Journal of the Science of Food and Agriculture J Sci Food Agric 81:22±29 (online: 2000)

Exposure to hypoxia conditions alters volatileconcentrations of apple cultivarsJonathan Dixon and Errol W Hewett*Institute of Natural Resources, College of Sciences, Massey University, Private Bag 11 222, Palmerston North, New Zealand

(Rec

* CoPalmContContContCont

# 2

Abstract: Brief periods of hypoxia at ambient temperatures (20°C) have potential for disinfestation

treatments or as pre-treatments to maintain fruit quality during extended storage. Nine cultivars of

apple (Cox's Orange Pippin, Fuji, Golden Delicious, Granny Smith, Paci®c Rose, Red Delicious,

Royal Gala, Splendour and Southern Snap) were exposed to hypoxia using an atmosphere of 100%

carbon dioxide for 24h at 20°C. Quantitative and qualitative analyses of volatile compounds were

undertaken after removal of fruits from a high-carbon-dioxide atmosphere and during 1 week at 20°C.

Concentrations of acetaldehdye, ethanol, ethyl acetate and ethyl esters were consistently enhanced by

hypoxia, while concentrations of acetate esters and aldehydes were depressed. Cultivars varied

considerably in response to high carbon dioxide, with Cox's Orange Pippin and Golden Delicious

having the least and Fuji and Red Delicious the greatest enhancement in ethyl esters. Fruits exposed to

hypoxia had larger odour unit scores than control fruits, suggesting that such changes in volatile

concentration may affect aroma and/or ¯avour. Enhanced ethyl ester concentrations in fruits exposed

to hypoxia may be due to increases in ethanol concentration that competitively inhibit formation of

non-ethyl esters. There may also have been a change in ester-forming enzyme activity and/or substrate

speci®city of the volatile biosynthetic pathway. This study has shown that short-term exposure to

hypoxia has the potential to change the aroma/¯avour of apples.

# 2000 Society of Chemical Industry

Keywords: Malus domestica (Borkh); hypoxia; ¯avour; volatiles; apple quality; solvent extraction; aroma; odourunits

INTRODUCTIONTypical apple aroma is the result of more than 300

volatile compounds1 including alcohols, aldehydes,

esters, ketones and ethers,2 where most are esters

(78±92%) and alcohols (6±16%). Most aroma com-

pounds are present in volatile emissions from apples,

but only a few de®ne characteristic apple aroma or

taste.1,3 Differences in apple aroma are apparent to

sensory panellists, where cultivars with strong `typical'

apple aroma are preferred.4

Aroma volatile production increases as apples ripen,

reaching a peak at the climacteric maximum.5 Vola-

tiles are synthesised from amino acids, membrane

lipids and carbohydrates.6 Fatty acids supply straight-

chain alcohols and acyl CoAs through b-oxidation.

Branched-chain volatiles are formed by metabolism of

amino acids, in particular isoleucine, leucine and

valine.7 Acetyl CoA is probably synthesised mainly

from pyruvate, as it is the substrate of the tricarboxylic

acid cycle.8 Formation of alcohols from fatty acids and

amino acids is by reduction of aldehydes to alcohols

catalysed by alcohol dehydrogenase (ADH, EC

1.1.1.1).6 Esters are produced by combining alcohols

eived 15 February 2000; revised version received 24 July 2000; acce

rrespondence to: Errol W Hewett, Institute of Natural Resourceserston North, New Zealandract/grant sponsor: Technology New Zealandract/grant sponsor: Frucor Processors Ltdract/grant sponsor: ENZAFRUIT (International)ract/grant sponsor: BOC Gases (New Zealand) Ltd

000 Society of Chemical Industry. J Sci Food Agric 0022±5142/2

and CoA derivatives of carboxylic acids in an oxygen-

dependent reaction catalysed by alcohol acyl CoA

transferase (AAT, EC 2.3.1.84).9 As acetyl CoA is the

most abundant CoA present in fruit tissue, the

majority of esters are acetate esters.

Hypoxic environments are those where the oxygen

(O2) concentration in the atmosphere surrounding

tissues or organs is insuf®cient to support aerobic

metabolism.10 Such conditions induce anaerobic

respiration where acetaldehyde and ethanol accumu-

late.11 Brief periods of hypoxia reduce postharvest

decay,12 help maintain general fruit quality,13 increase

volatile concentration in citrus and feijoa fruits14 and

are potential disinfestation treatments. Such treat-

ments use <2% O2 and up to 100% carbon dioxide

(CO2) or nitrogen (N2) atmospheres for up to 14 days

at about 20°C.15 Low-O2 conditions are considered

bene®cial for apple storage, as fruits maintain colour

and ®rmness longer than when stored in air. These

attributes de®ne apple fruit quality for storage and

marketing, but consumers require fruits that also have

acceptable ¯avour and aroma. Apples maintained in

low-O conditions for long periods (eg ultralow

2

pted 14 August 2000)

, College of Sciences, Massey University, Private Bag 11 222,

001/$30.00 22

Exposure of apples to hypoxia

controlled atmosphere (CA) storage) had reduced

volatile production with poor ¯avour and aroma

compared to fruits stored in air.16

Oxygen is considered to be an essential cofactor for

esteri®cation of alcohols in fruit tissue9 by supplying

NADH from aerobic respiration.17 In the absence of

O2, esteri®cation reactions stop and concentrations of

free alcohol increase. On return to aerobic conditions,

these alcohols are metabolised either to esters18 or to

shorter-chain compounds before esteri®cation19 or

they evaporate from the tissue. Therefore, after

removal of fruits from hypoxic conditions, there

should be an increase in concentration of a wide range

of esters in the apple aroma.

Sensory analysis indicated that fruits exposed to

low-O2/high-CO2 conditions had increased ¯avour.

Apples treated with 10±15% CO2 before CA storage

were rated by taste panellists as having better ¯avour

and texture than untreated fruits stored in CA only.20

Peaches and nectarines exposed to 86% CO2 or 97%

N2 for 1day at 20°C were preferred over untreated

fruits after 7 days at 20°C.21 Feijoa fruits treated with

98% N2�2% O2 for 24h at 20°C were rated sweeter

than control fruits after 7 days at 20°C.13 Golden

Delicious apples exposed to >95% CO2 for 24 or 48h

at 20°C had better ¯avour than untreated control

fruits after 2 weeks at 20°C, with a 24h treatment

being preferred.22 Such improvement in ¯avour may

be due to an increase in concentration of aroma/

¯avour volatiles induced by exposure to hypoxia.13

The effect of short-term exposure to hypoxia on apple

cultivars has not been characterised and it is possible

that it could increase the concentration of aroma and

¯avour volatiles, thus improving ¯avour acceptability

of apples to consumers.

Should exposure to hypoxia become a recom-

mended disinfestation treatment, then it would be

desirable for one set of treatment conditions to have

the same ef®cacy over a range of apple cultivars as well

as with new cultivars. Therefore quantitative and

qualitative changes in volatiles of a range of New

Zealand commercial apple cultivars were analysed

after exposure to a brief period of hypoxia.

EXPERIMENTALNine cultivars of apple (Malus domestica Borkh) were

assessed for their response to hypoxia. In 1996, fruits

were harvested from Hawkes Bay, New Zealand at

mid-commercial harvest on 13 February for Cox's

Orange Pippin (CO), 1 April for Fuji (F), 10 April for

Golden Delicious (GD), Granny Smith (GS), Sciros

(PR, Paci®c Rose2) and Red Delicious (RD) and 23

February for Royal Gala (RG) and Sciglo (SS,

Southern Snap2). Fruits were graded to export

standard, packed to count 100±125 (148±188g fresh

weight), transported unrefrigerated to Massey Uni-

versity, Palmerston North by road and placed at

0�1°C within 3 days of harvest. Splendour (SP)

apples were harvested from the Fruit Crops Unit,

J Sci Food Agric 81:22±29 (online: 2000)

Massey University (mid-May 1996), graded to count

125 and placed at 0�1°C. Before treatment, fruits

were removed from cool store and equilibrated to

20�3°C overnight. Fruits were divided into two

groups; one group was exposed to hypoxia (T) and

the other maintained in air as controls (C), with four

replicates of three apples each for each treatment.

Fruits were placed into 24 l Perspex chambers

connected to a manifold, in parallel, leading from a

high-pressure cylinder of pure CO2 (BOC Gases (New

Zealand) Ltd, New Zealand). During treatment,

chambers were purged continuously with humidi®ed

CO2 for 24h at 20�3°C. The initial CO2 ¯ow rate in

each chamber was 2.5 l minÿ1 for 2±3h (6.2 air

changes hÿ1), after which it was reduced to 1 l minÿ1

(2.5 air changes hÿ1). The CO2 and O2 contents of

chambers were monitored by gas chromatography,

every few minutes initially and every few hours

thereafter. A concentration of <0.5% O2 was reached

within 4h of commencing purging and remained

constant until chamber opening.

Groups of three fruits were assessed for CO2 and

ethylene production, ®rmness, weight loss, volatile

concentration of juice, and fermentation volatiles in

the juice headspace the day before treatment (dayÿ1),

on removal from treatment (day 0) and 1, 3, 5 and 7

days after treatment during ripening at 20�3°C.

Firmness was measured using a hand-held Ef®gi

penetrometer (model FT327) with an 11.1mm

measuring head on pared surfaces on opposing sides

of fruits at the equator. The average of the two

measurements was multiplied by 9.81 to convert

kilograms force (kgf) to newtons (N).

Carbon dioxide and ethylene were measured by gas

chromatography according to the following procedure.

Each replicate of three apples from each treatment was

weighed and, depending on the size of fruits, placed

into 1.6 l or 1.8 l glass preserving jars. Oneml of gas

was removed from each jar 15±30min after sealing,

using a plastic syringe (1cm3 Graduated Monoject1

syringe with detachable needle, 25 gauge�0.625inch,

Sherwood Medical, MO, USA). Carbon dioxide was

analysed using a thermal conductivity detector at

60°C and 90mA current in a Shimadzu 8A gas

chromatograph (GC) equipped with an Alltech CTR I

column (Alltech cat no 8700) at 30°C with hydrogen

at a ¯ow rate of 30ml minÿ1 as carrier gas. Ethylene

was measured using a ¯ame ionisation detector at

150°C, column 110°C and injector 190°C in a

Shimadzu 4B-PTF GC equipped with an F-1 grade,

80/100 mesh activated alumina 1.83m�3.18mm

column (Alltech cat no 80072). Nitrogen at a ¯ow

rate of 30ml minÿ1 was the carrier gas. The ¯ame was

maintained with hydrogen at 30ml minÿ1 and air at

300ml minÿ1. Carbon dioxide and ethylene concen-

trations were calculated as SI units.23

The three apples from each replicate of each

treatment were ground in a domestic juicer (Kenwood

Centrifuga model JE500, Kenwood Appliances, New

Zealand). Juice was collected and left at ambient

23

J Dixon, EW Hewett

temperature (20�3°C) for at least 30min before

volatile extraction and analysis.

The headspaces of apple juice from each treatment

were analysed for acetaldehyde, ethyl acetate and

ethanol. Thirtyml of apple juice was placed into 50ml

glass Ehrlenymer ¯asks sealed with Suba-Seal1 (No

33) rubber stoppers, maintained at 30°C in a water

bath. After at least 15min a 1ml gas sample, taken

from the ¯ask headspace, was measured using a ¯ame

ionisation detector at 180°C, column 45°C and

injector 110°C in a Pye Unicam GC ®tted with a

1.83m�3.18mm stainless steel column (Supelco cat

no 1-2212) containing a 10% Carbowax 20M coating

on 80/100 Chromosorb WAW support. Nitrogen at a

¯ow rate of 30ml minÿ1 was the carrier gas. The ¯ame

was maintained with hydrogen at 30ml minÿ1 and air

at 300ml minÿ1.

Volatile compounds were extracted from apple juice

using a diethyl ether/n-pentane solvent mixture (2:1

v/v, Analar BDH).24 Two 10ml aliquots of juice were

placed into separate 20ml scintillation vials (Wheaton

Scienti®c, NJ, USA) ®tted with a metal foil liner cap.

The internal standard (IS) consisted of 20ml lÿ1 octyl

acetate (Aldrich Chemical Company, WI, USA). Vials

were capped tightly and mixed with a vortex stirrer for

3±5s before storage at ÿ18°C until the aqueous phase

was frozen. The unfrozen solvent phase was decanted

from the aqueous phase, which was discarded. The

diethyl ether/n-pentane solvent extract was concen-

trated from 20ml to about 200ml (about 100-fold

concentration) and placed into a 250ml ¯at-bottom

glass insert (Sun International Trading cat no 200

232) in a 1.5ml glass screwtop autosampler vial (Sun

International Trading cat no 200 250) suitable for a

Hewlett Packard 5890 Series II Plus GLC autosam-

pler (Hewlett Packard 7673 controller and injector

and model 185968 100-sample carousel). Vials were

sealed with a plastic septum (Sun International

Trading cat no 200 368) before placement in the

carousel. Samples (1ml) were injected into the GLC.

Quanti®cation of volatiles in the solvent extract was by

comparison with authentic compounds made to a

Table 1. Carbon dioxide and ethylene productionand firmness for apple cultivars stored in air for 3days at 20°C, in control fruits and in fruits afterexposure to hypoxia for 24h. Mean of fourreplicates

Cultivar

Car

(nm

C a

CO 190.5

FU 143.1

GD 144.7

GS 135.2

PR 89.4

RD 112.1

RG 120.3

SP 128.8

SS 146.1

a Treatment:C, control; T

Values within a row for a

are not signi®cantly diffe

24

concentration of 200ml lÿ1 in the solvent mixture.

Odour unit values were calculated25 using published

aroma threshold values.26±28

Oneml of solvent extract was measured by capillary

gas chromatography using a Hewlett Packard 5890

Series II Plus GC connected to an IBM-compatible

personal computer equipped with Hewlett Packard

ChemStation V B.02.04 software. The capillary

column was a J&W 30m�0.32mm (id) fused silica,

DBWAX, 0.5mm ®lm thickness (Alltech cat no

93526). Injector and detector temperatures were 150

and 250°C respectively. The oven temperature was

held at 40°C for 5min, then programmed to 120°C at

5°Cminÿ1 and to 190°C at 20°Cminÿ1, with no

holding time, making a total run time of 24.5min.

Hydrogen was used as the carrier gas with a linear ¯ow

rate of 30cmsÿ1. A split injection mode was used with

a split ¯ow rate of 100ml minÿ1 and a split ratio of

15:1. The septum purge ¯ow rate was 5±6ml minÿ1.

Air and hydrogen ¯ow rates to the detector were 400

and 30ml minÿ1 respectively.

Each experiment on each cultivar was conducted as

a completely random design with sampling during

storage at 20°C as repeated measures. Means and

standard errors of the means for each cultivar and

treatment for headspace and juice volatiles were

graphed using the Origin v 5 graphics package

(Microcal Software Inc, USA). Data were subjected

to analysis of variance using SAS v 6.12 (SAS Institute

Inc, Cary, NC, USA). Signi®cant main effect means

were separated by Duncan's multiple range test at the

5% level of signi®cance. Odour units were calculated

as the sum of ratios of the concentration of a volatile

component in an extract and the average threshold

concentration of that volatile in water.29

RESULTS AND DISCUSSIONExposure of some commercial apple cultivars to

hypoxia reduced ethylene production 3 days after

treatment at 20°C but did not affect CO2 production

or fruit ®rmness (Table 1). This was similar to results

bon dioxide

ol kgÿ1sÿ1)

Ethylene

(nmol kgÿ1sÿ1) Firmness (N)

T a C T C T

a 158.8a 1.0a 0.5b 42.3a 43.7a

a 136.6a 0.3a 0.2b 70.5a 68.9a

a 153.8a 1.4a 1.7b 41.1a 43.9a

a 129.9a 0.7a 0.6a 70.0a 71.1a

a 45.6a 0.1a 0.0b 72.2a 72.2a

a 98.2a 0.5a 0.5a 61.7a 61.2a

a 136.3a 1.7a 1.4b 50.2a 52.4a

a 86.3b 0.2a 0.0b 75.5a 70.3b

a 132.5a 0.7a 0.5a 70.0a 70.2a

, exposed to hypoxia.

given cultivar for CO2, ethylene and ®rmness followed by a letter in common

rent at 5% using Duncan's multiple range test.

J Sci Food Agric 81:22±29 (online: 2000)

Exposure of apples to hypoxia

for Braeburn, GS and RD cultivars.30 While respira-

tion rate differed between cultivars, being greatest for

CO and least for PR fruits, CO2 production in control

fruits was not different from that in fruits exposed to

hypoxia. An exception was SP, which, when exposed

to hypoxia, had decreased CO2 production and fruit

®rmness compared to control fruits. No physiological

disorder or CO2 injury, external or internal, was

observed for any fruits exposed to hypoxia. Similar

effects on CO2 and ethylene production after treat-

ment with a wide range of CO2 concentrations and

exposure times have been reported for peaches,

nectarines,13 pears,31 strawberries32 and apples.33

Overall these results suggest that brief exposure of

major cultivars to hypoxia will have little effect on

ripening of apples.

High-CO2 treatments consistently enhanced con-

centrations of acetaldehyde, ethanol and ethyl acetate

in all cultivars examined, both on removal from

hypoxia (Table 2) and during ripening at 20°C (Table

3). Acetaldehyde, ethyl acetate and ethanol concen-

trations in control fruits varied according to cultivar.

There was no relationship between control concentra-

tions and magnitude of enhancement of acetaldehyde,

ethanol and ethyl acetate after hypoxia (Tables 2 and

3). For acetaldehyde, greatest enhancement was 1day

after removal from hypoxia, with GS fruits having the

least increase and RG fruits the greatest (Table 2).

Ethyl acetate concentrations were low on removal

Table 2. Concentration of volatiles (mmol lÿ1) from apple cultivars maintained in air

CO FU GD GS

C a T a C T C T C T

Alcohols

Ethanol 5621a 18100b 1486a 17623b ND 14511b 1699a 7432

Propan-1-ol 36a 39a 28a 91b 61a 142b ND 28

Butan-1-ol 1044a 1795b 975a 853a 1407a 2820b 43a 20

Pentan-1-ol 7a 17b 15a 15a 10a 26b ND ND

Hexan-1-ol 117a 259b 130a 122a 149a 313b 31a 25

2MBb 11a 31a 240a 354a 43a 130b 51a 41

Acetate esters

Ethyl acetate 3a 8b 1a 9b 1a 5b 1a 3

Propyl acetate 43a 48a ND ND 20a 19a ND ND

Butyl acetate 543a 383b 93a 46b 452a 332b ND ND

Pentyl acetate 12a 6a ND 3a 7a 7a ND ND

Hexyl acetate 120a 91b 23a 14a 74a 60a 11a 8

2MBAb 22a 17a 73a 36b 33a 31a ND ND

Ethyl esters

Ethyl propionate ND ND 4a 8a 4a 10a 7a 4

Ethyl butanoate ND 11b 10a 36b 11a 22b 13a 12

Ethyl pentanoate ND 2a 13a 12a 15a 29b 15a 11

Ethyl hexanoate ND ND ND 6b ND 6b ND ND

E2MBb ND ND ND ND ND 2a ND ND

Aldehydes

Acetaldehyde 85a 291b ND 279b ND 421b ND 56

Hexanal 75a 51a 29a 25a 92a 91a 41a 16

trans-2-Hexenal 476a 346a 156a 160a 166a 179a 283a 215

a Treatment:C, control; T, exposed to hypoxia.b 2MB, 2- and 3-methyl butan-1-ol; 2MBA, 2-methyl butyl acetate; E2MB, ethyl-2-m

Means within a row for a given cultivar followed by a letter in common are not sig

(concentration below 0.1mmol lÿ1).

J Sci Food Agric 81:22±29 (online: 2000)

from hypoxia (typically less than 10mmol lÿ1), but,

depending on cultivar, 3 days after removal from

hypoxia had increased in CO, FU, GD, PR, RD, SP

and SS fruits while remaining low in GS and RG fruits

(Tables 2 and 3).

Endogenous ethanol concentrations were high in

control CO, FU and RG apples during normal

ripening compared to other cultivars, in agreement

with previous results.34 Cox's Orange Pippin had the

highest endogenous concentration of ethanol amongst

cultivars, but hypoxia only induced a three fold

increase. Ethanol was not detected in control fruits

of PR and GD, but increased to 6.9 and 14.5mmol lÿ1

respectively on removal from hypoxia. Such increases

made ethanol the most abundant alcohol in the fruit

tissue.

Ethanol concentrations, although declining, re-

mained high for several days after treatment (Table

3). Similar enhancement in ethanol concentration has

been noted in blueberries,35 grapes36 and oranges37

following anaerobic treatments. The persistence of

high ethanol concentrations in hypoxic-treated fruit

could have provided a large pool of substrate for rapid

and sustained production of ethyl esters, even though

fruits had been returned to air. Therefore, in fruits

exposed to hypoxia, ethyl ester production has the

potential to be enhanced for some time after removal

to air.

In general, concentrations of alcohols were greater

at 20°C and on removal from hypoxia (day 0). Mean of four replicates

PR RD RG SP SS

C T C T C T C T C T

b ND 6879b 215a 4738b 4477a 9240b 87a 15682b 969a 4738b

b ND 22a 19a 91b 66a 106b 34a 74a 10a 94b

b 260a 186a 324a 488a 2546a 2622a 670a 835a 1326a 1464a

6a 5a ND 6b 17a 21a 16a 12a 25a 18a

a 47a 35a 44a 64b 180a 232a 94a 96a 187a 214a

a 33a 50a 28a 50a 52a 86a 111a 79a 76a 101a

b ND 4b 1a 7b 1a 2a 1a 6b 2a 10b

8a 8a 16a 31a 48a 47a ND ND 40a 33a

55a 39a 83a 87a 404a 188b 74a 63a 427a 228b

ND ND ND ND 2a ND ND ND 17a 8a

a 17a 16a 23a 28a 101a 37b 20a 21a 106a 79a

23a 17a 28a 25a 32a 14b 32a 6a 58a 44a

a ND 6a ND 9a ND ND 3a 7a ND 4a

a 12a 22b 3a 36b ND ND 13a 42b 8a 29b

a 15a 15a 13a 13a 6a 2a 13a 14a 19a 15a

ND 9b ND 15b ND ND ND ND 2a 12a

ND ND ND ND ND ND ND ND ND ND

b ND 127b ND 226b 50a 557b ND 222b 19a 336b

a 28a 13a ND ND 72a 68a 19a 0b 85a 61a

a 88a 98a 124a 128a 114a 128a 200a 147b 363a 220a

ethyl butanoate.

ni®cantly different at 5% by Duncan's multiple range test. ND, not detected

25

Table 3. Concentration of volatiles (mmol lÿ1) from apple cultivars maintained in air at 20°C and on removal from hypoxia (day 3). Mean of four replicates

CO FU GD GS PR RD RG SP SS

C a T a C T C T C T C T C T C T C T C T

Alcohols

Ethanol 3589a 17947b ND 4604b 4a 26892b ND 292a 30a 6381b 22a 9448b 184a 1693a ND 10130b ND 24120b

Propan-1-ol ND 34b 183a 173a 149a 151a 69a 60a 38a 55a 106a 135a 136a 196a 139a 72a 66a 107a

Butan-1-ol 1306a 2177a 857a 838a 1630a 2326a 60a 110a 253a 347a 365a 469a 3384a 4640a 1174a 624a 1416a 1697a

Pentan-1-ol 16a 16a 11a 13a 14a 15a 9a 2a 9a 12a 11a 13a 26a 28a 24a 11a 16a 27b

Hexan-1-ol 119a 143a 111a 80a 236a 229a 53a 68a 52a 57a 53a 70a 236a 281a 217a 60b 142a 234b

2MBb 21a 27a 297a 272a 79a 89a 224a 82a 70a 45a 112a 67b 106a 116a 169a 59b 71a 78a

Acetate esters

Ethyl acetate 1a 87b ND 9b 1a 46b 1a 1a 2a 27b 2a 26b 1a 8a 1a 16b 1a 38b

Propyl acetate 14a 14a ND ND ND ND ND 3a ND 6a 31a 3b 3a 35a 5a ND ND ND

Butyl acetate 482a 448a 63a 36b 390a 336a 1a ND 44a 39a 72a 63a 516a 534a 128a 37b 273a 164a

Pentyl acetate 13a ND ND ND 5a 2a ND ND ND 2a 2a 3a 9a 11a 4a ND 8a 7a

Hexyl acetate 98a 57a 18a 10b 111a 50b 10a 10a 15a 12a 30a 19b 137a 118a 57a 10b 58a 40a

2MBAb 25a 15a 74a 23b 47a 26b ND ND 44a 9a 111a 29b 62a 49a 59a 0b 59a 26b

Ethyl esters

Ethyl propionate 5a ND 6a 128b 7a ND 8a 52b ND 121b 3a 216b ND 10a 14a 151b 10a 98b

Ethyl butanoate 11a 25b 13a 207b 13a 17a 12a 74a 9a 238b 10a 353b 11a 15a 15a 244b 10a 346b

Ethyl pentanoate 17a 15a 11a 13a 14a 13a 14a 13a 13a 15a 19a 23a 17a 19a 17a 13a 20a 21a

Ethyl hexanoate 2a ND ND 35b ND ND ND 5a ND 24b 2a 65b 7a 2a ND 27b 2a 24b

E2MBb ND ND ND 52b ND ND 1a 7a ND 21b ND 49b ND ND ND 16b ND 17b

Aldehydes

Acetaldehyde 11a 346b ND 149b ND 130b ND 71a ND 35b ND 118b 11a 28a ND 265b ND 182b

Hexanal 47a 58a 71a 71a 207a 118b 75a 71a 69a 50a 28a 67a 161a 158a 97a 41b 107a 70a

trans-2-Hexenal 341a 306a 173a 181a 232a 155a 389a 338a 174a 111b 170a 175a 196a 211a 348a 171b 265a 192a

a Treatment:C, control; T, exposed to hypoxia.b 2MB, 2- and 3-methyl butan-1-ol; 2MBA, 2-methyl butyl acetate; E2MB, ethyl-2-methyl butanoate.

Means within a row for a given cultivar followed by a letter in common are not signi®cantly different at 5% by Duncan's multiple range test. ND, not detected

concentration below 0.1mmol lÿ1).

Figure 1. Concentration of selected aroma volatiles extracted from juice ofCO (*) and RD (~) apples exposed to 100% CO2 (� � � �) or air (—) for 24hand maintained at 20°C. Aroma volatiles shown represent compoundsthought to have important sensory characteristics in apple.46 Average andstandard error of four replicates.

J Dixon, EW Hewett

in apples on removal from hypoxia than in control

fruits, and cultivar-speci®c enhancement of particular

alcohols occurred (Table 2). For example, GD fruits

had enhanced concentrations of propan-1-ol, butan-

1-ol, pentan-1-ol, hexan-1-ol and 2-methyl butan-1-

ol on removal from hypoxia, while propan-1-ol,

pentan-1-ol and hexan-1-ol concentrations were

increased in RD fruits. In contrast, no enhancement

of alcohols was noted for PR and SP fruits. By 3 days

after removal from hypoxia there were few differences

in alcohol concentration between treatments,

although propan-1-ol concentration was increased in

CO fruits and pentan-1-ol and hexan-1-ol concentra-

tions were enhanced in SS fruits (Table 3). Signi-

®cant decreases in 2-methyl butan-1-ol were noted

for RD and SP fruits, while hexan-1-ol decreased in

SP fruits.

In general, concentrations of hexanal and trans-2-

hexenal in fruits exposed to hypoxia were unaffected

compared to control apples, both at removal from

hypoxia and after 3 days at 20°C (Tables 2 and 3).

Production of hexanal and trans-2-hexenal is an O2-

dependent process occurring by degradation of linoleic

and linolenic acids via the lipoxygenase pathway,6 and

this process might be inhibited by hypoxia.

The general effect of a brief exposure to hypoxia was

for some or all ethyl esters to be enhanced and for

acetate esters to decrease (Fig 1). On removal from

26 J Sci Food Agric 81:22±29 (online: 2000)

Exposure of apples to hypoxia

hypoxia there were decreased concentrations of butyl

acetate in CO, FU, GD, RG and SS fruits, hexyl

acetate in CO and RG fruits and 2-methyl butyl

acetate in FU and RG fruits compared to control

fruits (Table 2). There was no change in concentra-

tion of acetate esters for GS, PR, RD and SP fruits.

Acetate ester concentrations in fruits exposed to

hypoxia continued to decrease during storage in air

(Table 3). After 3 days at 20°C, FU, GD, RD, SP

and SS fruits had decreased concentrations of butyl,

hexyl and 2-methyl butyl acetate in high-CO2-

treated fruits compared with control fruits. The

decreases in acetate ester concentrations induced by

hypoxia were less than the increases in ethyl ester

concentrations, for example, butyl acetate decreased

3.4-fold while ethyl butanoate increased 34.5-fold in

SP fruits exposed to hypoxia compared to control

fruits.

Volatile concentrations in control fruits after 3 days

at 20°C were generally higher than in fruits at day 0

(Table 2), and this increase was not related to ethylene

production (data not shown). Alcohols and acetate

esters were present at day 0 and ethyl hexanoate and

ethyl-2-methyl butanoate appeared during ripening,

indicating that certain ethyl esters are synthesised as

ripening progresses. In general, enhancement of ethyl

esters reached a maximum 3 days after removal from

hypoxia and declined slowly thereafter (Fig 1). High-

CO2 treatment enhanced ethyl ester concentrations at

day 0 at 20°C in CO, FU, GD, PR, RD, SP and SS

fruits but had no effect on GS and RG fruits (Table 2).

Ethyl butanoate was enhanced in CO, FU, GD, PR,

RD, SP and SS fruits and ethyl hexanoate was

enhanced in FU, GD, PR and RD fruits. Ethyl

pentanoate was enhanced only in GD fruits on

removal from a high-CO2 atmosphere. After 3 days

at 20°C, concentrations of ethyl esters in control and

treated fruits were greater, and a different pattern of

ethyl ester enhancement was apparent, than in day 0

fruits (Table 3). Ethyl butanoate, ethyl hexanoate and

ethyl-2-methyl butanoate were present in increased

concentrations in FU, PR, RD, SP and SS fruits, while

only ethyl butanoate was increased in CO fruits. Ethyl

esters were no longer enhanced in GD fruits, and ethyl

pentanoate concentrations in hypoxic-treated fruits

were similar to concentrations in controls. Ethyl

propionate at day 0 was present in FU, GD, GS and

SP fruits, but after 3 days was present in CO, FU, GD,

GS, RD, SP and SS apples. Ethyl hexanoate only

appeared in volatile extracts of CO, RD, RG and SS

fruits after 3 days at 20°C. At day 0, control CO fruits

had no ethyl esters while RG fruits had only ethyl

pentanoate, but after 3 days there was a greater

number of ethyl esters for both cultivars. Depending

on cultivar and ester, the proportional increases in

ethyl esters following exposure to hypoxia were large,

being up to 86.5-fold for ethyl acetate in CO fruits,

34.5-fold for ethyl butanoate in SS fruits and 65-fold

for ethyl propionate in RD fruits 3 days after treatment

(Table 3).

J Sci Food Agric 81:22±29 (online: 2000)

These results support the hypothesis that ethyl ester

concentrations increase following hypoxia because

excess ethanol is preferentially available for reaction

with available acyl CoAs to form ethyl esters.38 Similar

increases in ethyl esters occurred in Delicious apples

after hypoxia for 30 days at 1°C,39 grapes after

carbonic maceration40 and oranges after 24h of

hypoxia.41

It is suggested that changes in volatile con-

centration following high-CO2 treatment could be

due to two effects on the volatile biosynthetic

pathway working either independently of one

another or in combination. First, the large increase

in ethanol concentration may competitively inhibit

formation of non-ethyl esters.38 Second, hypoxic

conditions induce new isozymes of enzymes that use

ethanol as the preferred substrate for formation of

ester compounds. New isozymes of ADH are

produced in response to hypoxia in grapes,40

tomatoes42 and avocados.43 High concentrations of

ethanol, induced by hypoxia, competitively use

available acyl CoAs, resulting in large increases in

ethyl esters and decreased non-ethyl ester concen-

trations. The fact that high-CO2 treatment induced

large increases in ethyl esters, decreases in non-ethyl

esters and increases in concentration of C3�alcohols on removal from hypoxia supports this

hypothesis. Similar increases in ethyl esters and

decreases in acetate esters occurred in Delicious

apples exposed to hypoxia for long periods at low

temperatures.39

Different fruits have different AAT isozymes each

with their own preference for speci®c alcohols and acyl

CoAs.44 Differences may exist between cultivars in

substrate speci®cities of AAT for alcohols or acyl CoAs

as has been found in strawberries.6 Exposure to

hypoxia induces changes in AAT activity,45 and it is

possible that, by analogy with ADH, such a treatment

may induce new isozymes of AAT with different

substrate speci®cities for alcohols or acyl CoAs. It is

also possible that a period of hypoxia induces changes

in the concentration of acyl CoAs by in¯uencing the

b-oxidation biosynthetic pathway in which long-

carbon-chain acyl CoAs are synthesised.6 However,

little is known of the factors that affect acyl CoA

concentrations and synthesis in fruits, and this would

be a most interesting and valuable topic for further

research.

The relative contributions to apple aroma by ethyl

and acetate esters are different; ethyl esters contribute

fruity apple-like characteristics, while acetate esters

contribute more fruity/solvent-like overtones.46 Odour

unit values, based on volatile concentrations presented

in Tables 2 and 3, indicate that those cultivars with

enhanced ethyl esters and decreased acetate esters had

the highest odour unit values (Fig 2). Exposure to high

CO2 increased the concentration of volatiles affecting

aroma intensity, with the variation among cultivars

ranging from a 1.5-fold decrease in GD to a 10-fold

increase in RD (Fig 2).

27

Figure 2. Odour unit values for apple juice from (a) control apples and (b)apples exposed to hypoxia, on removal from treatment (day 0) and after 3days at 20°C.

J Dixon, EW Hewett

CONCLUSIONA brief period of hypoxia enhanced ethyl esters and

anaerobic volatiles while suppressing acetate esters

and aldehydes for a wide range of apple cultivars.

Cultivars ranked from greatest to least for enhance-

ment in ethyl esters after 3 days at 20°C following

removal from hypoxia were RD, SS, SP, FU, PR, GS,

RG, CO and GD. Enhancement of ethyl esters may be

due to competitive inhibition by ethanol of biosynth-

esis of esters from other alcohols and/or to a change in

the enzymes which produce volatile compounds. Any

postharvest treatment using brief periods of hypoxia at

ambient temperatures, such as disinfestation, has the

potential to change the aroma and ¯avour of apples,

albeit transiently. These results indicate that enhance-

ment of volatiles induced by exposing apples to

hypoxia is transitory, lasting up to 7 days in some

cases. The long-term effect of hypoxia before or during

air or controlled atmosphere low-temperature storage

requires investigation.

ACKNOWLEDGEMENTSThe authors thank Technology New Zealand, Frucor

Processors Ltd, ENZAFRUIT (International) and

BOC Gases (New Zealand) Ltd for ®nancial assis-

tance.

REFERENCES1 Paillard NMM, The ¯avour of apple, pears and quinces, in Food

Flavours, Part C. The Flavour of Fruits, Ed by Morton ID and

MacLeod AJ, Elsevier Science, Amsterdam, pp 1±41 (1990).

2 Dimick PS and Hoskin JC, Review of apple ¯avorÐstate of the

art. CRC Crit Rev Food Sci Nutr 18:387±409 (1983).

3 Cunningham DG, Acree TE, Barnard J, Butts RM and Breall

PA, Charm analysis of apple volatiles. Food Chem 19:137±147

(1986).

4 Poll L, Evaluation of 18 apple varieties for their suitability for

juice production. J Sci Food Agric 32:1081±1090 (1981).

5 Song J and Bangerth F, The effect of harvest date on aroma

compound production from `Golden Delicious' apple fruit and

relationship to respiration and ethylene production. Postharv

Biol Technol 8:259±269 (1996).

28

6 Sanz C, Olias JM and Perez AG, Aroma biochemistry of fruits

and vegetables, in Phytochemistry of Fruit and Vegetables, Ed by

TomaÂs-BarberaÂn FA and Robins RJ, Oxford University Press,

New York, pp 125±155 (1997).

7 Heath HB and Reineccius G, Biogenesis of ¯avour in fruits and

vegetables, in Flavour Chemistry and Technology. AVI, West-

port, CT, pp 43±70 (1986).

8 Mathews CK and van Holde KE, Biochemistry, 2nd edn.

Benjamin/Cummings Pub Co Inc, Menlo Park, CA (1996).

9 Harada M, Ueda Y and Iwata T, Puri®cation and some

properties of alcohol acetyltransferase from banana fruit. Plant

Cell Physiol 26:1067±1074 (1985).

10 Ricard B, CoueÂe I, Raymond P, Sagilo PH, Saint-Ges V and

Pradet A, Plant metabolism under hypoxia and anoxia. Plant

Physiol Biochem 32:1±10 (1994).

11 Kader AA, Biochemical and physiological basis for effects of

controlled and modi®ed atmospheres on fruit and vegetables.

Food Technol 40:99±104 (1986).

12 Ke D, Goldstein L, O'Mahony M and Kader AA, Effects of

short-term exposure to low O2 and high CO2 atmospheres on

quality attributes of strawberries. J Food Sci 56:50±54 (1991).

13 Pesis E, Enhancement of fruit aroma and quality by acetaldehyde

or anaerobic treatments before storage. Acta Hort 368:365±373

(1994).

14 Shaw PE, Moshonas MG and Pesis E, Changes during storage of

oranges pretreated with nitrogen, carbon dioxide and acet-

aldehyde in air. J Food Sci 56:469±474 (1991).

15 Hallman GJ, Controlled atmospheres, in Insects Pests and Fresh

Horticultural Products: Treatments and Responses, Ed by Paull

RE and Armstrong JW, CAB International, Wallingford, pp

121±136 (1994).

16 Yahia EM, Liu FW and Acree TE, Changes in some odor-active

volatiles in controlled atmosphere-stored apples. J Food Qual

13:185±202 (1990).

17 Molina I, Salles C, Nicolas M and Crouzet J, Grape alcohol

dehydrogenase. II. Kinetic studies: mechanism, substrate, and

coenzyme speci®city. Am J Enol Vitic 38:60±64 (1987).

18 Yamashita I, Nemoto Y and Yoshikawa S, Formation of volatile

alcohols and esters from aldehydes in strawberries. Phytochem-

istry 15:1633±1637 (1976).

19 Rowan DD, Allen JM, Fielder S and Hunt MB, Deuterium

labelling to study aroma biosynthesis in stored apples, in CA

'97 Proceedings Volume 2: Apples and Pears, Ed by Mitcham EJ.

Univ California, Davis, CA, pp 227±233 (1997).

20 Tietjen WH and Hudson DE, Market quality of eastern-grown

`Golden Delicious' apples after prestorage CO2 treatment and

controlled-atmosphere storage. HortSci 19:427±429 (1984).

21 Lurie S and Pesis E, Effect of acetaldehyde and anaerobiosis as

postharvest treatments on the quality of peaches and nectar-

ines. Postharv Biol Technol 1:317±326 (1992).

22 Pesis E, Ampunpong C, Shusiri B and Hewett EW, Enhance-

ment of ethylene and CO2 production in apple fruit following

short-term exposure to high CO2. Postharv Biol Technol 4:309±

317 (1994).

23 Banks NH, Cleland DJ, Cameron AC, Beaudry RM and Kader

AA, Proposal for a rationalized system of units for postharvest

research in gas exchange. HortSci 30:1129±1131 (1995).

24 Larsen M and Poll L, Quick and simple extraction methods for

analysis of aroma compounds in fruit products, in Flavour

Science and Technology, Ed by BessieÁre Y and Thomas AF,

Wiley, Chichester, pp 209±212 (1990).

25 Teranishi R, Buttery RG and Schamp N, The signi®cance of low

threshold odor compounds in aroma research, in Flavor Science

and Technology, Ed by Martens M, Dalen GA and Russwurm

H Jr, Wiley, Chichester, pp 515±527 (1987).

26 Flath RA, Black DR, Guadagni DG, McFadden WH and

Schultz TH, Identi®cation and organoleptic evaluation of

compounds in Delicious apple essence. J Agric Food Chem

15:29±35 (1967).

27 Takeoka G, Buttery RG, Turnbaugh JG and Benson M, Odour

J Sci Food Agric 81:22±29 (online: 2000)

Exposure of apples to hypoxia

thresholds of various branched esters. Lebensm Wiss Technol

28:153±156 (1995).

28 Takeoka G, Buttery RG and Ling L, Odour thresholds of various

branched and straight chain acetates. Lebensm Wiss Technol

29:677±680 (1996).

29 Frijters JER, Some psychophysical notes on the use of the odour

unit number, in Progress in Flavour Research, Ed by Land DG

and Nursten HE, Applied Science Publishers, London, pp 47±

51 (1997).

30 Ampun W, Enhancement of aroma volatile compounds in

apples. PhD Thesis, Massey University, Palmerston North

(1997).

31 Wang CY and Mellenthin WM, Effect of short-term high CO2

treatment on storage of `d'Anjou' pear. J Am Soc Hort Sci

100:492±495 (1975).

32 Larsen M and Watkins CB, Firmness and aroma composition of

strawberries following short-term high carbon dioxide treat-

ments. HortSci 30:303±305 (1995).

33 Hribar J, Plestenjak A, Vidrih R and Simcic M, In¯uence of CO2

shock treatment and ULO storage on apple quality. Acta Hort

368:634±640 (1994).

34 Fidler JC and North CJ, The effect of periods of anaerobiosis on

the storage of apples. J Hort Sci 46:213±221 (1971).

35 Saltveit ME and Ballinger WE, Effects of anaerobic nitrogen and

carbon dioxide atmospheres on ethanol production and

postharvest quality of blueberries. J Am Sci Hort Sci

108:459±462 (1983).

36 Saltveit ME and Ballinger WE, Effects of anaerobic nitrogen and

carbon dioxide atmospheres on ethanol production and

postharvest quality of `Carlos' grapes. J Am Sci Hort Sci

108:462±465 (1983).

37 Pesis E and Avissar I, The post-harvest quality of orange fruits as

J Sci Food Agric 81:22±29 (online: 2000)

affected by pre-storage treatments with acetaldehyde vapour or

anaerobic conditions. J Hort Sci 64:107±113 (1989).

38 Ke D, Zhou L and Kader AA, Mode of oxygen and carbon

dioxide action on strawberry ester biosynthesis. J Am Soc Hort

Sci 119:971±975 (1994).

39 Mattheis JP, Buchanan DA and Fellman JK, Change in apple

fruit volatiles after storage in atmospheres inducing anaerobic

metabolism. J Agric Food Chem 39:1602±1605 (1991).

40 TesnieÁre CM, Romieu C and Vayda ME, Changes in the gene

expression of grapes in response to hypoxia. Am J Enol Vitic

44:445±451 (1993).

41 Shaw PE, Carter RD, Moshonas MG and Sadler G, Controlled

atmosphere storage of oranges to enhance aqueous essence and

essence oil. J Food Sci 55:1617±1619 (1990).

42 Chen AS and Chase T, Alcohol dehydrogenase 2 and pyruvate

decarboxylase induction in ripening and hypoxic tomato fruit.

Plant Physiol Biochem 31:875±885 (1993).

43 Kanellis AK, Solomos T and Roubelakis-Angelakis KA, Sup-

pression of cellulase and polygalacturonase and induction of

alcohol dehydrogenase isoenzymes in avocado fruit mesocarp

subjected to low oxygen stress. Plant Physiol 96:269±274

(1991).

44 Olias JM, Sanz C, Rios JJ and Perez AG, Substrate speci®city of

alcohol acyltransferase from strawberry and banana fruits.

ACS Symp Ser 596:134±141 (1995).

45 Fellman JK, Mattinson DS, Bostick BC, Mattheis JP and

Patterson ME, Ester biosynthesis in `Rome' apples subjected

to low-oxygen atmospheres. Postharv Biol Technol 3:201±214

(1993).

46 Rizzolo A, Polesello A and Teleky-VaÁmossy GY, CGC/sensory

analysis of volatile compounds developed from ripening apple

fruit. J High Resol Chromatogr 12:824±827 (1989).

29