1.56 1.56C12 26.20 ... ... 0.14 0.14 0.28 0.28 0.09 1.24 ...C10 17.20 ... 8.55 ... ... 2.80 ... ... 0.85 ... ...

C8 12.00 ... ... ... ... 0.35 3.88 0.58 0.97 ... ...Acetate-1-1*C 24.20 1.20 1.80 0.30 2.70 1.20 1.42Malonate-2-14C 22.20 1.35 0.30 ... 1.38 1.50 1.70

not occur. Since this would also be the case for theother 1-labeled sulbstrates the variety of fatty acidssynthesized from these substrates are probably elon-gation products. Since the C8-C,4 acids were elon-gated, they must have been converted to the thioesterderivative appropriate for elongation (whether it bea CoA or ACP derivative is not specified). How-ever, the process apparently went no further than achain length of C18 under present experimental con-ditions. Thin layer chromatography of total lipidfollowing incubation with palmitate or stearateshowed that most of the substrate was esterified inpolar complex lipids or in neutral lipid.

The saturated fatty acids made by tissue slicesfrom acetate were separated by GLC and degradedusing the technique modified from that of Harrisand James (9)., Using this technique, a series offatty acids differing in chain length by 1 carbonatom were produced from the original acid.

Radioactivity was present in every componentdown to the level of C1 0 from the w- end of the

Table III. Radioactivity:ntoass, Ratios for Tissue-Synthesized Fatty Acids Fromn Acetate-1-14CRatios were calculated using the ratio of the peak

heights on mass trace and radioactivity trace. ..." indi-cates not sufficient mnas!s to measure.

KMnO4Oxidationproducts C1 6

C28

C26C24

C22C20

C18C16C14C12CIO

C8

0.890.820.740.66 1

Parent comilpounId

18 20 C2 2C24 C28

20.000.55

5.9 0.291.221 1.6 0.37

15.52 1.06 3.6 0.620.45 2.40 0.69 1.7 1.850.50 0.79 0.56 0.6 1.110.85 0.50 0.72 0.9 1.000.75 0.50 0.34 ... 0.510.69 0.93 0.24 ... 0.28... ... ... ... 0.50

1 The figures for C22 were obtained after adding 4 mgof cold C22 to the isolated methyl ester before saponi-fication. Other figures were obtained -without additionof carrier.

chain. Neither miiass nor radioactivity were detectedunder our conditions in methyl esters of a chainlength less than C8, and a ratio for the latter com-ponent was nlot arrived at in most cases owing tothe small amounts of mass present, though in allcases some activity was discerned in this component.The figures for the even numibered degradationproducts are shown in table 1II. The ratios indicatede novo synthesis for both palmitate and stearate.In other components there was evidence of elonga-tion superimposed on a pattern typical of de novosynthesis. Although the magnitude of the effectvaried somewhat, all acids synthesized which pos-sessed a chain length greater than C18 had consider-ably greater relative amounts of radioactivity thanany of their degraded products.

Decarboxylation studies performed on a series ofsaturated fatty acids gave amine :C'O2 ratios between7.0 and 9.0 for palmitate and stearate, and thusconfirmed the conclusion that these products areformed de miovo from acetate. Results for the longerchain products were somewhat variable, and reliancehas been placed more on the data from degradationstudies.

Homogenate Studies. Homogenates of pea coty-ledon tissue wvere tested for synthetase activity. Thesynthetase capacity of the holmiogenates and super-natants was linear over a time period of 4 hr. WVhenmeasured over 3 hr, the relation between proteinconcentration and lipid incorporation was linearwith use of up to 16 mg protein per tube, as shownin figure 2. The rates of incorporation bv thehomogenate were closely similar to those of thesuipernatant. The same general relations were trueof the particulate fractions of the cell. U;p to 16.0mg/tube of "mitochondrial" protein and 15.0 mg/tubeof "microsomal" protein were used and gave linearrates of incorporation into lipid over a period of 3hr in each case. Using a time period of 3 hr,the incorporationi into lipid by particulate fractionswas linear to about 16 mg protein:tube. Malonatewas efficiently incorporated into lipid at about halfthis rate. There was a trend towards reduction ofsynthetase activ7ity as germination time progressed,but no peak of activity comparable with that observedusing whole tissue slices.

1641

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

w 0moleEr verPor':Ied

es h

5-

2hours

FIG. 2a. Incorporation of malcby homogenate and supernatant systime. Substrate: malonate-2-14C,dpm. Protein: Supernatant, 7.7 nmg. *, Supernatant. A, Homog

boxylase activity than the supernatant, but thatmalonvl CoA synthetase activity was much lower.The l0,OOOg pellet shoNed a different pattern, whiclhwas dependent on the pretreatment of the pellet. Iftreated in a manner which would be expected to

A ' keep the mitochondria intact, acetyl CoA was themost efficient precursor of lipid, and malonyl CoAwas relatively less effective. When treated with ahypotonic solution for 30 mim before incubation withsubstrate and co-factors, malonyl CoA was the onlysubstrate that was effectively used. This indicatedthat the hypotonic treatment had released and dilutedenzyme(s) and/or co-factors that were responsible

3 4 5 in part for the conversion of acetyl CoA to malonyl4 it l

CoA. Results also imply a limited movement of)nate-2-'aC into lipid malonv l CoA into the mitochondrial organelle in

140 aumnoles, 140,000 System A. Treatment wNith hypotonic medium al-

ng. Homogenate, 7.9 lowed ready entrance of malonyl CoA to the site ofrenate. synthetase activity.

The composition of the fatty acids formed byhomogenates from malonyl CoA and malonate was-C140, 2 %; C160, 25 %; C18.0, 70 %. There wereonly traces of very long chain acids. No unsaturatedfatty acids were made by homogenates from malonateor malonyl CoA. Clearly this pattern of incorpora-tion bore little resemblance to that typical of thewhole tissue (table I). Using various fractions andsubstrates, the results in table IV were obtained.The supernatant incorporated malonate and malonylCoA, but utilized acetate and acetyl CoA poorly.The 100,00Og pellet was as efficient as the super-natant using malonyl CoA. Incorporation of acetvlCoA was relatively greater than by the supernatant,whilst malonate was very poorly used. This indi-cated that the pellet was richer in acetvl CoA car-

40!

30mfp molesincorpor-oted 20

10l

0

3.2 6.4 9.6mg protein

FIG. 2b. Relation of protein concentiincorporation rate nmeasured over a 3 hrcrude homogenate and supernatant systemmalonate-2-14C, 140 mgmoles, 140,000 dpnnataint. *, Homogenate.

Table IV. Incorporation of Different Susbstrates IntoLiPid by Thrce Cell Fractions

Fractions were prepared from 2-day -pea cotyledons.Amounts of protein used: Supernatant, 6.4 mg; l0,OUOg"A" 16.0 mg; l0,OOOg 'B", 16.0 mg; 105g, 3.5 mg. Seetext for definition of "A" and "B". Substrates: acetate-1-14C 70 m,zmoles, 200,000 dpm; acetyl CoA-1-1,C 70mAumoles, 220,000 dpm; malonate-2-14C 70 mpmoles,70,000 dpm; malonyl. CoA-1-14C 70 m,kmoles, 75,000 dpm.

Supernatant 10,000g "A" 10,000g "B" 105gitiiiiioles incorporated per 10 mg

protein per hrAcetate 0.25 0.12 ... 0.35Acetyl CoA 0.24 0.71 <001 1.30Malonate 1.12 0.04 ... 0.10Malonvl CoA 2.70 0.37 3.00 3./7

In all these experimiients, use was made of Unl-labeled acetyl CoA in combination with malonyl

* CoA in an attempt to stimulate fatty acid synthesisby providing the primer acetyl CoA, but in practicethis addition made no observable increase in fattv

A acid synthesis possibly because there was enoughdecarboxylase activity in the tissue to provide

0 satturating amount of acetyl CoA.Further distinction between the 2 systems was

afforded by studies of their ppH dependence (fig 3).The sutpernatant system showed a sharp optimum atpH 7.0 whereas both pellets showed a much broadertolerance of pH variations, with an optimum for thelOO,OOOg pellet at 8.0 and for the 10,000g pellet at 7.0.

Patterns of fatty acid synthesis differed markedlybetween the 3 fractions used and upon whether the

12.....8 .0.. substrate used was malonyl CoA or acetyl CoA.12.8 16.0 Table V shows the distribution of radioactivity invarious fatty acids with a chain length ranging from

r period usingCd to C,6 as a restult of incubation of acetyl CoAs. Substrate: ancl malonyl CoA with the various cell fractions.

l. *, Super- It is apparent that the 1Og fraction, using malonylCoA, has the ability to synthesize the verv lonig

1642

1-

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

MIACEY AND STUMPF-LONG CHAIN FATTY ACID SYNTHESIS

chain fattyfractions. 17 % of C.,,.ditions fortemperature:ponents couthesis bv thi

3500k

3000 .

20001

cpm/aliquot

1000o

6.C

F[G. 3. p1tenis associatThe pH wascentration, an1 mil more thCoA, 70 niar3 hr, 30°. Thtype (10,000g10,000 pellet.

acids which is absent from the other size the very long chain fatty acids was not a generalFhe 10,0O0g fraction synthesized some property of particulate cellular components. Thebut even under the most favorable con- supernatant pattern was very similar to that foundtheir detection, namely the use of high from homogenates. It is apparent that the synthetics and an S.E. 30 column, the C22 26 com- capacity of the homogenates are largely dominatedIld not be detected as products of syn- by the supernatant. The differences between theis fraction. Thus, the ability to synthe- products made from malonyl CoA and acetyl CoA

were marked. With the latter substrate, neitherparticulate fraction synthesized fatty acids with achain length greater than C16. With acetyl CoA,50 % of the acids synthesized by the 10,000g fractionwas a C16:1 component; In contrast, with malonylCoA the main component was stearate, with somesmall amount of C20. A considerable shift in patternwas also observed as a result of using these differentsubstrates with the supernatant system, but the effectwas not as clear-cut. An unknown component ap-peared in the radiochromatograms from the 10(Pgpellet, having a retention time corresponding to acarbon number of 19.5. This component was ap-parently not a hydroxy-acid, as it did not separateas such on TLC, nor was it an unsaturated com-ponent since it remained in the same relative positionof gas-liquid chromatograms using both DRGS andS.E. 30 columns. It seemed probable that it couldbe an elongation product of an endogenous branchedchain.

Since the 105g fraction sedinmented in 2 distinctparts, the ability of both parts to incorporate malonylCoA into lipid was tested separately. It was foundthat both fluffy and heavy fractions incorporatedmalonyl CoA with about the same pattern of fatty

' ' ' acid production, but that the fluffy (lighter) fraction70 8.0 8.4 wvas considerably more efficient. Four determina-pH tions on each fraction yielded the following figures

H dependence of fatty acid synthestase sys- (in m,umoles incorporated per 10 mig pro-ein pered with 3 cellular fractions, as indicated 3 hr) ; fluffy, 10.82 ± 0.7; heavy, 7.44 +- 0.6.varied using HCI buffer, 0.05 AI final con- Experiments designed to detect elongationi fromird a total volume of incubations of 2.07 ml, palmitate-1_14C and stearate-1-14C and from palmityltan the normal volume. Substrate: malonyl CoA-1-14C utsing the lO5g pellet have yielded nega-noles, 75,000 dpm. Incubation conditions: results. the using ACP dedinesie 10,0O0g fraction used was of the "broken" tive results Experiments using A'CP derivativest). A, Supernatant. *, 105 pellet. V, should be done.

Experiments designed to detect interaction be-Table V. The Percent Distribuitiotn of Radioactivity in Fatty Acids After Inculbation of Acctyl CoA and Mfalonyl

CoA With Different Cell FractionsResults for the lOOOg pellet using acetvl CoA were obtained using pellets kept in 0.3 M sucrose during prepara-

tion and inlcubation, those for malonyl CoA ivere obtained using a pellet previuousl) dispersed in di'ute phosphate buffer.Material wa- bhulked over several experiments.

105g,nal CoA ac CoA

2

34555

10,00Ogmal CoA

325

65

1712203511

Supernatantac CoA mal CoA ac CoA

2 2 250 20 45484.

tr

75 53

3

C14:0C1 6:0

C18:0C1 :1Unl;noawnC.2OC.,C2C26

1643

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

Table VI. Possilble Interaction in Total Lipid Incor-poa-tiotn Between Supernatant, JMicrosomes, and

AIlitocizondriaThe substrate w-as 140 mcmoles

taimiing 150,000 dpm. Incorporation-- Supernatant, 7.3 mig protein. mteini. M = 10,000g, 99 mg proteiin.

Combination

S m MS mS M

m M.

S

III

M

malonyl CoA conl-time was 3 lir. S

Q5'g, 1.5 mg pro-

Incorporation/3 hr

14.919.319.47.3

10.51.64.6

sup)ernatant system. Since capacities for syinthesisof the 2 cell fractions are approximately equal onl aprotein basis, one would expect that in this case theincorporation of the miiixed fractions would be (lomiii-nated by that typical of the supernatanit system. bultthis was not so. The increase in total lipid svntlhe-sized by the combination of stipernatalnt pltls par-ticulate fractions couild be related to the insertion offatty acids to polar lipids in mlicrosomes.

Cofactor Rcqulirmeme1ts for- Cell Free Fractionis.The responise to various cofactors o(f 3 differentsystems usinlg 2 suibstrates is shovni in table VIII.

In this experimiient since a 10O)Ogl)pellet whiclh hadbeen treated wvith a hypotonlic solution was, ulsed,incorporation fromii acetyl Co.\ Nvas rather lowv. The

twveei1products of the supernatant -,vstemi anid the 2particulate fractions hlave vielded negative results.This experimen,t has been attempted in 2 waVs 1 ) byincubating supernatant and particulate cell materialtogether and separately aand testing the results forinteractions as they may affect the patterns of synI-thesis, 2) by incubating thle suipernatanit systemii withsubstrate to form the C,11 and C,1, components, andthen adding particulate preparations to the boiledsystem. In neither case has it vet been possible todetect interactions between the systems with respectto influencing the pattern of fatty acids synthesized.Interaction was however observed, as might be ex-pected, in the total lipid figures. Table VI showsthe result of recombining cellular fractioins oIn thetotal lipid synthesis from malonyl CoA. For in-stanlce, separately the 100,000g pellet and the super-natant system incorporated 1.6 and 10.5 mIlmoles ofmalonyl CoA but when combined the total incorpora-tion was 19.3 mjumoles, suggesting a synergisticinteraction occurring between the 2 fractions. TableVII shows the distribution of radioactivity betveenthe broad lipid classes of glycerides, free acid andpolar complex lipid as influenced by tlis interactioni.The supernatant systemii aione accuimultlated 58 % ofthe label in free acid, while when combined with the105g fraction most of the label was transferred fromthe free acid into the polar lipid fraction. Endog-enous synthesis in the 105g fraction cannot quantita-tivelv account for this effect, because in this experi-ment the protein associated with the 10Ug pellet wN-askept low at 1.5 mg compared with 7.3 nmg fromii the

Tabe VII. Lipid Distribution Patterns in Lipid Syn-thlcsizved by the Supernatant System and Suipcrnatant

+ 105g PelletProtein values anid other conditionis as for table VII.

Polar IlipidFree acidGIx cerides

Supernatant

255817

Supernatant+ 105g pellet

701119

Table VIII. 7T1e Effect of Limitfing Cofactors on 3Svn thltase Incorporating SNvstems

Substrates used: maloinyl CoA, 140 mOmoles 150,000dpmii, acetvl CoA, 140 m,umoles, 125,000 dpm. Protein:superiuatant, 10 7 nmg, 100,000g, 7 mg, 10,000g, 18.0 mg.

100,000:/ Sutperinataint l0,000g (B)n,ui,i)iO/eS sutbst rate incorporated per

10 inm pr-otein per hrMalonvI CoA a. cubstrateACP 4.90 0.83 0.86

-ATP 6.75 6.87 1.93-DPTH 8.30 6.17 2.70-TPNH 1.89 2.58 1.27--MgCl. 7.10 6.67 3.08-Bicarbonate 7 85 8.35 2.90Comiplete 8.10 8.14 3.13Acetvl CoA -a. .sul)trate-ACP 1.36 0.15-ATP 0.48 0.29 1

-DPNH 0.78 0.75 <0.01-TPNH 0.30 0.50

--MIgCl., )0.86 0 85-Bicarboinate 0.41 0.62Complete 1.72 0.86

ACP anl -ATP signiificanitlx depressed incor-px)rat ion.

resuilts obtained ii1dicatte that the sul)ernatant aindlO5g pellet w-ere more efficient oni a protein basis

thani the 1O,0O0g pellet. All systemis respoindedmarkedlv- to limitation of TPN'H. DlPNH did niothave the sanme effect, thotuglh its absence in tileincubation mediumii had sufficienit effect in the super-natant sx stem for it to be routinely included. Theadditioni of ACP piroduced a 10-fold stimulatioin of

lipid inicorporationi by the supern<atalit system, a

2-foldl one for the 100,00,g pellet aind a 3- to 4-foldone for the 10,000)g pellet. These results indicatethat the particulate system.s are ACP dependent butin valrying degrees I)resumablyl because of tile pres-enice of bouniid functional ACP. Usinlg acet)yl CoAtile differenices were moore obscured by the generallylow levels of inicorporation encounitered, blut thegeneral trenids remained similar. The expected re-

sponse to bicarboinate occurred using the supernatant

1644

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

MACEY AND STUM,IPF LONG CHAIN FATTY ACID SYNThIESIS

and 100,000g pellet. However, due to the low levelsof incorporation, no results were obtained usingacetyl CoA with the 10,000g (B) pellet.

While not yet studied in any detail, ,B-oxidationsystems are prominent in pea tissue homogenates,and are associated with both solu(ble and particulatematter after homogenizing.

Attemipts have been made to demonstrate pro-duction of C,8 l in cell free systems derived frompea tissue slices by incubations under a variety ofconditions, but so far without success. 'Chemicaldegradations on products of cell-free systems havebeen performed thus far only on the C16 and C18components made by the supernatant system, whichwere easily obtainable in the required large quanti-ties. Schmidt decarboxylations of these productsgave CO2 :amine ratios of 7.0 for C16 and 7.7 for theC18 component. These figures indicated predomi-nantly de novo synthesis. Degradation of C18 usingthe permanganate method revealed consistent radio-activity :mass ratios down to a level of C12, afterwhich the ratios were much reduiced.

Discussion

The synthesis of a large proportion (up to 80 %)of stearate by the cell free superinatant system andhomogenates from pea cotvledons is atypical of thetissue slice system which makes mostly Cl,1l fromacetate-1"C together with a smaller or equal amountof C1,., and also a series of very long chain fattyacids up to a chain length of C28. Both C16 0 andC18:0 were apparently synthesized de novo by tivsueslices and by the cell free system. Though it seemsprobably that C1,.1 synthesis proceeds by way of anoxidative desaturation of stearate, there is at themoment no direct evidence in vitro of this reaction.The above results do have relevance, however, tothe mode of synthesis of the very long chain fattyacids with a chain length up to n-C28.

The very long chain fatty acids synthesized fromacetate by tissue slices showed evidence of elongationhaving occurred because the pattern of distributionof radioactivity in them and the products of theirdegradation showed a larger radioactivity:mass ratioin the parent compound than in the degraded prod-ucts. In cases where no unlabeled diluent was addedto the fatty acid before degradation, the differencein the ratio approached an order of magnitude. Sucha result (table III) can be explained only if thema-in reaction occurring in the experimental periodwas the addition of 1 acetate unit to a pre-existingfatty acid Cn to form Cn12. The results do not pointto any one fatty acid as being the substrate forelongation reactions, but indicate that all endogenousfatty acids with a chain length >C,16 can performthis role.

The capacity for very long chaini fatty acid syn-thesis was confined to the 105g pellet and was notassociated with the 10,000g pellet. The exact nature

of the pellet as isolated has not yet been determined.Since the particulate fraction w-as the only locus ofproduction of fatty acids with a chaini length >C,8,which have been shown to be formed partially bvelongation reactions 'in tissue slices, it is likely,though not proved, that the 105g pellet is the site ofelongation of fatty acids of chainl lengthi C,8 andabove. Guchhait et al. (8) reported that pigeonliver microsomes were capable of elongation reac-tions, the mnain evidence being a stimulation ofmalonyl CoA incorporation by the addition of acylCoA derivatives of requisite chain length. In thepresent study, attempts to incorporate palmitylCoA-1-11C into the long chain fatty acids using thelO1g pellet failed, all the label remaining in thepalmitate when the latter was incubated under theusual conditions with cold malonyl CoA. Similarnegative results were obtained when the free fattvacids were used. Information relevant here is thedistribution of radioactivity among the products ofdegradation of the very long chain fatty acids syn-thesized by the cell free system, but this has not yetbeen obtaiined. It seems likely, however, that en-dogenous fatty acids within the 105g pellet form thesubstrate for elongation reactions involving malonylCoA. at least for the synthesis of the very long chainproducts. The 105g pellet synthesizes only a smallamount of C,,; no stearate is present in the synthe-sized products (table \T). Apart from the unknowncomponent, which is probably a product of elonga-tion of a branched chain, well over 90 % of theradioactive products of this system are long chainfatty acids C20 26. The presence of radioactivity inthe products of degradation of the tissue synthesizedacids down to a chain length of as low as C, indi-cated that during the experimental incubation periodof 5 hr, some of the very long chain acids weresynthesized de novo from acetate, and some wereproduced from elongation reactions, so that the finalmixture of acids extracted from the tissue wasderived from a mixed population with respect to itsrelation to the recently metabolized acetate. Whatseems therefore likely is that there is a de novosynthesis of acetate up to a chaiin length of C18,possibly by the soluble system, followed by a gradualaccumulation of longer chain fatty acids by a slowaddition of C2 units which may occur only aftertransfer of C18 to an elongation "complex" in themembrane.

The additions of acetate did not apparently occurcontinuously at any 1 locus as is the case for othersynthetase complexes, since if this were so, radio-activity :mass ratios would be similar down to acertain chain length, when there would then be anabrupt transition to the level of the primarily elon-gated substrate.

The above concept of the elongation reactions inpea tissue does explain some otherwise puzzlingresults. One of these, true in both pea cotyledontissue and barley leaf tissue, is that the actual massof the very long chain acid components is about 2

1645

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

orders of magnitude less than the mass of theC,1 18 components from which it is assumed theyare derived. In order to explain this, it is necessarvto invoke a compartmentation hypothesis. If onlysmall amounts of long chain acids are formed duringthe growth of a cell, and if labeled fatty acids wereformed exclusively by elongation reactions of thesesmall quantities, the relatively high specific activityof the very long chain acids would be explained.Secondly, it is difficult to explain why palmitate andstearate are not elongated to longer chain componentswhen supplied to the tissue with adequate suppliesof cold acetate, especiallv since activation at least tothe level of CoA ester must occur, as about 10 % ofthe acid taken up by the tissue is oxidized to CO2.It is, however, quite possible that the ACP deriva-tives of palmitate or stearate are the active sub-strates. There is now evidence accumulating thathigher plalnts do not have long chain acyl trans-acylases wlliclh would catalyze the transfer of thestearvi component of stearvl CoA to ACP. Thus,stearyl CoA could be rapidly formed but wouldeither be ,-oxidized or transferred to a complexlipid and remain out of the pool of acyl ACP neces-sary for elongation.

Breakdown of palmitate and stearate-1-14C toacetate units presumably occurred since about 10 %of the label from these substrates appeared in CO2.For sonme reason, however, resynthesis did not occurunder these conditions either in the present case ofpea tissue or barley leaf tissue (11) ; otherwisesy-nthesis of the very long chain fatty acids shouldhave been observed from the 1-labeled sutbstrates.Using uniformly labeled C,, as the substrate, Kolat-tukudy (12) demonstrated the production of somelong chain fatty acids up to a chain length of C96,.The origin of these was said to be bv direct elonga-tiotn from stearate. However, the very efficientconversion of palnmitate and stearate to nonacosane(12) whicl he reports, together with other evidencerecently stulmmarized (13) suggests that comparativestudies of different tissues mav reveal some inter-esting differences in elongation patterns.

A feature of some importance possessed bv the105g pellet systemn in peas is the synthesis anid re-lease of fatty acid homologues differing by only 2carbon atoms. It should be pointed out that thisability to svnthesize large quantities of stubstanceswhichl are apparent intermediates in tlle pathwav tothe final product is not typical of fatty acid syntlhe-tase systems. The supernatant system in peas, forexample, synthesizes oinly C16 and C,1 fatty acids,and the intermediates are generally not detectable.The same is true of most other systems whiclh lhtavebeen investigated. The pattern of radioactivity inthe degradation products indicating elongation ofendogenous acids, explains this feature of the svstemi.

Lipid synthesis in germinating peas, as in otherseeds, is probably important becauise of the necessityto synthesize cell membranes during this period. Itis of some interest that total lipid incorporation from

acetate increases rapidlv from the onset of imnbibition,while CO. output lags someiwhat behind l(fig 1).The role of the very long chain fatty acids in thegerminating seed is not clear, but they appear to beassociated equally with glycerides and polar lipids.No incorporation of radioactivity into less polarcompounds, such as long chain esters or paraffinscould be detected. It is possible that such very longchain acids are an essential component in cell mem-brane systems as well as in waxy coatings, and arepossibly converted to wax components in the leavesand stems of higher plants.

Acknowledgment

The authors are grateful to Dr. S. Matsumrura foradvice and assistance in preparation of E. coli ACP andto Mrs. Barbara Weigt for general assistance.

Literature Cited

1. ABRAHAM, S., K. J. MATTHES, AND I. L. CHAIKOFF.1961. Factors involved in the synthesis of fattyacids from acetate by a soluble fraction. Biochinm.Biophys. Acta 49: 268-85.

2. BRADY, R. O., R. M. BRADLEY, AND E. G. TRAMS.1960. Biosynthesis of fatty acids. I. S!tudieswith enzymes from liver. J. Biol. Chem. 235: 3093-98.

3. BURCHFIFELD, H. P. AND E. E. STORRS. 1962. Bio-chemical Applications of Gas Chromnatography.Academic Press, New York.

4. FREEMAN, C. P. AND D. WEST. 1966. Comp!eteseparation of lipid classes on a single thin-layerplate. J. Lipid Res. 7: 324-27.

5. GLEW, R. 1968. Ph.D. Thesis, University of Cali-fornia.

6. GOLDMAN, P., A. WV. ALBERT, AND P. R. VAGELOS.1963. The Condensation Reaction of Fatty AcidBiosynthesis. II. Requirements of the enzymesof the condensation reaction for fatty acid syn-thesis. J. Biol. Chem. 238: 1255-61.

7. GORNALL, A. G., C. J. BARDAWILL, AND M. M. DAVID.1949. Determination of serum proteins by meansof the biuret reaction. J. Biol. Chem. 177: 751-66.

8. GeCH HAIT, R. B, G. R PUTZ, AND J. W. POOrER.1966. Synthesis of long chain fatty acids by mi-crosomes of pigeon liver. Arch. Biochem. Biophys.117: 541-49.

9. HARRIS, R. V. AND A. T. JAMES. 1965. The fattyacid metabolism of Chlorella vulgaris. Biochim.Biophys. Acta 106: 465-73.

10. HAWKE, J. C. AND P. K. STUMPF. 1965. Fatmetabolism in higher plants. XXVII. Synthesisof long chain fatty acids by preparations of Hor-dcun vulgare L. and other Graminae. Plant Phy-siol. 40: 1023-32.

1 1. HAWKE, J. C. AND P. K. STUMPF. 1965. Fatmetabolism in higher plants. XXVIII. The bio-synthesis of saturated and unsaturated fatty acidsby preparations from barley seedlings. J. Biol.Chem. 240: 4746-52.

12. KOLATrTUKUDY, P. E. 1966. Biosy-nthesis of Xwaxin Brassica oleracea: Relation of fatty acids towax. Biochemistry 5: 2265-75.

1646

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.

MACEY AND STUMPF-LONG CHAIN FATTY ACID SYNTHElSIS

13. KOLATTUKUDY, P. E. 19;8. Bic synthesis of sur-

face lipids. Science 159: 498-505.14. LORCH, E., S. ABRAHANI, AND I. L. CHAIKOFF.

1963. Fatty acid synthesis by complex systems:The possibility of regulation by microsomes. Bio-chim. Biophys. Acta 70: 627-38.

15. MAJERUS, P. W., A. W. ALBERTS, AND P. R. Va-GELOS. 1965. Acyl carrier protein. VII. Theprimary structure of the substrate binding site.J. Biol. Chem. 240: 4723-26.

16. MANGOLD, H. K. 1961. Thin layer chromatographyof lipids. J. Am. Oil Chemists' Soc. 38: 708-27.

17. MORRIs, L. J. 1962. Separation of higher fattyaci(L isomers and vinylogues by thini layer chro-matography. Chem. Ind.: 1238-40.

18. NAGAI, J. AND K. BLOCH. 1966. Enzymatic desatu-ration of stearyl-acyl carrier proteini. J. Biol.Chem. 241: 1925-27.

19. PEARSON, R. JR., W. C. FINK, AND A. A. GORDUS.1965. Radiogaschromatography Peak Analyser. J.Gas Chromatog. 3: 381-83.

20. YANG, S. F. AND P. K. STUMPF. 1965. Fat me-

tabolism in plants. XXI. Biosynthesis of fattyacids by avocado imesocarp enzyme systems. Bio-chim. Biophys. Acta 98: 19-26.

1647

www.plantphysiol.orgon July 3, 2018 - Published by Downloaded from Copyright © 1968 American Society of Plant Biologists. All rights reserved.