tomato fruit cell - plant physiology. solubilization appearsto bea consequenceof autohydro-lysis...

Plant Physiol. (1989) 91, 816-8220032-0889/89/91/081 6/07/$01 .00/0

Received for publication December 27, 1988and in revised form June 1, 1989

Tomato Fruit Cell Wall1

1. Use of Purified Tomato Polygalacturonase and Pectinmethylesterase to IdentifyDevelopmental Changes in Pectins

James L. Koch2 and Donald J. Nevins*Department of Vegetable Crops, University of California, Davis, California 95616

ABSTRACT

Cell wall isolation procedures were evaluated to determinetheir effect on the total pectin content and the degree of methy-lesterification of tomato (Lycopersicon esculentum L.) fruit cellwalls. Water homogenates liberate substantial amounts of buffersoluble uronic acid, 5.2 milligrams uronic acid/100 milligramswall. Solubilization appears to be a consequence of autohydro-lysis mediated by polygalacturonase 11, isoenzymes A and B,since the uronic acid release from the wall residue can be sup-pressed by homogenization in the presence of 50% ethanolfollowed by heating. The extent of methylesterification in heat-inactivated cell walls, 94 mole %, was significantly greater thanwith water homogenates, 56 mole %. The results suggest thatautohydrolysis, mediated by cell wall-associated enzymes, ac-counts for the solubilization of tomato fruit pectin in vitro. Endog-enous enzymes also account for a decrease in the methylesteri-fication during the cell wall preparation. The heat-inacffvated cellwall preparation was superior to the other methods studied sinceit reduces,-elimination during heating and inactivates constitu-tive enzymes that may modify pectin structure. This heat-inacti-vated cell wall preparation was used in subsequent enzymaticanalysis of the pectin structure. Purified tomato fruit polygalac-turonase and partially purified pectinmethylesterase were usedto assess changes in constitutive substrates during tomato fruitripening. Polygalacturonase treatment of heat-inactivated cellwalls from mature green and breaker stages released 14% of theuronic acid. The extent of the release of polyuronides by poly-galacturonase was fruit development stage dependent. At thetuming stage, 21% of the pectin fraction was released, a valuewhich increased to a maximum of 28% of the uronides at the redripe stage. Pretreatment of the walls with purified tomato pecti-nesterase rendered walls from all ripening stages equally sus-ceptible to polygalacturonase. Quantitatively, the release of uron-ides by polygalacturonase from all pectinesterase treated cellwalls was equivalent to polygalacturonase treatment of walls atthe ripe stage. Uronide polymers released by polygalacturonasecontain galacturonic acid, rhamnose, galactose, arabinose, xy-lose, and glucose. As a function of development, an increase inthe release of galacturonic acid and rhamnose was observed (40and 6% of these polymers at the mature green stage to 54 and15% at the red ripe stage, respectively). The amount of galactoseand arabinose released by exogenous polygalacturonase de-creased during development (41 and 11% from walls of maturegreen fruit to 11 and 6% at the red ripe stage, respectively). Minoramounts of glucose and xylose released from the wall by exog-

' Supported, in part, by a gift from Chesebrough-Ponds Inc.2 Present address: Department of Agronomy, 621 Bradfield Hall,

Cornell University, Ithaca, NY 14853-0144.

enous polygalacturonase (4-7%) remained constant throughoutfruit development.

The appearance of PG3 during fruit maturation in tomatoand its correlation to ripening and softening events has beenwell documented (6, 9). Previous reports focused on thephysical properties of the enzyme including temperature ofinactivation and specific ion interactions (3). These featuresof PG action have generally been evaluated using substrates,usually citrus pectin, of unspecified composition. Little infor-mation is available concerning the effect of tomato fruit PGon constitutive wall pectins. In two reports, walls from maturegreen tomato fruit were employed to examine PG action (24,28), but the extent of pectin hydrolysis in walls prepared atdifferent developmental stages was not evaluated.

Similarly, most previous reports on PE have been per-formed using polymers extracted from diverse sources (21,23). One must be concerned that by virtue of extractionprocedures, these soluble polymers may be significantly mod-ified both in structure and in hydrolytic susceptibility bydeesterification or hydrolysis of glycosidic bonds.Methods of cell wall preparation are critical in the evalua-

tion of component polysaccharide structure. Chemical treat-ments with salts (8), acid (26), alkali (10, 13, 16) and organicsolvents (26) may modify or extract certain polymers fromthe wall. Similarly, the presence of active cell wall hydrolasesmay modify cell wall polymers. Pectic components of the cellwall are especially susceptible to chemical treatments. Heatand alkali treatments mediate ,-elimination reactions whichreduce overall polymer length (2, 16). Furthermore, alkalinepH can deesterify methyl groups (31) and associated phenolics(7) from the polysaccharide. Treatment of cell walls with

3 Abbreviations: PG, polygalacturonase II, isoenzymes A and B;PE, pectinmethylesterase; MG, mature green stage fruit; B, breakerstage fruit; T, turning stage fruit; P, pink stage fruit; R, red ripe stagefruit; HCW, heat-inactivated cell walls; AWR, autolyzed MG cellwall residue; Rha, rhamnose; GalUA, galacturonic acid; UA, uronicacid; CMC, I -cyclohexyl-3-(2-morpholinoethyl)-carbodiimidemetho-p-toluenesulfonate; RG I, rhamnogalacturonan I; CM, chloroform-methanol prepared walls; H-CM, heat-inactivated chloroform-meth-anol prepared walls; EtOH-CM, ethanol-treated chloroform-metha-nol prepared walls; DE, degree of methylesterification (mol %).

816 www.plantphysiol.orgon June 3, 2018 - Published by Downloaded from

Copyright © 1989 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

DEVELOPMENTAL CHANGES IN TOMATO PECTINS

organic solvents such as chloroform and acetone dehydratethe cell wall. Dehydration may alter hydrophobic interactionsand change hydrogen bonding between polymers. Physicalcharacteristics are modified during dehydration, and thesechanges may restrict subsequent rehydration. Accessibilitywithin complex polymers may impair the capacity ofenzymesto act on polysaccharide components.

Surprisingly little is known concerning cell wall modifica-tions during senescence of processing cultivars of tomatoes.Selection through plant breeding has generated processinglines that remain firmer throughout ripening while retainingother characteristics including ethylene production and en-zyme induction. While the ripening mutants of fresh markettomatoes such as rin and nor have been used to evaluatetexture and softening events related to the induction of spe-cific enzymes, the interpretation of the results is confoundedby substantial modifications of the ripening mechanism ofthese genotypes (29). This fact evokes certain reservations inthe use of these mutants for studying specific fruit softeningmechanisms. An investigation of processing varieties maylead to an enhanced understanding of the role that the cellwall plays in fruit softening and may provide a biochemicalbasis for interpreting changes in texture. Ultimately, an un-derstanding ofthe texture modification mechanism in tomatofruit may be employed to improve the quality of processedproducts.

In this study, the pectic components ofthe tomato fruit cellwalls were examined. Cell wall isolation methodology wasexamined to determine the extent that physical changes,which occur during cell wall isolation, affect the physiologicalinterpretation of pectin chemistry. Second, a comparison ofcell walls from different developmental stages was used toexamine the effect of PG on the pectin content of the fruitcell walls. By utilizing purified tomato fruit PG and PE, andcell walls extracted using a method which minimizes modifi-cation, the susceptibility of the cell wall pectin to PG wasexamined as a function of the degree of esterification of itsnative substrate.

MATERIALS AND METHODS

Plant Material

Field grown tomato fruit (Lycopersicon esculentum cv VF145B-7879) were harvested and visually sorted by ripeningstage (5). Fruit were surface-sterilized for 10 min in 0.1%hypochlorite, rinsed in water, halved, and the locular fluidwas removed with a spatula. The pericarp with skin attachedwas frozen in liquid N2 and maintained at -20°C.

Preparation of Tomato Fruit Cell Walls

Extraction oftomato fruit cell walls is outlined in Figure 1.All aqueous solutions contain 0.02% thimerosal (Sigma) tosuppress microbial growth. Fifty g (frozen weight) of tomatofruit pericarp were peeled and homogenized in a Waringblender with cold water or cold 50% (v/v) ethanol. Cold waterhomogenates were washed with 1 L cold water over a coarsesintered glass funnel, followed by 1 L chloroform-methanol(1:1), and 1 L acetone. The acetone washed cell wall residue

Figure 1. Flow chart for tomato pericarp cell wall preparation.

was then vacuum-dried at 22°C (CM walls). A subfraction ofthe acetone-washed walls was resuspended in 50% ethanol,heated to 80°C and, allowed to reflux 20 min to heat-inactivatecell wall hydrolases. The heat-treated cell walls were washedwith acetone as described above and vacuum-dried at 22°C(H-CM walls).

Ethanol homogenates were prepared using two methods.First, the homogenate was washed in 1 L 50% ethanol over acoarse sintered glass funnel followed by 1 L chloroform-methanol (1:1), and 1 L acetone. The acetone washed cellwalls were vacuum-dried at 22°C (EtOH-CM walls). Second,cell wall homogenates were heated to 80°C for 20 min toirreversibly inactive endogenous enzymes. Heated walls werewashed in 50% ethanol, then with acetone and vacuum-drivedat 22°C (HCW).

Preparation of Autolyzed Native Substrate

Because mature green (MG) fruit tissue contains substantialPE activity, but no PG, it was feasible to exploit autolysis todeesterify native pectins (24, 30). Frozen MG fruit (50 g) werepeeled and homogenized in cold water (4°C). The slurry waswashed in cold water (2 L) with aid of suction over a 70 ,umnylon mesh, washed in acetone (-20°C), and rinsed in coldwater. Excess water was expressed, and the wall material wasincubated in 200 mL of 100 mm acetate buffer (pH 5.2) for24 h. After autolysis, cell walls were washed in 20 mM acetatebuffer (pH 5.0) over a 30 uim nylon mesh to remove solublesugars, and the excess liquid was expressed. The AWR wasfrozen and maintained at -20°C.

Ethanol Inhibition of PG Activity

Cell walls from ripe fruit were prepared as described abovefor MG fruit to preserve autolytic activity. The capacity forautolysis was evaluated in 100 mm acetate buffer (pH 5.2)diluted with an appropriate amount of95% ethanol to achievefinal ethanol concentrations of 10, 20, 30, 40, 50, 60, 70, 80,and 90% (v/v). Following a 6 h incubation at 22°C, the cellwall suspension was centrifuged (1000g, 5 min) to sediment

817

www.plantphysiol.orgon June 3, 2018 - Published by Downloaded from Copyright © 1989 American Society of Plant Biologists. All rights reserved.


Plant Physiol. Vol. 91, 1989

the insoluble solids. The supernatant fluid was assayed forUA content according to Blumenkrantz and Asboe-Hansen(4).

Analysis of Wall Carbohydrate Components

Total UA content (1) and the degree of esterification (31)of the cell walls were determined after dehydrating nativewalls with acetone followed by vacuum drying. Soluble UAwere measured according to Blumenkrantz and Asboe-Han-sen (4). Neutral sugar composition was determined accordingto Hatfield and Nevins (12). Identification of galacturonicacid in polymers was accomplished after reduction ofcarboxylgroups with CMC (19). Quantitative conversion of galactu-ronic acid to galactose was confirmed by GLC of alditolacetates following TFA hydrolysis.

Isolation and Purification of PG and PE

PG was isolated from field-grown, ripe VF145B-7879 proc-essing tomatoes using the procedure of DellaPenna et al. (6).Following the elution ofPG with 10 mM a-methylmannosidefrom a concanavalin A-Sepharose 4B column, the PG fractionformed a single band on SDS-PAGE. Activity correspondingto PG was associated with the protein band on a nondenatur-ing gel (20).The void fraction from concanavalin A chromatography,

comprised ofPE, revealed five major proteins on SDS-PAGE.The PE fraction did not release any detectable carbohydratederived from cell wall polymers when assayed with tomatofruit walls as substrate. Glycosidase activity was absent evenafter 1 h incubation of this fraction when assayed accordingto Wallner and Walker (30) with the following p-nitrophenylsubstrates: a- and O-D-glucoside, a- and f3-D-galactoside, a-and f3-D-xyloside, a- and ,B-D-mannoside, fl-D-galacturono-side, a-L-rhamnoside, and fl-L-fucoside (a-D-xyloside and a-D-mannoside were purchased from Koch-Light Laboratories,all others from Sigma Chemical Co.). The PE fraction con-tained some a-amylase. No differences in PG or PE activitiesnor changes in qualitative characteristics of column profileswere observed when freshly harvested tissue was compared totissue frozen and stored at -20°C.

PG Hydrolysis

Aliquots of PG (5 ,ug, 800 ,tg GalA reducing equivalentsh-') were used in all experiments. The extent of PG-mediatedhydrolysis was determined after incubation of 5 to 10 mgHCW material in 2 mL 20 mm acetate (pH 5.0) for 24 h.After treatment, the insoluble cell wall material was collectedby centrifugation, 5OOg for 2 min, and the supernatant fluidanalyzed for carbohydrate content. Released polymers wereanalyzed for UA content as described above.

PE Pretreatment of HCW

Cell walls, with endogenous enzymes inactivated by heattreatment, were incubated with 5 ,ug PE, with activity capableof releasing 1.5 ,ug methanol min-'. The incubation wascontinued for 12 h in 2 mL 20 mm acetate buffer (pH 5.0).

The degree of polymer methylesterification was monitoredaccording to Wood and Siddiqui (31). Because ofthe presenceof a-amylase in the PE preparation, cell walls residues werewashed with 20 mL of the incubation buffer over a 30 ,umnylon mesh to remove products of starch digestion. Theretained residues were used for subsequent evaluations ofPGsusceptibility.

RESULTS

Total Cell Wall Pectin Content

Ethanol was employed to dehydrate tissue or precipitate ordenature proteins. The presence of 50% ethanol reversiblyinhibits PG activity in vitro thereby suppressing autolysis ofripe tomato fruit tissue (Fig. 2). The total UA content in thepreparation, which quantitatively reflects the amount of pec-tin in red ripe tomato fruit, was similar in all five isolationprocedures (Table I). The total UA of CM walls and HCWwalls is identical, but the difference in soluble uronides sug-gests that heat inactivation in the presence of ethanol mayprotect the pectin polymers from degradation and thus reducepectin loss during tissue preparation.

Buffer Soluble Pectin Content

When the soluble components of wall preparations areexamined after incubation in buffer at pH 5.0 (Table I), allpreparations except HCW contained some solubilized pectin.Since the total pectin content is the same in these extractions,we propose that solubilization of pectins is due to in vitro PGactivity during the aqueous phase of the cell wall preparation.This hypothesis is supported by the observation that theamount of soluble UA is similar between the CM preparation

10

izO

E00

*40.C

E

0 10 20 30 40 50 60 70 80 90 100

% Ethanol

Figure 2. Inhibition of the autolytic release of UA from ripe fruit cellwalls by ethanol. Autolytically active red ripe fruit cell walls weretreated with ethanol at different levels during the course of a 6 hautolysis treatment in 100 mm acetate buffer (pH 5.2) at 220C (0).Red ripe walls were treated with indicated ethanol levels for 1 h,centrifuged (1 000g, 5 min) to sediment cell walls, and the walls wereincubated in 20 mm acetate buffer for 6 h (0). No difference inautolysis rate was observed at acetate buffer concentrations between10 and 100 mm (data not shown). Values represent the mean of atleast three assays.

818 KOCH AND NEVINS




Table I. Total UA Content of Red Ripe Tomato Fruit Cell WallsWall preparations are identified in Figure 1. Values represent mean

+ SE of at least three measurements from two separate wall prepa-rations of ripe tomato fruit. Total UA was determined according toAhmed and Labavitch (1) after acetone substitution of water followedby vacuum drying. Soluble UA was determined by assaying thesupematant (1 000g, 5 min) following a 24 h incubation of cell wallsin 20 mm acetate buffer (pH 5.0) at 220C.

Wall Preparation Total UA Soluble UAmg UA/100 mg wall

HCW 18.5 ± 0.8 0.0 ± 0.0CM 18.4 ± 1.0 5.2 ± 0.3EtOH-CM 16.4 ± 1.2 2.7 ± 0.4H-CM 17.4 ± 1.9 1.0 ± 0.2Autolytically active walls 21.3 ± 2.3 6.0 ± 1.1

3:

E

00

E

6Time (h)

Figure 3. Autolytic release of UA from ripe fruit cell walls as a functionof temperature. Autolytically active red ripe fruit cell walls wereincubated in 100 mm acetate buffer (pH 5.2) at 4°C (0) and 25°C (0)for 2 to 12 h. The control (A) was the heat-inactivated cell wallpreparation (HCW) identified in Figure 1. At appropriate times, wallswere centrifuged (1 000g, 5 min), and the supematant fluid wasassayed for total UA according to Blumenkrantz and Asboe-Hansen(5). Values represent the mean ± SE of at least three determinations.

and that obtained by autolysis of red ripe fruit cell walls. Theamount of UA released by autolysis, 6.0 mg/100 mg wall(Fig. 3), and solubilized from CM walls, 5.2 mg/100 mg wall(Table I), is also similar to the amount of pectin released byspecific digestion of HCW with purified PG, 5.2 mg/100 mgwall.The amount ofUA solubilized by the autolysis of walls at

4°C for 30 min, ca. 1 mg/100 mg wall (Fig. 3), is the extentof hydrolysis anticipated during the aqueous phase of the CMextraction procedure. However, the amount ofUA solubilizedduring incubation in buffer after CM extraction, 5.2 mg/100mg wall (Table I), is much greater than the amount expected.Substantial PG activity might be preserved during the chlo-roform-methanol and acetone dehydration of the tissue or

alternatively, substantial ,-elimination occurs. The fact thatonly 1 mg UA/100 mg wall is solubilized from H-CM walls(Table I), a case where heat inactivation limits autolysis tothe time in an aqueous solution, reinforces the hypothesisthat PG activity is present in CM walls following dehydration.

The soluble pectins in cell wall extracts prepared in thepresence of ethanol constitute 2.7 mg UA/100 mg wall. Thisvalue is higher than the soluble components liberated fromHCW and H-CM. The increase in soluble UA after removalof ethanol suggests the presence of a heat-labile pectinase inthe acetone dehydrated CM and EtOH-CM walls. Theamount of UA liberated by heat in the absence of ethanol, 8mg/100 mg wall (data not shown), may reflect the amount ofpectin in the ripe fruit cell wall susceptible to ,8-eliminationreactions.

Degree of Methyl Esterification

Walls prepared by heat inactivation in ethanol (HCW)exhibit a DE of 94% in the tomato cultivar VF145B whilethose walls prepared in accordance with the CM scheme haveonly 56% DE (Table II). The lower DE in CM walls may bedue to autolytic action of PE during extraction. EtOH-CMwalls have a significantly higher DE, 72%, than CM walls. H-CM walls do not show any significant difference from CMwalls. We conclude that deesterification ofthe pectin, presum-ably by PE, occurs during the initial aqueous washes and littleor no deesterification occurs after chloroform-methanol andacetone dehydration of the cell walls. The action of PE israpid. Tomato fruit cell walls are deesterified within 10 minof the removal of ethanol inhibition (data not shown). Thus,in vitro PE activity may well account for the deesterificationof the pectin in CM walls. Pectin esterification from heat-inactivated cell walls, 94%, is approximately 20% higher thanpreviously observed (24, 28, 31). Differences in extractionmethodology may explain this discrepancy. The method forestimating methylester content does not discriminate betweenmethanol released from UA, proteins, or deterioration prod-ucts from alkali treatment of the cell wall (31). Following PEtreatment, 30% methylester content remains in heat-inacti-vated MG fruit cell walls, and MG fruit cell walls followingautolysis have a residual methylester content of 13%. Thisresidual methanol content in the cell wall preparation mayreflect the amount of non-UA methylesters or inaccessibilityof the substrate to PE.

Table II. Degree of Methyl Esterification of Mature Green Fruit CellWallsThe amount of methanol released from acetone dried cell walls

following a 30 min treatment with 0.5 M NaOH (31) was expressedas the degree of esterification, mol of methanol released by alkali permol UA. Values represent the mean ± SE of at least three measure-ments from two separate wall preparations. Wall preparations areidentified in Figure 1.

Wall Preparation Degree of Esterification Esterified UA

mol % mg UA/100 mg walla

HCW 93.8 ± 1.1 16.9 ± 1.0CM 55.9 ± 1.2 9.2 ± 1.2EtOH-CM 71.9 ± 1.0 13.6 ± 0.8H-CM 55.2 ± 4.1 9.6 ± 0.8AWR 11.2 ± 2.3 2.6 ± 1.1

a Assumes all NaOH-released methanol is from pectin.

819



Plant Physiol. Vol. 91, 1989

Composition of HCW Tomato Fruit Cell Walls

Cell walls of a processing tomato variety, VF 145B-7879,were analyzed for UA content and for the degree of uronidemethylesterification at selected ripening stages (MG, B, T, P,and R) (Fig. 4). The total UA content remained constant atabout 18% (UA weight/dry weight) throughout ripening. Thedegree of esterification decreased from 90% (mol methanol/mol UA) at MG and B stages to 35% at the pink and ripestages.

Changes in PG and PE-Susceptibile Polymers of TomatoFruit Cell Walls

Heat-treated tomato fruit cell walls were incubated in thepresence of PG until no additional UA was solubilized (24h). PG was capable of releasing 2.6 mg UA/100 mg fromheat-inactivated MG fruit cell walls, corresponding to 14% ofthe total UA (Fig. 5). PG mediated release of UA from wallsof the breaker stage was similar to that observed for MG

3:

E0

0

._

._

0L-

E

20-20-- -eV+

15-.

0MG B T P

Developmental StageR

Figure 4. Tomato fruit cell wall uronide composition. Total UA con-

tent of heat-inactivated cell walls (0). Methylesterified UA (). UAsusceptible to exogenous polygalacturonase (A). Values representthe mean ± SE of at least five separate tissue preparations.

a

30E0

0

.4

._

E

E

1UL

8--

6--

4

0

MG B T P R

Developmental Stage

Figure 5. UA released by treatment of walls with exogenous PG.Heat-inactivated cel wall (0). PE pretreated heat-inactivated cell wall(0). Values represent the mean ± SE of at least five preparations.

walls. An increase in the UA release mediated by PG was

observed at the turning stage. This ripening stage correspondsto the time of an initial decrease in uronide-esterification. PGtreatment of ripe fruit cell walls releases 5.3 mg UA/100 mgwall or 28% of the UA present (Fig. 5). In all cases, polymersreleased by PG have a degree of polymerization of between 2and 10 UA monomer units when subjected to fractionationon a DEAE column (data not shown) (17).

Pretreatment ofHCW with PE decreased the mol percentof methylesterified UA to about 30% at all stages of devel-opment. Subsequent treatment of the deesterified wall resi-dues with PG released 5.3 mg UA/100 mg wall from all stagesof ripening (Fig. 5). This value, 5.3 mg UA/100 mg wall,corresponds to the amount released from walls of the ripestage and also to that released upon PG treatment ofAWR.

Carbohydrate Composition of Polymers Released by PGand PE Treatment

Polymers released from HCW upon subjecting them to a

24 h treatment with PG were comprised ofGalUA, Gal, Rha,Ara, Xyl, and Glu (Table III). Monomeric components ofpolymers derived from MG and B fruit were predominantlyUA, 35-40%, and galactose, 27 to 33%. Polymers releasedfrom AWR also contained 40% UA but contained more

galactose (41%). As the fruit developed from the turning tothe ripe stage, there was a significant increase in the releaseof GalUA-containing and Rha-containing polymers coupledwith a decrease in Ara and Gal components. Fragmentsreleased by PG following PE pretreatment did not differ inneutral sugar composition from fragments released by PGtreatment alone (data not shown).

DISCUSSION

The cell walls of plants are dynamic in nature. A change inone constituent may lead to an altered physiological functionofa specific organ or tissue. It is essential therefore, to preservethe integrity ofthe cell wall during isolation to ensure minimal

Table Ill. Carbohydrate Composition of Cell Wall PolymersCarbohydrates were released by a 24 h polygalacturonase treat-

ment (5 zg PG at 220C). The developmental stages of fruit fromwhich walls are prepared is described in the text. No significantquantitative differences in polymer composition were observed be-tween PG treated walls and those walls pretreated with PE prior toPG treatment. The total refers to the quantity of UA released by PGtreatment of HCW (see Fig. 5). Values represent the mean of at leastthree separate digests.

Wall Source GaIUA Rha Gal Ara Total UA releasedmol % mg UA released/100 mg wall

AWR 40 6 41 11 5.31MG 42 6 31 10 2.70B 36 7 33 12 2.42T 42 10 28 10 3.83P 50 12 19 8 4.23R 54 15 1 1 6 5.33

a An additional 4 to 7% of each fraction was contributed by xyloseand glucose.

820 KOCH AND NEVINS

A ^




modifications of structural constituents. We have comparedthe pectic components from the cell wall of tomato fruitpericarp tissue, extracted using various procedures, to deter-mine the consequences of methodology on the total pectincontent and the degree of esterification. The purpose was todetermine the most valid wall isolation method for subsequentenzyme treatments. Purified tomato fruit PG and PE differ-entially degrade wall substrates isolated from fruit at selecteddevelopmental stages. To resolve the interaction this studyaddresses two additional objectives: (a) a partial identificationof the native substrate in the tomato fruit and (b) an assess-

ment of the degree of esterification of fruit pectins as a

function of developmental stage.The tomato fruit system offers a unique model for studying

pectin structure since tomato fruit are high in UA, 20% ofthe cell wall (Fig. 4), and have developmentally dependentchanges in their methylesterification, 90 mol % methylester-ification in mature green fruit tissue, and 30% in red ripetissue (Fig. 4). Preparation of pectin-rich cell walls withoutenzymatic activity often involves procedures that may modifycarbohydrate structure such as heating (14) or inactivationwith phenol:acetic acid:water (26). Tissues heat-inactivated in50% ethanol (HCW) have been used to identify starch content(15) and as an enzymatic control for autolytic reactions (25).Chloroform-methanol extracts have been extensively used toprepare pectin rich walls (1, 1 1) and for the isolation of cellwall polysaccharides for structural determination (12, 22).

Gross and Wallner (1 1) demonstrated that wall polymers,extracted in buffer from ripe tomato fruit cell walls, were

similar to a fraction in mature green fruit that was degradedby wall hydrolases extracted from ripe fruit. Furthermore,they detected no change in the water soluble fraction of rinfruit, a mutant which lacks significant amounts of PG. Gross(10) subsequently examined the composition of fruit cell wallsby fractionation using conventional procedures. The mostsignificant change noted in the wall composition during rip-ening was a decrease in the UA content of the covalentlybound pectin fraction. Huber (13) investigated the mol wtdistribution of EDTA-soluble pectin fractions and found a

decrease in size during development. These studies have ledto the proposal that the increase in pectin solubilization or

the change in pectin size is the consequence of the action ofPG in situ. Our results suggest that the preparation proceduresused might have allowed PG to act on the cell wall in vitro.Any conclusion that PG was active in situ prior to cell wall

extraction might be in error.Modifications of pectins may occur during the extraction

of cell walls. These modifications are a result of autolyticreactions primarily by PG and PE in extraction steps involvingaqueous phases. Although PE activity appears to decreaseafter dehydration, substantial PG may remain active. Extrac-tion ofthe cell wall in 50% ethanol inactivates PG and appearsto protect the pectin from ,B-elimination reactions duringsubsequent heat-inactivation. No soluble products of UA are

observed during incubation after heat-inactivation. ,B-Elimi-nation reactions are normally observed when methylpectinsare heated or incubated in neutral solutions (2, 16). Perhapsthe mode of inhibition of PG mediated autolysis by ethanolresults from a direct association of ethanol with the pectin

molecule. By replacing or displacing water molecules, enzy-matic activity is restricted by limiting accessibility of theenzyme. The displacement of water might also account for areduction of p-elimination reactions by limiting interactionsof water with methylpectins. The HCW extraction procedureappears superior to other extractions examined since it pro-vides optimum conditions for the preservation of cell wallstructure by inactivating constitutive enzymes that may mod-ify pectin structure.Tomato fruit cell wall uronides are highly esterified early

in development but become progressively deesterified duringripening. In vivo assays generally show PE activity is relativelyabundant and constant at all stages of ripening (Fig. 4).Autolysis results in deesterification to a residual 30% in AWRby endogenous PE within 10 min after ethanol inhibition ofenzyme activity is alleviated. This residual amount of esteri-fication may be due to enzyme specificity, restriction ofaccessibility of the enzyme by hydrophobic interactions be-tween pectin molecules due to methylesters, or due to Ca2"chelation of the pectin. The rate of in situ methyl-deesterifi-cation of wall uronides, observed by differences in the degreeof esterification of walls extracted at specific ripening stages,occurs at a much slower rate (4 d for equivalent dessertifica-tion of the wall pectins from the breaker to pink stages) thanin vitro PE activity. Hence, the rapid in vitro PE actionsuggests endogenous regulation of PE. Regulation may be theresult of allosteric interactions, compartmentalization of PEwithin the cell or by physical restriction of enzyme mobilitywithin the cell wall matrix. Susceptibility of the tomato fruitcell wall to PG certainly corresponds to a decrease in methy-lesterification of wall uronides. A relationship between theextent of uronide methylesterification and PG-mediated de-polymerization has been previously observed by Pressey andAvants (24). They demonstrated that cell wall protein frac-tions enriched in PE served to facilitate deesterification inwalls of MG fruit. The decrease in the esterification of MGfruit increased the susceptibility of the wall pectins to hydrol-ysis by PG-rich extract. The increase in UA release by PGafter treatment of cell walls with PE is in agreement with theresults reported herein since PG treatment of MG cell wallsallowed to deesterify by undergoing autolysis or the pretreat-ment of walls with PE releases significantly greater amountsofUA than PG hydrolysis of esterified walls.When HCW preparations derived from different ripening

stages are pretreated with PE, equivalent amounts of uronidesare released by subsequent PG treatment. This result supportsthe contention that there is relatively little change in thepectin content of tomato cell walls throughout development.Although PG is present in the cell wall just prior to breakerstage (6), the enzyme would appear to be limited in its capacityto release uronide components at this stage, presumably dueto the high level of substrate esterification.

Polysaccharide fragments containing GalA and Rha re-leased by PG may be similar to a rhamnogalacturonan derivedfrom sycamore suspension cultures (19, 27). The most ob-vious difference in neutral sugar composition between thesolubilized tomato polymers and sycamore RGI is the pres-ence of minor quantities of Glc and Xyl polymers in tomato(4-7% each). The presence of Xyl and Glc residues in PG

821



Plant Physiol. Vol. 91,1989

digests of MG fruit cell walls has been previously reported(24), but the physiological significance of the release of thesecomponents remains uncertain.The most significant change in the neutral sugar content of

PG released polymers is the decrease in Gal as the fruit ripens.It has been proposed that the decrease in Gal during devel-opment may be due to a decline in the incorporation of Galduring the onset of senescence (18). This hypothesis is con-sistent with the current results provided PG activity is presentin situ. PG would instigate the turnover of RG I, whilecontinued synthesis of Gal-poor polyuronides would serve tomaintain the total uronide level in the wall during ripening.The establishment of the presence of in situ PG activity ispresented in a subsequent paper in this series. A secondinterpretation for the change in Gal may be the presence ofgalactanase or galactosidase in the cell wall not detected inprotein extracts employed in this study.

In conclusion, direct modification of the tomato cell wallpolygalacturonan appears essential for its effective degrada-tion by PG in vitro. One way to achieve this modificationwould be to alter the degree of methylesterification of thesubstrate. Deesterification ofthe polyuronide would favor PGactivity. Certain differences in the composition of polymersreleased by PG as a function of developmental stage indicatethat the polygalacturonan substrate also undergoes modifica-tions, either by selective hydrolysis or polyuronide turnover.The primary structure of the polymer must ultimately beresolved to fully evaluate the significance of these events.

ACKNOWLEDGMENTS

The authors wish to express their gratitude to Drs. John M.Labavich and L. Carl Greve for helpful discussions and scientificsupport.

LITERATURE CITED

1. Ahmed A, LabavitchJM (1977) A simplified method for accuratedetermination of cell wall uronide content. J Food Biochem 1:361-365

2. Albersheim P, Neukom H, Deuel H (1960) Splitting of pectinchain molecules in neutral solutions. Arch Biochem Biophys90: 46-51

3. Ali ZM, Brady CJ (1982) Purification and characterization ofthe polygalacturonases of tomato fruits. Aust J Plant Physiol9: 155-169

4. Blumenkrantz N, Asboe-Hansen G (1973) New method for quan-titative determination of uronic acids. Anal Biochem 54: 484-489

5. Color classification requirements in tomatoes (1975) UnitedStates Department of Agriculture Visual Aid TM-L-1. TheJohn Henry Co., Lansing, MI

6. DellaPenna D, Alexander DC, Bennett AB (1986) Molecularcloning of tomato fruit polygalacturonase: analysis of polyga-lacturonase mRNA levels during ripening. Proc Natl Acad SciUSA 83: 6420-6424

7. Fry SC (1983) Feruloylated pectins from the primary cell wall:their structure and possible functions. Planta 157: 111-123

8. Grant GT, Morris ER, Rees DA, Smith PJC, Thom D (1973)Biological interactions between polysaccharides and divalentcations: the egg-box model. FEBS Lett 32: 195-198

9. Grierson D, Tucker GA (1983) Timing of ethylene and polyga-

lacturonase synthesis in relation to the control of tomato fruitripening. Planta 157: 174-179

10. Gross KC (1984) Fractionation and partial characterization ofcell walls from normal and non-ripening mutant tomato fruit.Physiol Plant 62: 25-32

11. Gross KC, Wallner SJ (1979) Degradation of cell wall polysac-charides during tomato fruit ripening. Plant Physiol 63: 117-120

12. Hatfield R, Nevins DJ (1986) Characterization of the hydrolyticactivity of avocado cellulase. Plant Cell Physiol 27: 541-552

13. Huber DJ (1983) Polyuronide degradation and hemicellulosemodifications in ripening tomato fruit. J Am Soc Hortic Sci108:405-409

14. Huber DJ, Nevins DJ (1979) Autolysis of the cell wall ,B-D-glucan in corn coleoptiles. Plant Cell Physiol 20: 201-212

15. Huber SC, Israel DW (1982) Biochemical basis for partitioningof photosynthetically fixed carbon between starch and sucrosein soybean (Glycine max Merr.) leaves. Plant Physiol 69: 691 -696

16. Kiss J (1974) ,B-Eliminative degradation of carbohydrates con-taining uronic acid residues. Adv Carbohydr Chem 29: 229-303

17. Koch JL (1988) Characterization of tomato fruit polyuronides.Changes in their susceptibility to polygalacturonase duringripening. PhD dissertation, University of California, Davis, pp1-154

18. Lackey GD, Gross KC, Wallner SJ (1980) Loss of cell wallgalactan may involve reduced rate of synthesis. Plant Physiol66: 532-533

19. Lau JM, McNeil M, DArvill AG, Albersheim P (1985) Structureof the backbone ofrhamnogalacturonan I, a pectic polysaccha-ride in the primary cell walls of plants. Carbohydr Res 137:111-125

20. Lisker S, Retig N (1974) Detection of polygalacturonase andpectin lyase isoenzymes in polyacrylamide gels. J Chromatogr96: 245-249

21. Markovic O, Machova E, Slezarik A (1983) The action oftomatoand Aspergillus foetidus pectinesterases on oligomeric sub-strates esterified with diazomethane. Carbohydr Res 116: 105-111

22. Morrison JC, Greve LC, Labavitch JM (1987) The role of cellwall-degrading enzymes in the formation of gum ducts inalmond fruit. J Am Soc Hortic Sci 112: 367-372

23. Pressey R, Avants JK (1972) Multiple forms of pectinesterase intomatoes. Phytochemistry 11: 3139-3142

24. Pressey R, Avants JK (1982) Solubilization of cell walls bytomato polygalacturonases: effects of pectinesterases. J FoodBiochem 6: 57-74

25. Rushing JW, Huber DJ (1984) In vitro characterization of to-mato fruit softening. The use ofenzymatically active cell walls.Plant Physiol 75: 891-894

26. Selvendran RR (1975) Analysis of cell wall material from planttissues: extraction and purification. Phytochemistry 14: 1011-1017

27. Talmadge KW, Keegstra K, Bayer WD, Albersheim P (1973)The structure of plant cell walls. I. The macromolecular com-ponents ofthe walls ofsuspension-cultured sycamore cells witha detailed analysis of the pectic polysaccharides. Plant Physiol51: 158-173

28. Themmen APN, Tucker GA, Grierson D (1982) Degradation ofisolated tomato cell walls by purified polygalacturonase in vitro.Plant Physiol 69: 122-124

29. Tigchelaar EC, McGlasson WB, Buescher RW (1978) Geneticregulation of tomato fruit ripening. HortScience 13: 508-513

30. Wallner SJ, Walker JE (1975) Glycosidases in cell wall-degrad-ing extracts of ripening tomato fruits. Plant Physiol 55: 94-98

31. Wood PJ, Siddiqui IR (1971) Determination of methanol and itsapplication to measurement of pectin ester content and pectinmethylesterase activity. Anal Biochem 39: 418-428

822 KOCH AND NEVINS



tomato fruit cell - plant physiology. solubilization appearsto bea consequenceof autohydro-lysis...

Documents