Environmental stresses affect tomato microsomal membrane function differently than natural ripening and senescence

Environmental stresses affect tomato microsomal membrane function differently than natural ripening and senescence

Postharvest Biologyand Technology Postharvest Biology and Technology 6 (1995) 257-273 ELSEVIER Environmental stresses affect tomato microsomal membr...

1MB Sizes 1 Downloads 17 Views

Postharvest Biologyand Technology Postharvest Biology and Technology 6 (1995) 257-273

ELSEVIER

Environmental stresses affect tomato microsomal membrane function differently than natural ripening and senescence T Palma, A.G. Marangoni, D.W. Stanley * Department of Food Science, University of Guelph, Guelph, Ont. NlG 2W1, Canada

Accepted 21 December 1994

Abstract

On the premise that environmental stresses may induce changes in microsomal membrane function characteristic of accelerated senescence, the effects of physical damage (PD), chilling injury (CI), heat shock (HS) and controlled atmosphere (CA) on tomato fruit (Lycupersicon esculentum cv. Trust) were compared to changes occurring during normal ripening and senescence. A physiological pattern for ripening and senescence of tomato fruit was established based on weight loss patterns, colour development, ion leakage, lipid fatty acid profiles of microsomal membrane lipids, microsomal K+ stimulated ATPase activity and electrophoretically separated protein patterns of microsomal membranes. Control fruit displayed increases in redness, weight loss, ion leakage and saturation index (SI) of membrane lipids, as well as the appearance of membrane-associated polygalacturonase (polygalacturonase isozyme II - PGII) and its B-subunit and decreased microsomal membrane K+ stimulated ATPase activity during a three-week storage period. Relative to controls, PD and CI fruit displayed increased ion leakage and a decreased rate of red colour development, increased ATPase activity and SI. Membrane-associated PGII appeared sooner in PD fruit, but its appearance was delayed in CA and CI fruit, relative to controls. Tomatoes exposed to HS and CA treatments, however, displayed ripening and senescence physiological symptoms similar to control fruit. Thus, environmental stresses induced membrane changes that were expressed differently than seen during senescence. Keywords:

Chilling injury; Heat shock; Controlled atmosphere; Polygalacturonase

1. Introduction

Experimental evidence has shown that senescence of various fruits, vegetables and flowers is associated with the degradation of biological membranes, and that * Corresponding author. Fax: 519 824-6631. 09255214/95/$09.50 0 1995 Elsevier SSDIO925-5214(94)00058-l

Science

B.V. All rights reserved.

258

7: Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273

loss of membrane integrity is an early and fundamental feature of senescence (reviewed by Thompson, 1988; Stanley, 1991; Leshem et al., 1992; Paliyath and Droillard, 1992). Membrane deteriorative processes include elevated phase transition temperatures of bulk membrane lipids and decreased bulk membrane lipid fluidity (Thompson, 1988; Marangoni et al., 1990) hydrolysis of membrane phospholipids to constituent free fatty acids, and peroxidation of constituent polyunsaturated fatty acids with a corresponding production of free radicals (Thompson, 1988; Paliyath and Droillard, 1992). These free radicals catalyze further deteriorative reactions that lead to cellular death and a decline in fruit, vegetable and flower quality. Plant materials are exposed to a variety of environmental stresses during growth and after harvest. Stresses such as chilling injury (CI) lead to membrane degradation resembling accelerated senescence via biochemical mechanisms similar to those described above (Marangoni et al., 1989; Todd et al., 1992). Moreover, resistance to CI has been shown in tomato fruit to be related to the physical properties of biological membranes (Marangoni and Stanley, 1989; Marangoni et al., 1989) and could be induced by acclimating the fruit to low temperatures by gradual exposure to decreasing temperatures (Marangoni et al., 1990). Marangoni et al. (1990) also observed that the endoplasmic reticulum of chilled tomato fruit lost its integrity in a similar fashion to the lamellar unstacking observed during heat shock (Brodl, 1989). It is tempting, therefore, to propose that losses of fruit, vegetable and flower quality due to environmental stresses are the result of membrane degenerative processes characteristic of an accelerated senescence. It has previously been suggested that environmental stress of various types is able to induce or accelerate many physiological changes in the plant cell that resemble the senescence syndrome (Nooden, ,1988; Parkin et al., 1989). With this hypothesis in mind, the objective of this research was to characterize physiological and biochemical changes that occur in the microsomal membranes of tomatoes during natural ripening and senescence, and compare them to the physiological and biochemical changes that occur in tomatoes exposed to various environmental stresses. 2. Material and methods 2.1. Plant material

Greenhouse tomato plants were grown at ~20°C locally and fruit at the “breaker” stage picked during the summer and fall. They were sterilized in dilute sodium hypochlorite, stored overnight at 20°C and 95% relative humidity (RH) and used the following day (day 0). 2.2. Storage conditions

Time-stress combinations were designed to confer moderate but observable (by sensory evaluation, data not shown) injury response within two weeks of treatment.

T. Palma et al. I Postharvest Biology and Echology

6 (1995) 257-273

259

Half of the harvested fruit were randomly designated as controls and ripened at 20°C and 95% RH. Weight loss, redness, ion leakage determinations and membrane extractions were performed. 2.3. Treatments

Four stress treatments including physical damage (PD), heat shock (HS), controlled atmosphere (CA) storage and CI were applied. PD

Loosely packed fruit were mounted on a shaking device set at two cycles per second with a displacement of 2.5 cm. Tomatoes were shaken for 8 h, with the orientation of the box changed by 90” hourly, and then transferred to control conditions (20°C and 95% RH). Roth treatment and unshaken controls were analyzed at 0, 7 and 14 days. HS

Tomatoes were held at 45°C for 8 h at 95% RH, and then moved to control conditions. Both heat treated and unheated controls were analyzed at 0, 7 and 14 days. CA

Tomatoes were stored in 4-l desiccators (20°C) with a continuous flow (ca. 20 ml min’) of a 4% 02 + 96% Nz mixture. The gas mixture was bubbled through distilled water in order to maintain high RH. After 7 days, fruit were removed to control conditions. Both CA treated and control fruit were analyzed at 0, 7, 14 and 21 days.

cz Tomatoes were chilled at 4°C and 95% RH for 14 days and then transferred to control conditions for 14 days. Both CI treated and control fruit were analyzed at 0, 7, 14 and 21 days; CI treated fruit were also analyzed at 28 days. 2.4. Analyses

For each set of analyses at each weekly sampling date ten fruit were selected at random from the control and treatment storages. These were weighed, halved axially, and a composite from one set of samples (ten halves) was homogenized as detailed below. Part of this homogenate was used for a redness determination; that remaining was the starting point for the membrane isolation procedure. Five determinations of membrane fatty acid profiles and ATPase activity were performed on this composite membrane isolate. From the other set of ten halves, 36 tissue discs were prepared that were then used for ion leakage determinations (six sets of six discs).

260

T. Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273

Weight loss

Ten individual fruit were weighed the day of harvest and weekly thereafter weight loss was expressed as a percentage of the initial weight (day 0).

and

Redness determination

The composite sample of ten tomato halves was rinsed and the pericarp homogenized in a Waring blendor at 4°C with chilled buffer [l : 1, w/v, fruit : buffer, 0.1 M MOPS, pH 7.2,0.33 M sorbitol, 30 mM ascorbic acid, 5 mM EDTA, 5 mM EGTA, 3 mM sodium metabisulphite, 1 mM dithiothreitol (DTT)] and 0.5% (w/w) insoluble polyvinylpyrrolidone (PVP). The homogenate was filtered through three layers of cheesecloth and a 300-ml aliquot of filtrate was used for colour determination. Redness was determined with an Agtron Process Analyzer Model M-35-D (Agtron Inc., Sparks, Nev.) following the manufacturer’s instructions. Ion leakage

A set of 36 pericarp discs of 1.5 cm diameter and ca. 0.5 cm thickness, trimmed of locular tissue, was obtained with a cork borer for each treatment-time combination. Six replicates of six discs each were placed in tubes with 25 ml of 0.4 M mannitol solution. Following 5 h at 24”C, conductivity was measured with a Conductivity Meter (Model CDM3, Radiometer, Copenhagen). Tubes were left in a -18°C freezer for at least 16 h, and after thawing the conductivity was remeasured. This value was taken as 100% leakage. Ion leakage values were expressed as a percentage of the conductivity after the freeze-thaw cycle (Kuo and Parkin, 1989). Preparation of microsomal membranes

Membranes were prepared using the procedure of Marangoni and Stanley (1989). For each treatment, a 360-ml aliquot of the filtered tissue extract from redness determination was centrifuged at 4°C for 20 min at 10,000 g and the resulting supernatant recentrifuged for 90 min at 70,358 g. The pellet was manually resuspended in 2.0 ml of pH 7.2 buffer (0.1 M MOPS and 0.33 M sorbitol) at 4°C and stored at -30°C until further use. This sample was referred to as pericarp microsomal membranes or PMM. Enzyme assays

Samples of frozen PMM were warmed to 20°C for all enzyme assays and diluted with pH 7.2 MOPS-sorbitol buffer to the desired protein concentration. Total protein content was determined using the Bio-Rad DC Protein Assay (Bio-Rad, Mississauga, Ont.) following the supplier’s microplate assay protocol. Potassiumstimulated and nitrate-sensitive ATPases as well as Triton-X-lOO-stimulated UDPase activity measurements were as described in Briskin et al. (1987), except that the phosphate determination was adopted from Chifflet et al. (1988). The cytochrome C oxidase assay was performed as described by Yoshida (1979). Three replicate determinations on each treatment-time combination membrane preparation were performed except for K+-stimulated ATPase where five replicates were measured.

T. Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273

261

N-terminal sequencing Gels [7.5% sodium-dodecyl

sulfate (SDS)-polyacrylamide] containing the protein were electroblotted with a Mini Tram Blot (electrophoretic transfer cell; Bio-Rad) following all manufacturer’s protocols, onto Immobilon-P transfer membranes (Millipore). Sequencing was carried out by Edman degradation (Matsudaira, 1990) with a Milligen/Biosearch 6600 Pro Sequencer Protein Sequencing System (Millipore) using a SequeTeg column (Millipore). Crosslinking to Immobilon-P was performed using the SequeNet attachment process (Pappin et al., 1990). Electrophoresis

SDS-PAGE (polyacrylamide gel electrophoresis) protein separations were performed in a Mini-Protean II Dual Slab Cell (Bio-Rad) using the method of Laemmli (1970) following all manufacturer’s protocols. Each lane was loaded with 10 pg of membrane protein. Gels (12% SDS polyacrylamide) were stained with silver according to Merril (1990). Molecular weight markers consisted of commercial low molecular weight SDS-PAGE standards (Bio-Rad). Lipid analysis

The lipids from a 300-500-4 sample of PMM from each treatment-time combination were extracted following the procedure of Higgins (1987). Methylation of free fatty acids was carried out as directed using methanolic base reagent (Supelco, Bellefonte, Pa.). Fatty acid composition was determined by gas chromatography (Marangoni et al., 1990) based on three determinations for samples that had been concentrated 15x under a stream of 02-free N2. Saturation index (SI) was calculated according to the following formula: % saturated % monoenes + 2x % dienes + 3x % trienes where % saturated = C saturated fatty acids; % monoenes = C unsaturated fatty acids with one double bond; % dienes = C fatty acids with two double bonds; and % trienes = C fatty acids with three double bonds. SI =

3. Results and discussion 3.1. Normal ripening and senescence

Weight loss (Fig. 1) and red colour (Fig. 2) increased during ripening and senescence of control tomatoes. Initial redness values varied among the controls used for the different treatments as was expected from fruits grown under different light conditions (i.e., different times of the year): the degree of lighting at the beginning of ripening influences subsequent colouration (Varga and Bruinsma, 1986). Weight loss was attributed to loss of water resulting from transpiration. As observed in many fruits (Wilson and McMurdo, 1981), ion leakage of control fruit gradually increased during ripening and senescence (Fig. 3). Since ion leakage is a measurement of overall cellular membrane integrity (Wilson and McMurdo, 1981) this is evidence of membrane deterioration.

T Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273

262

‘ii i%n

12

12

10

10

6

6

6

6

4

4

2

2

_

0 COMROL

3

W

iij

12

3

10

TREATMEN?

TREATMENT

ATMOSPHERE

CMILLING

T

10

6

6

6

6

4

4

2

2

_

n

Ti

0 TREATMENT

n Day

INJURY

12

0, w Week I,

Week 2,

q

cxxma

TREATMENT

Week 3 and 13 Week 4.

Fig. 1. Weight loss of tomato fruit as a function of storage period during normal ripening senescence (control), and following applied environmental stresses (treatment). Bars represent average of ten whole tomato fruit and their corresponding standard deviations. Thick bar indicates 95% confidence interval for all data.

and the the

Microsomal membranes were characterized by marker enzyme assays. Potassiumstimulated ATPase (plasma membrane marker) was purified 3.3-fold over the initial extract, nitrite-sensitive ATPase (tonoplast marker) was purified 4.5fold over the initial extract, Triton-X-lOO-stimulated UDPase (Golgi marker) was purified 4.4fold over the initial extract while cytochrome C oxidase (mitochondrial marker) was not present at all in the microsomal membrane preparation. The composition of PMM changed during ripening and senescence. Gas-liquid chromatographic analysis of the membrane lipids (Table 1) revealed a general loss in polyunsaturated fatty acids and a concomitant increase in SI throughout the storage period. For all four treatment controls, losses in linolenic acid (18 : 3) were most predominant, while linoleic acid (18 : 2) losses were less clearly defined. Lipid peroxidation has been postulated as a major membrane deteriorative process during senescence (Thompson, 1988). The changes observed, particularly losses in linolenic acid, are indicative of membrane peroxidation and, since ion leakage also increased, are in agreement with the concept of membrane degradation due to senescence. SDS electrophoresis of PMM showed that after 7 to 14 days at 20°C a doublet

7: Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273 PHYSICAL

DAMAGE

263

HEAT SHCJX a 7

6

6

5

5

Q

4

4

.z 3

3

3

6

2

2

5 u

1

1 0

T10

5

TREATMENT

TREATMENT

CHILLING

-ATMOSPHERE

8

a-l

8

d

7

7

0

6

6

B

5

5

4

4

3

3

2

2

IT

1

1

0

0

CONTROL

TREATMEM

w

Day

0.

q Week

1,

Week

2.

q

Week

3 and 17 Week

INJURY

TREATMENT

4

Fig. 2. Redness (converted Agtron units) of homogenized tomato fruit pericarp filtrate as a function of storage period during normal ripening and senescence (control), and following applied environmental stresses (treatment). Bars represent the average of a composite sample of ten whole tomatoes.

appeared at about 45 kD (Fig. 4). These bands were identified as the endopolygalacturonase monomer (PGII) and its /?-subunit by N-terminal sequencing of the protein separated and extracted from the SDS-PAGE gel (data not shown). PG is an enzyme associated with the softening process of many fruits, including, to a limited extent, tomatoes. The N-terminal sequence corresponded closely with that reported for mature PGII and its /?-subunit by other researchers (Grierson et al., 1986; Sheehy et al., 1987; Zheng et al., 1992). Membrane-associated PGII may represent endoplasmic reticulum or Golgiassociated PGII being synthesized and in transit to the cell wall. Biggs and Handa (1989) previously reported the localization of PG mRNA on membrane-bound polyribosomes or ripening tomato pericarp. It is reasonable that the PG translation products, PGII and the #?-subunit, relatively hydrophobic polypeptides, would also be membrane-associated. The activity of K-+-stimulated ATPase decreased during ripening and senescence (Fig. 5). This has also been observed in flowers as a feature of senescence (Borochov et al., 1986), and has been postulated to be due to loss of membrane functionality (Palta, 1990).

264

I: Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273 PHYSICAL

HEATSHCCK

DAMAGE 100,

loo-

90 -

so 80 -

coNrm_

% iii 0

TREATMEM

TREATMENT

CHILLING

CXMRXEGATMOGPHERE 100

100

90

90

50

80

70

70

80

60

50

50

40

40

30

30

20

20

10

10

0

0

n Day 0, q Week

1,

Week

2. •i

Week 3 and 17 Week

lNJU!=lY

4.

Fig. 3. Ion leakage (% of total tissue conductivity) as a function of storage period during normal ripening and senescence (control), and following applied environmental stresses (treatment). Bars represent the average of six replicates from a composite sample of ten tomatoes and their corresponding standard deviations. Thick bar indicates the 95% confidence interval for all data.

From these results, a pattern of normal ripening and senescence was established for tomato fruits stored at 20°C and 95% RH using data from red colour development, weight loss, ion leakage, ATPase activity, SI of membrane fatty acids, and the appearance of PG. Although some seasonal differences were observed in ripening patterns for greenhouse-grown tomatoes harvested at different times of the year (i.e., control fruit for each of the four treatments), trends were similar in all fruit for these parameters. Weight loss, red colour and ion leakage increased with ripening and senescence and microsomal membranes exhibited losses of polyunsaturated fatty acids and decreased ATPase activity. A protein associated with microsomal membranes and attributed to PG also appeared during ripening and senescence. These measurable effects constituted the pattern of senescence used for assessment of stress-induced effects. 3.2. Treatment effects PD tomatoes lost more weight than their controls, while with HS fruit no significant difference was found (Fig. 1). However, for CA tomatoes, fruit weight

T Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273

265

Table 1 Fatty acid composition (weight %) of tomato microsomal membrane lipids Acid

Control

Treatment Week 1

Week 2

Physical damage c12:o 0.09 c14:o 0.54 Cl6:O 46.76 C18:O 2.85 C18: 1 1.38 C18:2 31.34 C18:3 16.78 c2o:o 0.27 Sl 0.44

0.53 0.98 53.34 3.93 3.52 27.96 9.73 0.00 0.66

0.00 0.00 50.14 4.09 7.50 31.90 6.36 0.00 0.60

Heat shock c12:o 6.16 c14:o 5.01 C16:O 35.98 C18:O 6.40 C18:l 0.73 C18:2 33.18 C18:3 10.51 c2o:o 1 42 Sl 0.56

12.98 8.46 27.31 10.06 1.90 29.00 6.47 3.83 0.79

14.31 10.32 25.38 10.99 0.89 27.39 5.70 5.02 0.91

Chilling injury c12:o 1.35 c14:o 4.55 C16:O 45.90 C18:O 2.99 C18:l 0.00 C18:2 25.46 C18:3 13.45 c2o:o 0.29 SI 0.70

6.54 3.39 SO.85 10.93 4.51 18.06 4.20 1.51 1.38

4.13 3.19 46.60 7.07 7.73 24.45 3.14 3.08 0.98

Controlled atmosphere c12:o 8.59 10.41 c14:o 6.62 8.82 C16:O 32.63 29.92 C18:O 5.45 6.89 C18:l 1.33 2.24 C18:2 30.76 28.40 C18:3 9.39 6.47 c2o:o 5.23 6.84 SI 0.64 0.80

12.30 9.43 29.33 6.77 4.94 26.51 3.34 7.38 0.96

Day 0

Week 3

Week 4

2.07 2.50 55.13 3.73 5.00 21.69 7.99 1.87 0.90

2.36 3.50 51.36 8.47 8.77 19.75 3.73 2.07 1.14

i .72 3.12 57.78 16.86 9.48 7.72 3.31 0.00 2.21

5.96 3.71 42.14 5.73 2.44 30.88 8.24 0.91 0.66

6.73 4.10 45.24 4.54 4.13 29.65 4.65 0.97 0.80

Week 1

Week 2

0.54 46.16 2.85 1.38 31.34 16.78 0.27 0.44

0.00 0.00 48.99 4.99 3.40 32.87 9.75 0.00 0.54

0.00 0.99 56.83 4.17 3.41 28.28 5.72 0.00 0.81

6.76 5.01 35.98 6.40 0.73 33.18 10.51 1.42 0.56

9.39 7.17 30.09 9.37 3.86 31.78 5.64 2.71 0.70

16.33 10.95 23.36 11.48 2.11 27.20 4.22 4.37 0.96

2.21 2.34 54.56 12.79 8.98 17.07 1.41 0.64 1.53

1.35 4.55 45.90 2.99 0.00 25.46 13.45 0.29 0.70

1.37 1.16 58.34 13.33 9.21 12.77 3.19 0.64 1.69

11.23 8.80 28.37 6.88 4.33 30.52 2.39 1.41 0.86

8.59 6.62 32.63 5.45 1.33 30.76 9.39 5.23 0.64

9.12 6.71 32.35 5.80 0.46 30.67 8.78 6.11 0.68

Week 3

Day 0

0.09

changed only slightly while in storage, and even after removal to air (2o”C, 95% RH). The weight loss of tomato fruit was also slowed by chilling; however, when CI fruit were transferred to 20°C weight loss increased markedly and became comparable to that of control fruit.

266

T Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273

Physical c

Heat

Damage

C

t +

Shock _

t

*

97

-66

Controlled c

Chilling

Atmosphere t

Injury

14

01230123s

0

1231

Fig. 4. Electrophoretic separation patterns of tomato fruit pericarp tissue proteins as a function of storage period during normal ripening and senescence applied environmental stresses (treatment). Left lanes are controls, right lanes lane (S) corresponds to molecular weight markers. 0 = day 0, I = week 1, 2 4 = week 4.

234s microsomal membrane (control), and following are treatments, far right = week 2, 3 = week 3,

Red colour development was somewhat delayed in PD relative to control fruit (Fig. 2). HS fruit developed red colour at a slightly slower rate and to a somewhat lesser extent than their control, a possible consequence of the inhibition of lycopene formation at temperatures above 32°C (Varga and Bruinsma, 1986). Fruit remained green during CA storage, but when removed to air these samples developed red colour quickly and reached comparable levels to the control. In contrast, CI fruit developed colour only slowly after their removal from the chilling stress. This is a characteristic CI symptom of tomatoes (Jackman et al., 1992). Ion leakage of PD fruit was similar to that of controls during the first week of storage, but increased noticeably during the second week (Fig. 3). This contrasted with HS fruit in which ion leakage was comparable to the control through the second week of storage at 20°C. Ion leakage decreased during CA storage; levels remained below the control during the subsequent ambient ripening period. The

l? Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273 PHYSICAL

267

HEAT StUXK

DAMAGE 8

I

7

a

cT3NTRu

TREATMENT

TREATMENT

4 E 3 b

81



7

5

6

6

i5

is

C0NlROLLED

CHILLING

ATMOSPHERE

am

5

5

4

4

2

3

3

2

2

1

1

0

0 TREATMENT

TREATMENT

n Day

INJURt

31

0.

q Week

1,

Week

2,

q Week

3 and

q Week

4.

Fig. 5. K+-stimulated ATPase activity as a function of storage period during normal ripening and senescence (control), and following applied environmental stresses (treatment). Bars represent the average of a combined microsomal membrane preparation derived from ten tomatoes and the corresponding standard deviation for five determination of this sample. Thick bar indicates the 95% confidence interval for all data.

level of ion leakage exhibited by CI fruit did not differ from the controls during chilling and was maintained through the first week of subsequent storage at 20°C. During the second week at 20°C ion leakage increased markedly in CI fruit relative to controls. Membrane lipid SI increased similarly for PD and HS fruit relative to their respective controls during storage (Table 1). In contrast, SI values were constant during CA storage and increased only slightly upon removal from CA storage. Chilling treatment led to increased SI during the first week of storage, followed by a decrease to the initial value in the second week. Decreased SI has also been observed by several authors during acclimation of fruits to cold storage (Christiansen, 1984; Goto et al., 1984; Marangoni et al., 1990). The SI rose slightly in the first week of ripening once the chilling stress was removed, and doubled during the second. A membrane-associated protein identified as PG in control fruit was also observed electrophoretically in microsomal membranes derived from fruit exposed

268

T. Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273

to all four stresses (Fig. 4) but at differing times during storage. In PD fruit, PG appeared one week before it did in the controls; whereas in CA and CI fruit PG appearance was delayed until the removal of fruits to control conditions. In HS fruit, membrane-associated PG appeared at the same time as in controls. There are several lines of evidence suggesting that PG mRNA expression requires the induction of a ethylene-inducible factor or factors (Grierson and Tucker, 1983; Maunders et al., 1987; Oeller et al., 1991). High temperatures have been shown to inhibit ethylene biosynthesis (Biggs et al., 1988; Picton and Grierson, 1988) and PG expression (Picton and Grierson, 1988), while wounding, a response that would have been expected through the PD treatment, induces ethylene biosynthesis and, consequently, also membrane-associated PG production (Biggs, 1987). Both CI and CA delay or completely inhibit ethylene biosynthesis and the concomitant ripening of tomato fruit (Field, 1990; Forney and Lipton, 1990); induction of ripening processes and of autocatalytic ethylene production occurs only after the stress is removed and fruit are transferred to ambient temperature conditions. The appearance of PGII as a function of the various stresses is thus consistent with the effects of these stresses on ethylene biosynthesis. ATPase activities in HS and CA fruit were similar to that in their respective controls (Fig. 5). PD and CI fruits, however, exhibited generally higher activities that did not decrease after removal of the stress or during subsequent ripening compared to controls. Iswari and Palta (1989) have proposed ATPase to be the main locus of membrane functional alterations during low temperature stress. Membrane lipid changes may modulate ATPase activity during cellular response to chilling (Palta, 1990). During cold storage, ATPase activity increased (Fig. 5), possibly due to acclimation to low temperatures; relative increases in Cl8 : 2 and Cl8 : 3 were also observed (Table 1). However, after removal from chilling conditions, the changes in ATPase activity did not mirror changes in fatty acid composition or SI of the membrane fraction. Thus, even if partial acclimation occurred, it did not prevent manifestation of CI symptoms. Also, changes in the saturation of the membrane did not seem to affect the measured ATPase activity of the CI treated fruit. 3.3. General effects of stress Physical damage of tomato fruit led to greater weight loss and delayed development of red colour compared to control fruit; softening and weight loss occurred earlier, and ion leakage increased above control levels after one week of storage. Membrane associated-PG protein was apparent one week prior to the control, and the SI increased dramatically, a reflection of higher 18 : 3 and 18 : 2 fatty acid losses. ATPase activity, however, remained at high levels, in contrast to what would be expected during accelerated senescence where changes in the structure of the membrane may inhibit membrane-associated enzyme activity. Weight loss, appearance of PG and changes in membrane lipid composition were similar in HS and control tomatoes. Red colour development was somewhat attenuated and the appearance of membrane-associated PG after two weeks of ripening was at reduced levels compared to control fruit. Any damage induced

7: Palma et al. IPosthalvest Biology and Technology 6 (1995) 257-273

269

by HS appears to be unrelated to either the ripening process or to membrane functionality, but may involve some other physiological processes such as the inhibition of enzymes involved in lycopene synthesis. These results agree with former work (Picton and Grierson, 1988) in which HS tomato fruit failed to develop red colour and PG expression was inhibited. Ripening was notably delayed during CA storage and fruit failed to ripen normally after being transferred to air, as judged by decreased weight loss and red colour development. Initial membrane lipid composition was maintained during CA, and loss of unsaturated fatty acids, characteristic of ripening and senescence, was inhibited during subsequent storage under control conditions. CI fruit underwent acclimation as indicated by increases in ATPase activity, polyunsaturated fatty acid levels, particularly linoleic acid. Demandre et al. (1986) reported oleate desaturase activity in potato microsomes. Desaturation at the sn-2 position of phosphatidylcholine, from oleic acid to linoleic in the presence of NADH was observed by these authors. Wada et al. (1990) proposed that chilling tolerance may be partly due to an F12 nonspecific desaturase of 18: 1. The increases in linoleic acid levels accompanying acclimation are consistent with enzyme catalyzed desaturation of membrane lipids. This defense mechanism, however, was only partially successful in preventing damage during subsequent storage under control conditions. Injuries manifested themselves in terms of uneven fruit ripening and accelerated membrane senescence as indicated by a dramatic increase in SI and ion leakage upon removal of the chilling stress. For all chill-sensitive fruit there is a characteristic time-temperature combination that induces CI (Jackman et al., 1988). This may explain the acclimation process observed during the first week of cold storage: for this tomato variety one week at 4°C was not sufficient to induce injury as judged by sensory analysis (data not shown). Current knowledge of senescing membranes suggests that compositional alterations leading to microviscosity changes determine membrane functionality (Thompson, 1988; Stanley, 1991; Leshem et al., 1992; Paliyath and Droillard, 1992). Part of this functionality may be reflected in the activity of membrane transport proteins such as ATPases. In the present work, a significant quantitative relationship (r = -0.78, P < 0.01, 13 df) was observed between the amount of tissue ion leakage and microsomal membrane ATPase activity in control fruit. Thus, dysfunction of ATPases may be a key factor responsible for the observed increases in ion leakage in senescing tomato fruit. This effect was not observed when data from all four stress treatments were examined. No statistically significant correlations were found between either ion leakage and SI or ATPase activity and SI for control fruits. The general lipid composition of the membrane, as reflected in its degree of saturation, therefore, does not seem to play a role in the observed increase in ion leakage or decrease in ATPase activity during ripening and senescence of controls. There was, however, a significant correlation between both the amount of tissue ion leakage (r = -0.80, P -c 0.01, 13 df) and ATPase activity (r = 0.76, P < 0.01, 13 df) and the percentage of linolenic acid present in the microsomal membranes of control fruit. Linolenic acid (18: 3) is extremely susceptible to oxidation and would be degraded by oxidative enzymes such as lipoxygenases quite readily. As

270

T Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273

well, no more 18: 3 would be synthesized by desaturases, in contrast to 18: 2, thereby decreasing the usefulness of this latter fatty acid as a biochemical index of senescence. It would seem that linolenic acid content rather than the SI is a better indicator of the senescent status of the microsomal membrane fraction. Losses of linolenic acid are indicative of membrane degradation via senescent processes, that in turn are responsible for increases in ion leakage and decreases in the activity of membrane-associated enzymes such as ATPases. These effects were observed for the control fruit only and do not reflect the changes in stressed fruit. Grouped data for PD and CI, and HS and CA did not show any meaningful relationships, except for a significant negative correlation (r = -0.74, P < 0.05, 6 df) between the ATPase activity and linolenic acid content of microsomal membranes for the grouped PD and CI stresses. In both these treatments ATPase activity remained high or increased during the ripening period after stress was applied (Fig. 4). In this case the reversal of the anticipated decline in ATPase activity was accompanied by increased linolenic acid content, supporting the observation that levels of this fatty acid are useful indicators for the physiological status of microsomal membranes undergoing senescence and stress. 4. Conclusions Fig. 6 summarizes the results for all control fruit and presents the physiological pattern of normal ripening and senescence. Metabolic processes initiated by environmental stresses were found to either prevent or slow those reactions expected to accompany senescence. Metabolic reactions to stress would thus appear to be distinct from senescence. Results from this study suggest that physiological reactions of the tomato fruit to PD, HS, CA and CI stresses were unique, and while some features were similar, these stresses did not manifest themselves as accelerated senescence phenomena. Different routes of stress damage may be taken depending upon the injury suffered that may or may not resemble those of natural senescence, perhaps depending on the rate and magnitude of the applied stress. It is apparent that the effect of various stresses on fruit physiology must be more thoroughly investigated. Macroscopic characteristics may not be a satisfactory indicator of stress or accelerated senescence. Where does aging and/or injury stop and where does senescence start? How is senescence triggered? Is it possible that stress exists as a continuum, from acute to chronic, and that depending on where along this spectrum experiments are conducted, different physiological responses will result? If so, then it is crucial to define which effects predominate along this continuum and where exactly senescence is triggered, independent of other physiological effects. Future work will include a more detailed examination of each type of stress, since the different stresses clearly lead to different physiological effects. Acknowledgements

This work was supported in part by the Natural Sciences and Engineering Research Council of Canada and the Ontario Ministry of Agriculture and Food.

iT Palma et al. I Postharvest Biology and Technology 6 (1995) 257-273



I

I

I

I

0

1

2

3

271

150-

. ~IONLEAKAGE ~MSRANE

ATPam ACTIVITY

~EhWFIANE 0

LINOLENIC ACID (X OF TOTAL LIPID)

,

I

I

0

1

2

RIPEYNG

3

TIYE (WEEKS)

Fig. 6. Biochemical pattern of ripening and senescence. fruit of four replicate experiments and its corresponding weight loss are smaller than the symbols.

Each value represents the mean from control standard error. Error bars for redness and

The authors gratefully acknowledge a critical review of the manuscript provided by Dr. Robert Jackman. References Biggs, MS., 1987. Regulation Thesis, Purdue University,

of polygalacturonase gene expression West Lafayette, Ind.

during

tomato

fruit ripening.

Ph.D.

212

T Palma et al. /Postharvest Biology and Technology 6 (1995) 257-273

Biggs, M.S. and Handa, A.K., 1989. Temporal regulation of polygalacturonase gene expression in fruits of normal, mutant, and heterozygous tomato genotypes. Plant Physiol., 89: 117-125. Biggs, M.S., Woodson, W.R. and Handa, A.K., 1988. Biochemical basis of high temperature inhibition of ethylene biosynthesis in ripening tomato fruits. Physiol. Plant., 72: 572-578. Borochov, A., Tirosh, T and Mayak, S., 1986. The fate of membrane proteins during flower senescence. Acta Hortic., 181: 75-80. Briskin, D.I?, Leonard, R.T and Hodges, T.K., 1987. Isolation of the plasma membrane: Membrane markers and general principles. Methods Enzymol., 148: 542-557. Brodl, M.R., 1989. Regulation of the synthesis of cellular proteins during heat shock. Physiol. Plant., 75: 439-443. Chifflet, S., Torriglia, A., Chiesa, R. and Tolosa, S., 1988. Phosphate assay in concentrated protein solutions. Anal. Biochem., 168: l-4. Christiansen, M.N., 1984. Temperature stress and membrane lipid modification. In: B.N. Timmermann, C. Steelink and EA. Loewus (Editors), Phytochemical Adaptations to Stress. Plenum Press, New York, N.Y., pp. 177-195. Demandre, C., Tremolieres, A., Justin, A.M. and Mazliak, I?, 1986. Oleate desaturation in six phosphatidylcholine molecular species from potato tuber microsomes. Biochim. Biophys. Acta, 877: 380-386. Field, R.J., 1990. Influence of chilling stress on ethylene production. In: C.Y. Wang (Editor), Chilling Injury of Horticultural Crops. CRC Press, Boca Raton, Fla., pp. 235-253. Forney, C.F. and Lipton, W.J., 1990. Influence of controlled atmospheres and packaging on chilling sensitivity. In: C.Y. Wang (Editor), Chilling Injury of Horticultural Crops. CRC Press, Boca Raton, Fla., pp. 251-267. Goto, M., Minamide, T, Fuji, M. and Iwata, T, 1984. Preventive effect of cold-shock on chilling injury of mume (Japanese apricot, Pncnus mume Sieb. et Zucc.) fruit in relation to changes of permeability and fatty acid composition of membrane. J. Jap. Sot. Hortic. Sci., 52: 210-218. Grierson, D. and Bicker, G.A., 1983. Timing of ethylene and polygalacturonase synthesis in relation to the control of ripening. Planta, 157: 174-179. Grierson, D., Tucker, G.A., Keen, J., Ray, J., Bird, CR. and Schuch, W., 1986. Sequencing and identification of a cDNA clone for tomato polygalacturonase. Nucleic Acids Res., 14: 8595-8603. Higgins, J.A., 1987. Separation and analysis of membrane lipid components. In: J.C.B. Findlay and W.H. Evans (Editors), Biological Membranes, A Practical Approach. IRL Press, Oxford, pp. 103138. Iswari, S. and Palta, J.P., 1989. Plasma membrane ATPase as a site of functional alteration during cold acclimation and freezing injury. In: PH. Li (Editor), Low Temperature Stress Physiology in Crops. CRC Press, Boca Raton, Fla., pp. 123-137. Jackman, R.L., Yada, R.Y., Marangoni, A.G., Parkin, K.L. and Stanley, D.W., 1988. Chilling injury. A review of quality aspects. J. Food Qual., 11: 253-278. Jackman, R.L., Gibson, H.J. and Stanley, D.W., 1992. Effects of chilling on tomato fruit texture. Physiol. Plant., 86: 600-608. Kuo, S.J. and Parkin, K.L., 1989. Chilling injury in cucumbers (Cucumb sativa L.) associated with lipid peroxidation as measured by ethane evolution. J. Food Sci., 54: 1488-1491. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227: 680-685. Leshem, Y.Y., Shewfelt, R.L., Willmer, C.M. and Pantoja, O., 1992. Plant Membranes. A Biophysical Approach to Structure, Development and Senescence. Kluwer Academic Press, Boston, Mass. Marangoni, A.G. and Stanley, D.W., 1989. Phase transitions in microsomal membranes from chilling sensitive and chilling resistant tomato plants and fruit. Phytochemistry, 28: 2293-2301. Marangoni, A.G., Smith, A.K., Yada, R.Y. and Stanley, D.W., 1989. Ultrastructural changes associated with chilling injury in mature green tomato fruit. J. Am. Sot. Hortic. Sci., 114: 958-962. Marangoni, A.G., Butuner, Z., Smith, J.L. and Stanley, D.W., 1990. Physical and biochemical changes in the microsomal membranes of tomato fruit associated with acclimation to chilling. J. Plant Physiol., 135: 653-661.

T Pahna et al. I Postharvest Biology and Technology 6 (1995) 257-273

273

Matsudaira, I?, 1990. Limited N-terminal sequence analysis. Methods Enzymol., 182: 602-613. Maunders, M.J., Holdsworth, M.J., Slater, A., Knapp, J.E., Bird, C.R., Schuch, W. and Grierson, D., 1987. Ethylene stimulates the accumulation of ripening related mRNA in tomatoes. Plant Cell Environ., 10: 177-184. Merril, C.L., 1990. Gel staining techniques. Methods Enzymol., 182: 477-488. Nooden, L.D., 1988. The phenomenon of senescence and aging. In: L.D. Nooden and A.C. Leopold (Editors), Senescence and Aging in Plants. Academic Press, San Diego, Calif., pp. l-50. Oeller, P.W., Min-Wong, L., Taylor, L.P., Pike, D.A. and Theologis, A., 1991. Reversible inhibition of tomato fruit senescence by antisense RNA. Science, 254: 437-439. Paliyath, G. and Droillard, M.J., 1992. The mechanism of membrane deterioration and disassembly during senescence. Plant Physiol. Biochem., 30: 789-812. Palta, J.P., 1990. Stress interactions at the cellular and membrane levels. HortScience, 25: 1377-1381. Pappin, D.J.C., Coull, J.M. and Kester, H., 1990. Solid-phase sequence analysis of proteins electroblotted or spotted onto polyvinylidene difluoride membranes. Anal. Biochem., 187: 10-19. Parkin, K.L., Marangoni, A., Jackman, R.L., Yada, R.Y. and Stanley, D.W., 1989. Chilling injury. A review of possible mechanisms. J. Food Biochem., 13: 127-153. Picton, S. and Grierson, D., 1988. Inhibition of expression of tomato-ripening genes at high temperature. Plant Ccl1 Environ., 11: 265-272. Sheehy, R.E., Pearson, J., Brady, C.J. and Hiatt, W.R., 1987. Molecular characterization of tomato fruit polygalacturonase. Mol. Gen. Genet., 208: 30-36. Stanley, D.W., 1991. Biological membrane deterioration and associated quality losses in food tissues. Crit. Rev. Food Sci. Nutr., 30: 487-553. Thompson, J.E., 1988. The molecular basis for membrane deterioration during senescence. In: L.D. Nooden and A.C. Leopold (Editors), Senescence and Aging in Plants. Academic Press, San Diego, Calif., pp. 51-83. Todd, J.E., Paliyath, G. and Thompson, J.E., 1992. Effect of chilling on the activities of lipid degrading enzymes in tomato fruit microsomal membranes. Plant Physiol. Biochem., 30: 517-522. Varga, A. and Bruinsma, J., 1986. Tomato. In: S.P. Monselise (Editor), CRC Handbook of Fruit Set and Development. CRC Press, Boca Raton, Fla., pp. 461-481. Wada, H., Gomobos, Z. and Murata, N., 1990. Enhancement of chilling tolerance of a cyanobacterium by genetic manipulation of fatty acid desaturation. Nature, 347: 200-203. Wilson, J.M. and McMurdo, A.C., 1981. Chilling injury in plants. In: G.J. Morris and A. Clarke (Editors), Effects of Low Temperatures on Biological Membranes. Academic Press, New York, N.Y., pp. 145-172. Yoshida, S., 1979. Freezing injury and phospholipid degradation in vivo in woody plant cells. Plant Physiol., 64: 241-246. Zheng, L., Heupel, R.C. and DeIlaPenna, D., 1992. The B subunit of tomato fruit polygalacturonase isoenzyme, 1. Isolation, characterization, and identification of unique structural features. Plant Cell, 4: 1147-1156.