Plant Physiol. (1980) 65, 1067-10720032-0889/80/65/1067/06/$00.50/0

Ethylene Action and Loss of Membrane Integrity during PetalSenescence in Tradescantia'

Received for publication July 2, 1979 and in revised form November 27, 1979

JEFFREY C. SUTTLE' AND HANS KENDEMSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, Michigan 48824

ABSTRACT

Senescence of isolated petals of Tradescantia is accompanied by a largeincrease in membrane permeability, and application of ethylene hastensthe onset of this increase. There is a 1- to 2.5-hour lag between ethyleneapplication and the onset of anthocyanin efflux (an indicator of increasedmembrane permeability). Simultaneous application of 0.1 miimolar cor-dycepin or cycloheximide with ethylene abolishes the response to ethylene.Analysis of phospholipid levels in these petals during senescence has shownthat the increase in membrane permeability is accompanied by a massiveloss of phospholipids. Factors which enhance or retard the rate of antho-cyanin efflux exert a corresponding effect on the rate of phospholipid loss.The composition of the phospholipid fraction remains unchanged duringsenescence. The activity of phospholipase D declines during senescencewhereas that of acyl hydrolase remains essentialiy constant.

Increased membrane permeability is a characteristic attribute ofsenescing plant tissues (2, 6, 11, 14, 15). Desiccation of leaves,wilting of petals, and enhanced efflux of cellular constituents suchas vacuolar pigments, sugars, and electrolytes are all gross mani-festations ofmore subtle changes in membrane integrity that occurduring senescence. Because many aspects of membrane permea-bility are associated with the composition and organization of thelipid components of membranes (12), considerable attention hasbeen focused on changes in membrane lipids (primarily phospho-lipids) during senescence. The loss of membrane integrity thatoccurs during leaf senescence is correlated with a large decline inthe phospholipid content (2). However, because of the long inter-vals between the determinations, it is difficult to discern whetherchanges in phospholipid content precede the increase in cellularpermeability, or vice versa.Ephemeral flowers offer a unique opportunity to study the

temporal sequence of biochemical changes attending senescence.Senescence of isolated segments of Ipomoea flowers is accompa-nied by an abrupt increase in membrane permeability (6) and alarge decline in phospholipid content (1). However, because phos-pholipid content has not been directly correlated to the onset ofpermeability changes within this tissue, it is again difficult toassess the role of phospholipid loss in the observed increases inmembrane permeability.During senescence, anthocyanin is released from isolated petals

of Tradescantia, and pretreatment of the petals with 10 ,ul/l of

'This research was supported by the United States Department ofEnergy under Contract EY-76-C-02-1338.

2 Present address: United States Department of Agriculture Metabolismand Radiation Research Laboratory, State University Station, Fargo,North Dakota 58105.

ethylene hastens the onset of this process (14). Because anthocy-anins are localized in the vacuole, these petals offer a uniqueopportunity to assess directly the integrity of cellular membranes,in particular that of the tonoplast, as well as the role of ethylenein effecting membrane degradation. Here we further describe theresponse of mature petals of Tradescantia to ethylene and thenature of the biochemical changes which result in the loss ofmembrane integrity during senescence.

MATERIALS AND METHODS

Plant Materials and Ethylene Analysis. Cloned plants of ahybrid Tradescantia (clone 02) were grown in an environmentalchamber as previously described (14). Petals were isolated fromfully opened flowers which were excised from the plant early onthe morning of flower opening. Prior to use, the petals were storedin a glass Petri dish containing a disc of water-saturated filterpaper to prevent desiccation. Ethylene production was determinedin 1-ml gas samples by GC (14).

Simultaneous Determination of Anthocyanin and ElectrolyteLeakage. Two groups of 18 petals each were floated on glass-distilled H20 for 1 h. The petals were then placed on 8 ml of glass-distilled deionized H20 in 50-ml Erlenmeyer flasks. The flaskswere sealed with serum vial caps, and ethylene was added to oneflask to yield a final concentration of 10 tLl/l in the head space. Atthe appropriate times, the flasks were opened, and both theconductance and A at 575 nm ofan aliquot of the bathing mediumwas determined as described (14). Following these determinations,the aliquot of bathing medium was returned to the flask and theflask resealed. Following resealing, ethylene was again added tothe respective flask to a final concentration of 10 ,A/l.

Effect of Cycloheximide and Cordycepin on Ethylene-inducedEfflux. Groups of 15 petals were floated on 5 ml glass-distilledH20 or 0.1 mm cycloheximide (Sigma) in 50-ml Erlenmeyer flasksequipped with side arm cuvettes as described (14). The flasks weresealed, and ethylene was introduced to a final concentration of 10,ul/l. In one flask, only 4.5 ml distilled H20 were initially included,and after 2 h exposure to ethylene, 0.5 ml of 1 mm cycloheximidewas injected through the serum vial cap. Analogous procedureswere used in the study of the effect of cordycepin (Sigma) and theother inhibitors of transcription.

Determination of Total Phospholipid Levels. Groups of 12petals were floated on 3 ml distilled H20 (or the appropriateinhibitor solution) in 50-ml Erlenmeyer flasks. Ethylene-treatedpetals were exposed to 10 pl/l of ethylene for 90 min prior totransfer to 50-ml flasks. At the appropriate times, the ethylenecontent of the headspace was determined as well as the A of thebathing medium at 575 nm. The petals were removed and ex-tracted in boiling isopropanol (1 ml) in a 5-ml conical glass tissuehomogenizer. The extract was transferred to a new 18 x 150-mmdisposable glass test tube to which 4 ml chloroform and 1 mlmethanol were added. The combined organic phase was thenpurified as before (1). Aliquots of this organic phase were evapo-

1067 www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.

SUTTLE AND KENDE

rated and the phosphorus determined according to Rouser et al.(10).

Phospholipid Composition. Batches of 150 petals were extractedas described above in 10 times the volume of each solvent.Following purification, the organic phase was evaporated andthen taken up in 1 ml chloroform. This was applied to a silicicacid column and fractionated according to Beutelmann and Kende(1) to obtain a phospholipid-enriched fraction. This fraction wasevaporated and redissolved in 50 Al of chloroform-methanol (2:1,v/v), and 25 1I were applied to a silica gel TLC plate (precoatedplastic sheets, Sil Gel, without binder, Brinkmann Inst.). Theplates were developed in one dimension in chloroform-methanol-acetic acid-water (85:15:10:3.5, v/v). The phospholipids were iden-tified by co-chromatography with authentic standards and bygroup-specific reagent sprays such as molybdenum, ninhydrin,and Dragendorf reagent (13). For quantitative determination, thephospholipids were localized with iodine vapor and scraped fromthe plate. The phospholipids were eluted, and the phosphoruscontent was determined as before.

Characterization of Endogenous Phospholipase Activity. En-dogenous phospholipases were characterized by homogenizinggroups of 12 petals in 0.3 ml of 0.1 M acetate buffer (pH 5.5)containing 5 mM DTT, 25 mm CaCl2, and 13.5 nCi of [U-'4C]-phosphatidylcholine (1.8 Ci/mmol, New England Nuclear). Thesehomogenates were allowed to stand for various lengths of time,and the reaction was stopped by the addition of 1 ml of boilingisopropanol. The organic phase was purified as above and evap-orated under a stream of N2. It was redissolved in 75 ,ul ofchloroform-methanol (2:1, v/v). An aliquot was chromatographedas before on silica gel plates. Improved resolution was accom-plished by two successive developments in the same direction. Thefirst solvent system was acetone-petroleum ether (3:1, v/v), thesecond chloroform-methanol-acetate acid-water (80:15:10:3.5, v/v). Following TLC, the plates were scanned for radioactivity. Theradioactive zones were scraped from the plates and the radioactiv-ity was determined by scintillation counting.

Phospholipase D Activity. Phospholipase D activity was deter-mined by extracting 1 g (fresh weight) of petals inlO ml of 0.1 Macetate buffer (pH 5.5) containing 5 mm DTT and 25 mm CaCl2(extraction buffer). The extract was allowed to stand for 1 h at 4 Cand was then centrifuged for 10 min at 13,000g. The supernatantwas used as a source of enzyme. The substrate (phosphatidylcho-line) was prepared by adding 0.9 mg phosphatidylcholine (Sima)containing 22.5 nCi phosphatidylcholine-choline-methyl-[ 4C1-methyl (50 mCi/mmol, New England Nuclear) in ether to a 50-ml flask. Following evaporation of the ether, 2 ml of the extractionbuffer were added to the flask, and the mixture was sonicatedfor 10 min. Three ml of the enzyme preparation were added to theflask, and the mixture was incubated at 28 C for I h. Followingincubation, I ml of the reaction mixture was removed and ex-tracted four times with petroleum ether. Activity is expressed aswater-soluble radioactivity from which a blank control preparedas above, but without enzyme, has been subtracted.

Quantitation of Phospholipase Activity. For each determinationof phospholipase activity, two groups of 12 physiologically equiv-alent petals were extracted. One group was homogenized in 1.0ml boiling isopropanol, and the amount of chloroform-methanol-soluble phosphorus was measured as described before. The othergroup was homogenized in 0.3 ml of 0.1 M acetate buffer (pH 5.5)containing 5 mM DTT and 25 mm CaCl2 (extraction buffer). Thehomogenate was allowed to stand for 1 h at 28 C at which time 1ml boiling isopropanol was added to stop the reaction. The lipidphosphorus was purified and assayed as described above. Thedifference in lipid phosphorus between the two determinationswas taken as a measure for endogenous phospholipase activity.

RESULTS

Characteristics of Cellular Efflux during Senescence. Duringthe course of senescence, isolated petals of Tradescantia exhibitedlarge increases in the rates of both anthocyanin and total electro-lyte efflux (14). The pattern of efflux of both types of cellularconstituents mirrored each other during senescence (Fig. 1). Theonset of the increased rate of efflux in untreated petals began after15:30 h on the day of flowering. If petals were continuously treatedwith 10 ,ul/l of ethylene beginning at 11:00 h, the increased rate ofefflux began after 13:30 h or approximately 2 h earlier. Again, therates of efflux of both types of compounds paralleled each otherthroughout the experiment. Because the content of anthocyaninin these petals remained constant during senescence, the increasedrate of efflux of this compound could not be explained on thebasis of concentration-dependent diffusion. The increased rate ofefflux must therefore be the result of increased membrane perme-ability.The Nature of Ethylene-enhanced Cellular Efflux. The results

presented in Figure 1 also demonstrate that there was a lagbetween the time of ethylene application (11:00 h) and the onsetof increased cellular efflux (13:30 h). The nature of this lag periodwas investigated using the protein synthesis inhibitor cyclohexi-mide. Figure 2 shows that the simultaneous application of 10 pl/1 ethylene and 0.1 mm cycloheximide resulted in the completeinhibition of the increase in anthocyanin efflux. However, if petalswere exposed to ethylene for 2 h prior to the application ofcycloheximide (still within the lag period) the response was onlypartially inhibited. Further experiments (not presented) showedthat simultaneous application of 10 ,ul/I ethylene and 0.1 mMcordycepin, an inhibitor oftranscription, also resulted in inhibitionof the ethylene-enhanced cellular efflux. On the other hand,simultaneous application of ethylene and the following transcrip-tion inhibitors was found to have no effect on the subsequentincrease in pigment efflux: 15 ,ug/ml actinomycin D, 0.1 mM 5-fluorouracil, or 0.1 mm 6-methylpurine.

Phospholipid Loss and Anthocyan_n Efflux. Because phospho-lipid loss has been implicated in both ethylene-enhanced andsenescence-related increases in cellular permeability (1, 12), wenext investigated the relationship between increases in anthocy-anin efflux and total phospholipid content in senescing petals ofTradescantia. The onset of membrane deterioration (as judged byincreased pigment efflux) began after 12:30 h (Fig. 3). Analysis ofphospholipid levels (Fig. 3C) showed that, prior to the onset ofincreased permeability, the phospholipid content first rose andthen fell sharply as pigment efflux increased. When petals were

pretreated for I h with 10 pJ/1 ethylene, the onset of pigmentefflux could be detected after 11:30 h, or approximately 60 minearlier than in control petals (Fig. 3B). The onset of phospholipid

-4040

0m 120Ca 100W 8004 600-1u 40z 200C' 0

140

/R

3 1510 19TIME OF DAY(h)

0.4

0.34

0.2 r..it)4

0.1

0

FIG. 1. Leakage of anthocyanin and electrolytes during natural and

ethylene-induced senescence. Efflux of anthocyanin and total electrolyteswas determined in two groups of 18 petals floated on deionized, glass-distilled H20. Ethylene-treated petals (- - -) were continuously exposedto 10 td/l of ethylene beginning at 11:00 h. (-A): anthocyanin efflux;(OL): electrolyte efflux.

Plant Physiol. Vol. 65, 19801068

www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.

ETHYLENE AND MEMBRANE INTEGRITY

41

13 15TIME OF DAY(h)

FIG. 2. Effect of cycloheximide on anthocyanin efflux in petals contin-uously exposed to 10 ul/1 ethylene. Ethylene treatment was begun at 10:30h. Cycloheximide (final concentration 0.1 mM) was added at the time ofethylene application (--- -) or 2 h afterwards (-M-), and the subsequentefflux of anthocyanin monitored.

c

CM0

Sc)r-In

cf)

Ic-

0a-

0

08 10 12 14 16 18TIME OF DAY(h)

FIG. 3. Effects of a 1-h pretreatment with 10 1.d/l ethylene on thesubsequent rates of (A) ethylene production, (B) anthocyanin efflux, and(C) phospholipid levels in mature petals. Ethylene was administered from09:00-10:00 h. Each point represents the assay of a separate group of 12petals. Values are expressed on a per g fresh weight basis. --- -) ethylene-treated petals; ( ): control petals.

decline commenced with the increase in pigment efflux, and theinitial rise in phospholipid content was much reduced, as com-

pared to the controls (Fig. 3C).Both AVG3 and a 4% CO2 atmosphere retarded the rate of

3 Abbreviation: AVG: aminoethoxyvinylglycine (L-2-amino-4-[2'-ami-noethoxy]trans-3-butenoic acid).

anthocyanin efflux in mature petals of Tradescantia (14). Petalswhich were continuously exposed to both of these inhibitors ofsenescence exhibited a much reduced rate of pigment efflux whencompared to control petals, and this was accompanied by acorresponding reduction in the rate of phospholipid loss (Fig. 4).As mentioned before, cycloheximide is very effective in arresting

the increase in pigment efflux normally seen during natural orethylene-induced senescence in Tradescantia (Fig. 2). Cyclohexi-mide also arrested the decline in phospholipid levels normallyobserved during petal senescence (Fig. 5).These results demonstrate a quantitative correlation between

phospholipid loss and increased membrane permeability. Further,they indicate that the observed increase in permeability is due tothe decline in the phospholipid content. These results do notexclude the possibility that the increase in membrane permeabilityleads to a release of phospholipases from a cellular compartment(vacuole?) and that these phospholipases cause the increased rateof phospholipid loss. To test this possibility, the following exper-iment was performed: petals were first treated with cycloheximidein order to prevent any increases in phospholipase content andwere then placed under a N2 atmosphere, which induced cellularleakage and loss of compartmentation. Whereas anoxia resultedin a constant, but much increased rate of anthocyanin efflux, nosubstantial loss of phospholipid was observed (Fig. 5). Thus, thedecrease in phospholipid content cannot be explained solely byan increase in vacuolar efflux.

Phospholipid Composition. Table I shows the composition ofthe phospholipid fraction prior to and following the initial increasein membrane permeability. Phosphatidylethanolamine (65%) andphosphatidylcholine (23%) were the major phospholipid speciesin mature petals. Phosphatidylglycerol (8%) and phosphatidyli-nositol (5%) were minor species. Analysis of the phospholipidcomposition following the onset of membrane leakage showedthat the per cent composition remained unchanged during this

20OA

c 14_ CONTROLst10_6:2 _ §AVG+CO02~~~~~~~4-B

3J

P.- 2-_g/S

08 10 12 14 16 18TIME OF DAY(h)

FIG. 4. Effects of continuous exposure to 0.1 mM AVG and a 4% CO2atmosphere on (A) ethylene production, (B) anthocyanin efflux, and (C)phospholipid levels in mature petals. Each point represents the assay of aseparate group of 12 petals. Values are expressed on a per g fresh weightbasis. (---): AVG + C02; ( ): control.

1069Plant Physiol. Vol. 65, 1980

www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.

Plant Physiol. Vol. 65, 1980

1.81.61.41.21.0

4 0.80.60.40.2

cnm 400zXj 300

IL 20

.iI15

.7

x~~~~~~~1-

i i i i i iB

I I I I I I 109 11 13 15 17 19

TIME OF DAY(h)

FIG. 5. Effect of anaerobiosis on (A) anthocyanin leakage and (B) lipidphosphorus in petals continuously exposed to 0.1 mm cycloheximide.Cycloheximide treatment was begun at 09:00 h and anaerobiosis wasinitiated at 10:30 h. Each point represents the assay of a separate group of12 petals. Values are expressed on a per gram fresh wt basis. (--- -): N2+ cycloheximide; ( ): cycloheximide.

Table I. Phospholipid Composition prior to Senescence and at anAdvanced Stage of Senescence in Isolated Petals of Tradescantia

Groups of 150 petals were homogenized in boiling isopropanol, and aphospholipid-enriched fraction was isolated as described in the text.Phospholipid composition was determined after separation by TLC.Time of Extraction pCa PE PG PI

h % ofphospholipid09:00 23.0 64.6 7.6 4.816:00 22.0 66.0 9.0 3.0

a PC: phosphatidylcholine; PE: phosphatidylethanolamine; PG: phos-phatidylglycerol; PI: phosphatidylinositol.

phase of senescence.Enzymes of Phospholipid Catabolism. The catabolic sequence

of phospholipid breakdown in senescing petals was characterizedby homogenizing the petals in a small volume of buffer containing[U-_4C]phosphatidylcholine (about 13.5 nCi). The time course ofphospholipid breakdown in crude homogenates isolated frompetals at an advanced stage of senescence is shown in Figure 6.With time, there was a decline in the phosphatidylcholine contentwhich was accompanied by a corresponding increase in free fattyacids. Analysis of the rate of hydrolysis in this system showed thatit was highest during the first 5 min of incubation whereupon itdeclined. Phosphatidic acid, a possible intermediate of phospho-lipid breakdown, and lysophosphatidylcholine were not observed,which indicated that acyl hydrolase rather than phospholipase Dactivity predominated during this phase of petal senescence. Phos-pholipase D activity was indeed found to decline dramaticallyduring the course of petal senescence (Table II). The endogenouscapacity for phospholipid destruction in these petals was deter-mined at three stages of senescence by taking advantage of thefact that endogenous phospholipids are readily hydrolyzed incrude homogenates of plant tissues unless special precautions aretaken (4). The experimental design was as follows: (a) batches ofpetals to be assayed were divided into two equivalent groups; (b)one group was homogenized in boiling isopropanol to determinethe initial phospholipid content; (c) the other group was homog-enized in buffer and allowed to stand for I h at 28 C prior to

I-

I-

u0ar

-

-Jw

0 0.5 1.0R f VALUE

FIG. 6. Characterization ofendogenous phospholipase activity in crudehomogenates of petals isolated at an advanced stage of senescence. Thecatabolic sequence of phospholipid breakdown was followed by homoge-nizing groups of petals in buffer containing [U-'4Clphosphatidylcholine(13.5 nCi). The homogenates were allowed to stand for (A) 5 min, (B) 15min, (C) 30 min, and (D) 60 min. The phospholipid fraction was isolatedand subjected to TLC. Bars indicate the position of authentic standards.LPC: lysophosphatidylcholine, PC: phosphatidylcholine, PA: phosphatidicacid, and FFA: free fatty acids.

Table II. Phospholipase D Activity during Senescence of Isolated Petals ofTradescantia

The enzyme was assayed by measuring the release of water-solubleradioactivity following a 1-h incubation (28 C) with I ,umol phosphatidyl-choline containing ['4Clcholine-methyl-phosphatidylcholine (22.5 nCi).Activity is expressed as water-soluble cpm over an enzyme blank after foursuccessive extractions of the aqueous phase with petroleum ether.

Time of Phospholipase D InitialExtraction Activity Activity

h cpm %09:00 17,715 10013:00 2,935 1715:30 1,590 9

killing in boiling isopropanol; and (d) the difference in chloro-form-methanol soluble phosphorus was taken as a measure of theendogenous phospholipase activity. The endogenous capacity todegrade phospholipids remained essentially unchanged duringpetal senescence (Table III).

1070 SUTTLE AND KENDE

www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.

ETHYLENE AND MEMBRANE INTEGRITY

Table III. Endogenous Phospholipase Activity in Crude Homogenates ofPetals Isolated at Various Stages of Senescence

Chloroform-methanol-soluble phosphorus was determined before andafter a 1-h incubation (28 C) in 0.3 ml acetate buffer (pH 5.5) as describedin the text. Phospholipase activity is expressed as the difference in chlo-roform-methanol-soluble phosphorus between the initial and final deter-minations.

Lipid PhosphorusTime of

Extraction Initial Final PhospholipaseValue Value Activity

h tg g' fresh wt10:30 44.99 29.09 15.9014:00 38.01 27.21 10.8016:00 32.82 17.61 15.21

DISCUSSION

Isolated petals of Tradescantia exhibit an increase in bothanthocyanin and electrolyte efflux during senescence (14). Theresults presented in Figure 1 show that the rates of efflux of thesetwo compounds proceed in parallel, which suggests that the in-crease in membrane permeability is of a general nature. Thisindicates that gross alterations in membrane integrity occur duringsenescence. It is not known if the increase in tonoplast permeabil-ity, as demonstrated by the increase in anthocyanin efflux, alsoleads to an enhanced efflux of other, possibly more physiologicallyrelevant, vacuolar constituents such as enzymes.

As in other senescing tissues (6, 1 1), application of ethylene toisolated petals of Tradescantia has been shown to hasten the onsetof the increase in membrane permeability (14). The results pre-

sented in Figures 1 and 2 demonstrate that ethylene does notdirectly affect membrane integrity but rather that the gas exerts itseffect on permeability through cellular metabolism. This conclu-sion is consistent with the observed increase in ethylene sensitivityof both flower and fruit tissues during maturation (6, 14).The lag of 90-150 min between the application of ethylene and

the increase in anthocyanin efflux (Fig. 2), together with theinhibitory action of cycloheximide (Fig. 2), indicates that ethyleneaction in senescing petals of Tradescantia requires protein synthe-sis. Furthermore, it appears that proteins synthesized within 2 hof ethylene application are essential in mediating the increase inmembrane permeability because application of cycloheximideafter this period only partially inhibits ethylene-enhanced antho-cyanin efflux. In addition, the ability of cordycepin, an inhibitorof transcription, to block, ethylene-enhanced anthocyanin effluxindicates that RNA synthesis may also be required for ethyleneaction. It seems that the increase in membrane permeability causedby the application of ethylene is a secondary effect of the gas

which is mediated by other cellular processes requiring both RNAand protein synthesis.Data indicate that the observed increase in membrane perme-

ability is the result of phospholipid loss (Figs. 3 and 4). Factorswhich hasten the onset of pigment efflux (ethylene pretreatment)or inhibit the increase in permeability (CO2 and AVG or cyclo-heximide) exert a corresponding effect on the rate of phospholipidloss. Because only total phospholipid levels have been determined,it is not known if the rate of loss of phospholipids in the varioussubcellular organelles proceeds in parallel or if the membranes ofcertain organelles are preferentially affected. These results are ingeneral agreement with other investigations (1, 2, 12) which havedemonstrated that loss of membrane integrity during senescence

is accompanied by losses of phospholipids.The observed decline in phospholipids during senescence in

petals of Tradescantia can be explained by a decrease (or cessation)of phospholipid synthesis, by increased phospholipid catabolism,

or by both. Inhibitors of respiration have been shown to block thesynthesis of phospholipids in plant tissues (3). Cycloheximide andsustained anaerobiosis together have little effect on phospholipidlevel in these petals (Fig. 5). This observation indicates that theloss of phospholipids during petal senescence is due to an increasedrate of catabolism. Further support for this conclusion can bederived from the data presented in Table I. Although there is a70%o reduction in phospholipid content during petal senescence,the phospholipid composition remains essentially unchanged,demonstrating that all phospholipids are lost in parallel.

Surprisingly, the in vitro assays of endogenous phospholipaseactivities have failed to demonstrate any increase in these activitiesduring senescence (Tables II and III). Analysis of Figure 3 showsthat the maximum rate of phospholipid decline during petalsenescence is 4.77 jig phospholipid degraded g-' fresh weight h-',and the results presented in Table III show that enough phospho-lipase activity is present in petals prior to the onset of membranedeterioration to account for this maximum rate of phospholipidloss. Although loss of phospholipids during senescence is welldocumented, very little information is available concerning therole of endogenous phospholipases in this process (5). It has beenshown that in senescing mung bean cotyledons the catabolicsequence of phosphatidylcholine breakdown is the following:phosphatidylcholine -- phosphatidic acid -- lysophosphatidicacid -* free fatty acid (7). Figure 6 indicates that a differentsequence of catabolism occurs in senescing petals of Tradescantia.In senescing petals, phosphatidylcholine is directly deacylated toyield free fatty acids without accumulation of a lyso intermediate.The addition of 40 ,ug unlabeled lysophosphatidylcholine to the invitro assay system (not shown) did not increase the amount ofradioactivity associated with lysophosphatidylcholine, indicatingthat it is not a transient intermediate in this catabolic system. Thefact that no phosphatidic acid has been detected during the in vitrocharacterization indicates that phospholipase D activity is verylow in senescing petals of Tradescantia, and Table II substantiatesthis conclusion. The observed decline in phospholipase D activityduring petal senescence is consistent with earlier investigations (9)which demonstrated that the activity of this enzyme is highest inyoung, actively growing tissues and that it declines with age.

Senescence in mung bean cotyledons is also associated with amarked decline in phospholipid levels but without any demon-strable increase in phospholipase activities (7). These results havebeen taken as an indication that phospholipases are sequesteredin healthy tissues and that this compartmentation is lost duringsenescence thereby leading to an enhanced rate of catabolism.The vacuole has been proposed as the plants' equivalent of theanimal lysosome (8) and therefore is a prime candidate for thesubcellular localization of phospholipase activities. Figure 5 showsthat disruption of the tonoplast (as evidenced by enhanced antho-cyanin leakage) is not sufficient to initiate phospholipid catabo-lism in these petals. However, it is not known if enhanced antho-cyanin efflux is also accompanied by enhanced efflux of vacuolarproteins (i.e. hydrolases).

In summary, the action of ethylene in increasing membranepermeability in senescing petals of Tradescantia appears to bemediated by processes requiring both RNA and protein synthesis.Furthermore, these results suggest that the observed increase inmembrane permeability is a direct result of an increase in phos-pholipid degradation presumably caused by an increase in activityof preexisting phospholipases.

LITERATURE CITED

1. BEUTELMANN P, H KENDE 1977 Membrane lipids in senescing flower tissue ofIpomoea tricolor. Plant Physiol 59: 888-893

2. FERGUSON CHR, EW SIMON 1973 Membrane lipids in senescing green tissue. JExp Bot 24: 307-316

3. FERGUSON CHR, EW SIMON 1973 The effect of iodoacetate on phospholipidlevels and membrane permeability. J Exp Bot 24: 841-846

Plant Physiol. Vol. 65, 1980 1071

www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.

1072 SUTTLE A

4. GALLIARD T 1970 The enzymic breakdown of lipids in potatoe tuber by phos-pholipid- and galactolipid-acyl hydrolase activities and by lipoxygenase. Phy-tochemistry 9: 1725-1734

5. GALLIARD T 1973 Phospholipid metabolism in photosynthetic plants. In G BAnswell, J N Hawthorn, RMC Dawson, eds, Form and Function of Phospho-lipids, Elsevier, New York, pp 253-284

6. HANSON AD, H KENDE 1975 Ethylene-enhanced ion and sucrose efflux inmorning glory flower tissue. Plant Physiol 55: 663-669

7. HERMAN E, M CHRISPEELS 1978 Catabolism of phospholipids and compartmen-tation of phospholipases in cotyledons. Plant Physiol 61: S-15

8. MATILE P 1978 The biochemistry and function of vacuoles. Annu Rev PlantPhysiol 29: 193-213

9. QUARLES RH, RMC DAWSON 1969 The distribution of phospholipase D in

ID KENDE Plant Physiol. Vol. 65, 1980

developing and mature plants. Biochem J 112: 787-79410. ROUSER G, S FLESCHER, A YAMAMOTO 1970 Two dimensional thin layer

chromatographic separation of polar lipids and determination ofphospholipidsby phosphorus analysis of spots. Lipids 5: 494496

11. SACHER J 1973 Senescence and postharvest physiology. Annu Rev Plant Physiol24: 197-224

12. SIMON EW 1974 Phospholipids and plant membrane permeability. New Phytol73: 377-420

13. SKIPSKI VP, M BARCLAY 1969 Thin layer chromatography of lipids. MethodsEnzymol 14: 530-598

14. SUTTLE JC, H KENDE 1978 Ethylene and senescence in petals of Tradescantia.Plant Physiol 62: 267-271

15. THIMANN KV 1978 Senescence. Bot Mag Tokyo Special Issue 1: 19-43

I _ w wNI

www.plantphysiol.orgon March 5, 2020 - Published by Downloaded from Copyright © 1980 American Society of Plant Biologists. All rights reserved.