Plant Physiol. (1991) 97, 798-8030032-0889/91 /97/0798/06/$01 .00/0

Received for publication April 22, 1991Accepted May 17, 1991

Abscisic Acid Metabolism in Salt-Stressed Cells ofDunaliella salina

Possible Interrelationship with fB-Carotene Accumulation

A. Keith Cowan* and Peter D. RoseDepartment of Botany, Schonland Botanical Laboratories (A.K.C.), and Department of Biochemistry and

Microbiology, Goldfields Biotechnology Laboratory (P.D.R.), Rhodes University, Grahamstown 6140, South Africa

ABSTRACT

The interrelationship between abscisic acid (ABA) productionand ti-carotene accumulation was investigated in salt-stressedcells of the halotolerant green alga Dunaliella salina var bardawil.Cells were supplied with either R-[2-14C]mevalonolactone or [14C]sodium bicarbonate for 20 hours and then exposed to increasedsalinity (1.5 to 3.0 molar NaCI) for various lengths of time. Incor-poration of label into abscisic acid and phaseic acid and thedistribution of [14C]ABA between the cells and incubation mediawere monitored. [14C]ABA and [14C]phaseic acid were identifiedas products of both R-[2-14C]mevalonolactone and [14CJsodiumbicarbonate metabolism. ABA metabolism was enhanced by hy-persalinity stress. Actinomycin D, chloramphenicol, and cyclohex-imide abolished the stress-induced production of ABA, suggest-ing a role for gene activation in the process. Kinetic analysis ofboth ABA and j-carotene production demonstrated two stagesof accelerated a-carotene production. In addition, ABA levelsincreased rapidly, and this increase occurred coincident with theearly period of accelerated fl-carotene production. A possible rolefor ABA as a regulator of carotenogenesis in cells of D. salina istherefore discussed.

The unicellular green alga Dunaliella salina var Bardawilresponds to salinity stress by regulating the flux of carbonbetween starch production in the chloroplast and the synthesisof glycerol in the cytoplasm (4). A further consequence of saltstress in this organism includes the accumulation of fl-caro-tene (2, 7). These physiological responses would appear tofunction as mechanisms designed to protect the organismagainst conditions of stress, in particular high-intensity irra-diation (3) and either hypo- or hypersalinity (12), and couldtherefore be mediated by hormonal activity.ABA has been characterized as an endogenous compound

in extracts ofDunaliella (20), and levels ofthis phytohormoneincrease when cells are exposed to salinity stress (1 1). Thislatter observation might therefore suggest that ABA is in-volved in mediating some of the responses of this alga tochanges in salinity. Furthermore, an increase in the levels ofboth fl-carotene and ABA in response to salt stress might beindicative of a carotenoid origin of ABA in the phylumPhycophyta. Recent evidence, from studies using higher planttissues, favors the biosynthesis of ABA from all-trans-violax-

anthin, a process involving cleavage of 9-cis-neoxanthin toyield xanthoxin, which is then converted to ABA via ABA-aldehyde ( 14, 16, 1 8, 24). This proposed biosynthetic pathwayfor ABA in higher plants emphasizes the interrelationship thatcould exist between fl-carotene accumulation and ABA pro-duction in cells of Dunaliella exposed to salinity stress.For ABA to play a physiological role in cells of Dunaliella,

endogenous levels of this hormone need to be regulated byboth the rate of synthesis and the rate of catabolism orsequestration of the hormone in an inactive form. Further-more, ABA should influence one or more physiological proc-ess, thereby allowing cells to tolerate prevailing stressful con-ditions. In a recent publication, Hirsch et al. (11) reportedattempts to address these possibilities. These authors demon-strated elevated levels of ABA from ['4C]mevalonic acid andconversion of ABA to PA', dihydrophaseic acid, and glucoseconjugates. Unfortunately, they were unable to demonstratean effect of ABA on any of the physiological processes inves-tigated. Nevertheless, their studies do provide good circum-stantial evidence of a role for ABA in the response of algae toconditions of salt stress. In this paper we describe studies thatwere carried out to reassess ABA metabolism, and evaluatethe interrelationship between A-carotene and ABA accumu-lation, in cells of D. salina exposed to salinity stress.

MATERIALS AND METHODS

Growth Conditions

Dunaliella salina var Bardawil Teod., strain CCAP 19/30,was obtained from Culture Collection for Algae and Protozoa,United Kingdom. Cells were cultivated in a defined growthmedium containing, unless otherwise stated, 1.5 M NaCl, 5mM MgSO4, 0.3 mm CaCl2, 5 mM KNO3, 0.2 mM KH2PO4,1.5 ,uM FeC13, 6 AM EDTA, 50 mm NaHCO3, and a tracemetal mixture all at pH 8.0 as described by Ben-Amotz andAvron (2). Cells were cultured in a constant environmentchamber at 25°C under continuous illumination ( 165 Amolm-2 s') from cool-white fluorescent lamps. Cell numberwas determined daily using an Improved Neubauer hemocy-tometer, and cells were harvested when the culture hadreached a density of between 27 and 33 x 104 cells/mL.

Abbreviations: PA, phaseic acid; MVAL, mevalonic acid lactone.798

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.

ABSCISIC ACID METABOLISM IN DUNALIELLA SALINA

Application of Radiochemicals and IncubationProcedures

Cells of D. salina were harvested by centrifugation at 20°Cin a Sorvall RC-5 refrigerated centrifuge at 3000g for 10 min.The algal pellet was washed in 1.5 M KCI to reduce "clump-ing" and recentrifuged at 3000g for 10 min. The washed pelletwas resuspended in fresh growth medium and, unless other-wise stated, 6.0 mL (equivalent to 42 x 106 cells) were

incubated in 25-mL Erlenmeyer flasks containing either R-[2-'2C]MVAL (specific activity 2.03 GBq/mmol), NaH['4Cj03 (specific activity 1.91 GBq/mmol), or (R,S,)-[2-'4C]ABA(specific activity 1.30 GBq/mmol), which were all obtainedfrom Amersham International, Buckinghamshire, England.Incubations were carried out in a shaking reaction incubatorat 25°C under continuous low-light illumination (42 gmol -

s-2 '). After 20 h, the uptake of radioactive substrateswas stopped by centrifugation and repeated washing (threetimes) of cells in equal volumes of 1.5 M KCI. Pellets were

resuspended in fresh growth medium, containing 3.0 M NaCl,and incubated as above for various lengths oftime as specifiedin "Results." Incubations were terminated by centrifugationand washing in fresh culture medium of identical NaCl mo-

larity. Chl content was determined (1) and cells, incubationmedia, and washings were extracted and analyzed for ABAand its acidic products as described below.

Incubations with Inhibitors

Cycloheximide (0.2 ,g * mL-') and chloramphenicol (50,ug * mL-') were supplied to low-light-illuminated cells 4 hbefore exposure to salinity stress, and actinomycin D (50 ,ug* mL') was added 12 h before exposure. Incubation flaskscontaining chloramphenicol or actinomycin D and their re-

spective controls were covered with yellow cellophane as

described by Lers et al (13).

Extraction, Purification, and Analysis of RadioactiveABA and PA

Algal material, harvested by centrifugation, was usuallyextracted by sonication in ice-cold methanol/ethyl acetate(50:50, v/v) containing butylated hydroxytoluene (20 mg -

L-'). Extracts were filtered through Whatman No. 1 filterpaper, and the residue was washed with excess solvent. Thefiltrate was reduced to dryness in vacuo at 35°C and resus-

pended in 0.5 M potassium phosphate buffer (pH 8.5). Theaqueous phase was partitioned three times against equal vol-umes of diethyl ether to remove neutral and basic impurities.Aqueous extracts were then acidified to pH 2.5 and parti-tioned three times against equal volumes of ethyl acetate toextract the acids. Water was removed by freezing and filtra-tion, and the ethyl acetate-soluble acids were applied to a

prerinsed Sep-Pak C,8 cartridge (Waters Associated, Milford,MA). Incubation media containing the respective washingswere acidified to pH 2.5 and likewise partitioned three timesagainst equal volumes ofethyl acetate. These were also appliedto a prerinsed Sep-Pak C,8 cartridge. ABA and PA were elutedfrom the cartridges using a profile similar to that describedby Pierce and Raschke (17). Purified samples were chromat-ographed on thin layers of silica gel GF254 (Merck) developed

twice to 15 cm in toluene/ethyl acetate/acetic acid (25:15:2,v/v/v). Plates were air dried, divided into 30 equal strips, andeluted with absolute methanol, and the levels of radioactivitywere determined in a Beckman LS 5801 scintillation counter,programmed for automatic quench correction, following ad-dition of 10 mL of cocktail (2, 5-diphenyl oxazole pyrophos-phate in toluene 5 g * L-'). Radioactive ABA and PA wereidentified according to the criteria of Milborrow (15) andZeevaart and Milborrow (23) using biosynthetically preparedPA (8) and authentic ABA (Sigma) as markers. Putative ABAand PA, well resolved by TLC in toluene/ethyl acetate/aceticacid (25:15:2, v/v/v), were eluted with H20-saturated ethylacetate and methylated with ethereal diazomethane and re-chromatographed by TLC in hexane/ethyl acetate ( 1:1, v/v).After reelution with ether, ABA methyl ester was reduced toan equal mixture of the 1',4'-trans and 1,4-cis diols usingNaBH4 in aqueous methanol at 0°C, which was then separatedby TLC in benzene/ethyl acetate/acetic acid (15:3:1, v/v/v).Likewise, PA methyl ester was reduced to an equal mixtureof dihydrophaseic acid and its 4'-epimer, following treatmentwith NaBH4, which was then resolved by TLC in hexane/ethyl acetate (1:1, v/v), and the plates were developed threetimes to 15 cm.

Determination of ABA and ,8-Carotene

To enable appropriate comparisons to be made it wasimportant to measure levels of ABA and (3-carotene in thesame extract. Once harvested, cells were resuspended in 50mL of fresh growth medium, containing 3.0 M NaCl, to givea culture density of 13 x 105 cells . mL-'. Cells were incubatedas described above. Cell cultures of the same density growingin 1.5 M NaCl medium were included as controls. At eachassay interval, aliquots were removed for the determinationof Chl (1) and dry weight. Cells were extracted for ABA and(3-carotene in ice-cold acetone (-20°C) containing 2,6-Di-tert-butyl-p-cresol (butylated hydroxytoluene, 20 mg * L-'), addedas an antioxidant, by sonication and stored at -20°C undernitrogen in the dark until analyzed.For analysis of 13-carotene, acetone extracts were reduced

to dryness, redissolved in 70% methanol, and applied to Sep-Pak C18 cartridges. Oxygenated carotenoids were eluted with5 mL 90% methanol, Chl with 5 mL 100% methanol, andcarotene with 5 mL 100% acetone as described by Eskins andDutton (10). The concentration of 1-carotene in the acetoneeluates was determined according to the method of Ben-Amotz and Avron (2).The levels of endogenous ABA were determined in similar

acetone extracts containing a known amount of high specificactivity [3H]ABA (added to correct for recovery) and ABA-methyl ester (Sigma), as an internal standard. Acetone wasremoved in vacuo at 35°C and the aqueous residue purifiedusing Sep-Pak C18 cartridges as described above. The incuba-tion media and washings were similarly purified followingacidification. Aliquots of these eluates were analyzed by re-verse-phase HPLC using a 0 to 100% methanol gradientcontaining 0.5% acetic acid throughout and quantified bypeak integration and calibration with reference compounds.The liquid chromatography apparatus consisted of a high-pressure ternary pump (Spectra-Physics SP 8800), a Rheodyne

799

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.

Plant Physiol. Vol. 97, 1991

0

011E0~co C~

MilM ABA A.8- FLPA

l15M 3OM

)n4- T

7kLEB.

15M

--w--n-I

08-

'T iE

co0-41-Cn

Figure 1. The synthesis and metabolism of ABA by cells of D. salinavar bardawil exposed to different salinities. Aliquots of cells (6.0 mL,equivalent to 0.145 mg Chi * mL-1) at a culture density of 7 x 106cells . mL-1 were supplied R-[2-14C]MVAL (366.6 kBq) and incubatedfor 20 h at 250C before exposure to hypersaline conditions. Followingpreincubation in 1.5 M NaCI medium, cells were washed by centrifu-gation and resuspended in fresh medium containing either 1.5 M NaCI(for control cells) or 3.0 M NaCI (stressed cells) and incubated for anadditional 4 h at 250C. Cells were separated from the medium bycentrifugation, and [14C]ABA and [14C]PA were extracted and ana-lyzed as described in "Materials and Methods." A, Cells; B, culturemedium. Data are expressed as the means of three replicates ± SD.

7125 injector (20-,uL loop), and a 5-,gm ODS Spherisorbcolumn (4.6 x 100 mm i.d.). Flow was 1.0 mL * min-', andpeaks were monitored at 254 nm with a linear UV- 106 fixedwavelength detector coupled to a Spectra-Physics SP 4290integrator. The fractions corresponding to authentic ABA andPA were pooled, reduced to dryness, and methylated withexcess ethereal diazomethane. Unequivocal identification ofthese products as ABA and PA was established by combinedcapillary GC-MS.GC-MS was performed on a Hewlett-Packard 5890 instru-

ment, using a fused-silica capillary column (12 m x 0.32 mmi.d.) of HP-1 programmed from 120C at 5oC * min-' withHe as the carrier gas (1.5-2 mL * min-'), coupled to an HP5988A MS system. Electron impact mass spectra were re-corded at 70 eV and a source temperature of 250°C. Identifi-cation of the compounds as methylated ABA and methylatedPA was established using an Hewlett-Packard data processingstation.

RESULTSABA Metabolism in Salt-Stressed Cells

Analysis of the Sep-Pak C18-purified acids from bothstressed and nonstressed cells supplied with R-[2-'4C]MVAL,

3OM

Tl

and the respective incubation media, resulted in the distri-bution of radioactivity shown in Figure 1. The data from thisexperiment allow several observations to be made. First, cellsof D. salina are able to transform labeled MVAL into ABAand PA, suggesting that both the ABA anabolic and catabolicpathways are operational in this organism. Second, bothprocesses are enhanced by hypersaline conditions. Third, cellsdo not appear to accumulate appreciable amounts of eitherABA or PA in response to salt stress. Fourth, any ABA andPA produced is preferentially partitioned into the culturemedium, and this partitioning effect increases in response tosalt stress. Thus, the overall effect of hypersalinity stress onABA metabolism in cells of D. salina would appear to beenhancement, manifested by accumulation of ABA and PAin the culture medium. To investigate this aspect in moredetail the rate of appearance of ABA in the culture mediumwas determined. Furthermore, a kinetic study was carried outto determine the distribution of labeled ABA between cellsand the incubation medium.The results presented in Figure 2 show that the rate of

appearance of labeled ABA in the incubation medium in-creased dramatically within 30 min of exposure to salinitystress. Thereafter, the rate of appearance of ABA declined tolevels similar to those observed in the control. Detailed kineticstudies (Fig. 3) revealed that with time the amount of labeledABA in the incubation medium increased and that the rateof appearance was similar to that observed in Figure 2. Incontrast, levels of radioactive ABA associated with the cellsdid not change significantly during the course of incubation.However, a slight increase was observed immediately after

04

V,-.

E 0 2co

0-

o 1 5M NaCl* 3OM NaCt

6 3Time after transition (h)

Figure 2. The rate of appearance of labeled ABA in the culturemedium from cells of D. salina var bardawil following exposure tosalinity stress. Aliquots of cells (4.0 mL, equivalent to 0.25 mg Chl -

mL-1) at a culture density of 9 x 1 o6 cells . mL-' were supplied withR-[2-14C]MVAL (366.6 kBq) for 20 h and then transferred to freshmedium containing either 1.5 or 3.0 M NaCI for 3 h. Cells wereremoved by centrifugation, and the amount of [14C]ABA present inthe culture medium was determined as described in the "Materialsand Methods." Rates were calculated by subtracting the initial [14C]ABA amount from the amount of [14C]ABA at the end time point ofeach interval. This value was expressed as a function of time, in h,and plotted against the time point of each assay interval. Conditionsof incubation and analysis were as described in Figure 1. Data arerepresentative of two independent experiments.

COWAN AND ROSE800

v

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.

ABSCISIC ACID METABOLISM IN DUNALIELLA SALINA

Table I. Effect of Inhibitors of Translation and Transcription on theIncorporation of Label from either [2-'4C]MVAL or NaH[14CJ03 intoABA by Cells of D. salina Exposed to Salinity Stress

Aliquots of the algal culture (6.0 mL, equivalent to 42 x 1 06 cells)were supplied with either R-[2-'4C]MVAL or NaH[14C]O3 (both 366.6kBq) and incubated at 250C under continuous low illumination (42zmoI . m-2* S-1). Inhibitors of transcription and translation were appliedat the appropriate times as described in "Materials and Methods,"after which cells were exposed to increased salinity for 4 h (from 1.5to 3.0 M NaCI medium) and ABA was extracted and analyzed asdescribed in Figure 1

Treatment NaH['4C]03 [2-14C]MVALBq-mg' Chl (%)a

Control 386.2 (0) 428.7 (0)Actinomycin D (50 gg-mL-1) 164.4 (57.4) 166.1 (61.3)Cycloheximide (0.2 Ag -mL-1) 185.9 (62.2) 151.1 (64.8)Chloramphenicol (50 Ag - mL-') 171.8 (55.5) 194.3 (54.7)

a Percentage inhibition relative to control.

0-06

0*4

-I_Ecrco

-0

02

E

-0-1 &co

-O

0 2 4Time after transition (h)

Figure 3. Time course of ABA accumulation and rate of productionin cells (0) and culture medium (0) following exposure of D. salinavar bardawil to salinity stress. Application of R-[2-14C]MVAL andincubation conditions were as described in Figure 1. A, Distributionof ['4C]ABA between cells and culture medium during the first 4 hafter exposure; B, rate of [14C]ABA production during the 4-h stressperiod. Appearance rates were calculated as described in Figure 2.Data are representative of two independent experiments.

the imposition of hypersaline conditions. This increase ap-peared to suggest an alteration in the levels and/or activitiesof the ABA-biosynthesizing enzymes induced by hypersalineconditions. To elucidate this aspect further, transcriptional(actinomycin D) and translational (chloramphenicol and cy-cloheximide) inhibitors were added at the appropriate timeand at concentrations known to inhibit the induction of f-

carotene accumulation (13).Hypersalinity induction of labeled ABA production was

inhibited by actinomycin D, cycloheximide, and chloram-phenicol. These data therefore indicate that both transcriptionand de novo protein synthesis are required for the increasedproduction of ABA induced by hypersalinity. Furthermore,the results in Table I show that incorporation of label intoABA was similar for cells supplied with either NaH['4C]03 or

R-[2-'4C]MVAL. This coupled with the inhibition of ABAbiosynthesis by chloramphenicol strongly suggests that plas-

tid-localized enzymes are involved in the induction of ABAproduction by cells of D. salina exposed to hypersalineconditions.

Kinetics of ABA and fl-Carotene Accumulation

The results presented in Figure 4 show that in response tostress ,B-carotene content of cells of D. salina declined within6 h but had returned to basal levels 24 h after the impositionof stress. Thereafter, the concentration of fl-carotene in non-

stressed cultures remained constant, whereas fl-carotene instressed cells began to accumulate by day 3. The early declinein fl-carotene content immediately following the impositionof stress might reflect a stress situation caused by centrifuga-tion during transfer of cells. However, the decline in -

t,120

cmE

60

C: 60-0

C-

L-i

0

6 i 3 4Time after transition (d)

Figure 4. Kinetics of p3-carotene accumulation by cultures of D. salinaexposed to increased salinity. After harvest, 50-mL aliquots of cellsat a culture density of 13 x 105 cells. mL-1 were incubated in either1.5 (0) or 3.0 M (0) NaCI medium at 25°C. d3-Carotene content was

determined after extraction of cells in acetone, and data representthe means of three replicates ± SD.

A.

B.

04

Li

E_04

.w0.

ol

0-

.02-L-i

E

coi

0.I I~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1.....

801

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.

Plant Physiol. Vol. 97, 1991

X;

4-

0-

0 1 2 3Time after transition (d)

Figure 5. Kinetics of ABA accumulation by cells of D. salina exposedto increased salinity. ABA levels were monitored following transfer ofcells to fresh incubation medium containing either 1.5 (0) or 3.0 M (0)NaCI. Conditions of incubation were as described for Figure 4. Datarepresent the means of three replicates ± SD.

carotene content was less marked in control cells, suggestingthat this response was, at least in part, due to increasedsalinity.Oxygenation of carotenes yields xanthophylls such as vio-

laxanthin and neoxanthin, the postulated precursors of ABA(22). No change in bulk oxo-carotenoid content was detectedin cultures exposed to either 1.5 or 3.0 M NaCl (data notshown). However, a significant increase in levels of ABA wasdetected in cultures of D. salina exposed to 3.0 M NaCl (Fig.5). ABA levels reached a maximum 8 h after exposure tohypersaline conditions and then declined to a basal level of 3ng * mg-' dry weight. A similar but less dramatic responsewas detected in control cultures at 1.5 M NaCl, again, sug-

401

'EJcm

20

01

d i 4 --

4 5~Time after transition (d)

Figure 6. Course of ABA accumulation in the culture medium fromcells of D. salina exposed to increased salinity as described for Figure5. Cells were transferred to fresh medium (see Fig. 4) containingeither 1.5 (0) or 3.0 M (0) NaCI, and levels of ABA were monitoredduring a 5-d period as described in "Materials and Methods." Dataare expressed as the means of three replicates + SD.

1 2 3 4Time after transition (d)

Figure 7. The rate of ,B-carotene and ABA accumulation by cells ofD. salina exposed to hypersaline conditions. Rate of accumulationwas calculated from each time point (Figs. 4 and 5) by subtractingthe initial concentration from the final concentration divided by thetime taken to reach the final concentration. These values were plottedagainst the end time point of each time interval. Data are expressedas the means of three replicates ± SD.

gesting a stress condition imposed by centrifugation and celltransfer. Nevertheless, the massive accumulation of ABA incells exposed to hypersaline conditions strongly suggests thatABA production increased in response to salt stress. More-over, the ABA content of the medium from cells incubatedat 3.0 M NaCl increased throughout the incubation period,whereas ABA levels in the medium from control cells did notchange significantly (Fig. 6). These data indicate that underconditions of hypersalinity there is an immediate response interms of increased ABA production and that the bulk of thisABA is partitioned into the culture medium.

In attempting to elucidate the interrelationship between ,B-carotene accumulation and ABA production, we calculatedthe rate of change in both A-carotene and ABA production.The result shown in Figure 7 indicates two stages of acceler-ated fl-carotene accumulation and that the initial stage ofincreased f-carotene production coincides with the only stageof accelerated ABA production.

DISCUSSION

Results from the present investigation indicate that cells ofD. salina accumulate fl-carotene in two stages in response tohypersaline conditions. The first stage commences immedi-ately after exposure to hypersalinity and coincides with in-creased endogenous ABA levels. This stage lasts for approxi-mately 24 h, after which fl-carotene content remains constantuntil the second stage ofaccumulation begins. By comparison,cellular levels of ABA remained constant, whereas levels ofABA in the culture medium increased almost linearly overthe remainder of the 5-d incubation period. These data are inkeeping with earlier reports concerning the effects of salinityand high-light intensity on f-carotene accumulation by cells

I I

802 COWAN AND ROSE

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.

ABSCISIC ACID METABOLISM IN DUNALIELLA SALINA

of D. salina (7, 13) and suggest an interrelationship between,B-carotene and ABA production.Evidence has been obtained to suggest that ABA is derived

from the oxo-carotenoid, 9-cis-neoxanthin (16). The overpro-duction of carotenoids by cells of D. salina in response tostress (2, 3, 6, 7, 13) might therefore be regulated, at least inpart, by the synthesis ofABA and the subsequent partitioningof ABA and some of its metabolic products into the culturemedium. In higher plant tissues, ABA has been shown toinduce its own conversion to PA (21). The results presentedin this investigation show that amounts of both PA and ABAisolated from the culture medium of salt-stressed cells werehigher than in the medium from nonstressed cells. Thisobservation supports the enhanced catabolism of ABA inresponse to stress.

Results from studies using labeled precursors and inhibitorsof both transcription and translation suggest that the induc-tion of ABA production occurs by de novo synthesis. Simi-larly, the induction of ,B-carotene accumulation in responseto stress is known to occur by de novo synthesis (13). Fur-thermore, several stress-induced proteins have been isolatedwhich could fulfill a carotenogenic function (19), possibly byinfluencing the metabolism ofABA. In addition, norflurazon,which is known to inhibit ABA biosynthesis in higher plants(9), also prevents the accumulation off-carotene in Dunaliella(5, 6), further supporting an interrelationship between ABAproduction and (3-carotene accumulation in cells of D. salina.It has been suggested that the early stage of (-carotene accu-mulation is mediated by the activation of a gene(s) encodingan enzyme(s) responsible for ,B-carotene synthesis (13). Resultsfrom the present study suggest that enhanced ABA productionand catabolism during the early stage of ,B-carotene accumu-lation might shift the chemical equilibrium to favor carbonflux through the isoprenoid pathway, the end result beingenhanced carotenoid biosynthesis, possibly mediated by geneactivation and/or alterations in enzyme synthesis and/oractivity.The second stage of f-carotene accumulation may result

from continued ABA synthesis, ABA catabolism, and subse-quent partitioning of ABA and its acidic products into theculture medium. Cells do not appear to accumulate significantquantities of ABA during this stage, and label from supplied(R,S)-[2-'4C]ABA was not incorporated by cells of D. salina(data not shown). This accumulated information stronglysuggests an interrelationship between ABA production by,and fl-carotene accumulation in, salt-stressed cells of D. sal-ina. Thus, increased production of ABA and the enhancedconversion of ABA to PA and subsequent release of thesecompounds into the culture medium could function to initiatethe accumulation of carotenoids, in particular, fl-carotene.Studies are currently underway to determine (a) whether thisresponse is similar for other stress stimuli and (b) to quantifychanges in the levels of all major photosynthetic pigments inrelation to changes in ABA levels. It is hoped that thesestudies will further substantiate the possible interrelationshipbetween the production of ABA and the onset of fl-caroteneaccumulation in cells exposed to conditions of stress.

ACKNOWLEDGMENTSThe authors wish to thank the Joint Research Council (Rhodes

University) for funding this research project. Mr. Aubrey Sonnemanis thanked for expert handling of the mass spectrometer.

LITERATURE CITED

1. Arnon DI (1949) Copper enzymes in isolated chloroplasts: poly-phenoloxidase in Beta vulgaris. Plant Physiol 24: 1-15

2. Ben-Amotz A, Avron M (1983) On factors which determinemassive ,B-carotene accumulation in the halotolerant alga Dun-aliella bardawil. Plant Physiol 72: 593-597

3. Ben-Amotz A, Avron M (1989) The wavelength dependence ofmassive carotene synthesis in Dunaliella bardawil (Chlorophy-ceae). J Phycol 25: 175-178

4. Ben-Amotz A, Avron M (1990) The biotechnology of cultivatingthe halotolerant alga Dunaliella. Trend Biotechnol 8: 121-126

5. Ben-Amotz A, Gressel J, Avron M (1990) Massive accumulationof phytoene induced by norflurazon in Dunaliella bardawil(Chlorophyceae) prevents recovery from photoinhibition. JPhycol 23: 176-181

6. Ben-Amotz A, Lers A, Avron M (1988) Stereoisomers of f-carotene and phytoene in the alga Dunaliella bardawil. PlantPhysiol 86: 1286-1291

7. Borowitska MA, Borowitska LJ, Kessly D (1990) Effects ofsalinity increase on carotenoid accumulation in the green algaDunaliella salina. J Appl Phycol 2: 111-119

8. Cowan AK, Railton ID (1987) The catabolism of (±)-abscisic acidby excised leaves of Hordeum vulgare L. cv Dyan and itsmodification by chemical and environmental factors. PlantPhysiol 84: 157-163

9. Creelman RA (1989) Abscisic acid physiology and biosynthesisin higher plants. Physiol Plant 75: 131-136

10. Eskins K, Dutton HJ (1979) Sample preparation for high-per-formance liquid chromatography of higher plant pigments.Anal Chem 51: 1885-1886

11. Hirsch R, Hartung W, Gimmler H (1989) Abscisic acid contentof algae under stress. Bot Acta 41: 21-53

12. Kirst GO (1990) Salinity tolerance of eukaryotic marine algae.Annu Rev Plant Physiol Plant Mol Biol 41: 21-53

13. Lers A, Biener Y, Zamir A (1990) Photoinduction of massive ,B-carotene accumulation by the alga Dunaliella bardawil. Kinet-ics and dependence on gene activation. Plant Physiol 93:389-395

14. Li Y, Walton DC (1990) Violaxanthin is an ABA precursorin water-stressed dark-grown bean leaves. Plant Physiol 92:551-559

15. Milborrow BV (1974) Biosynthesis of abscisic acid by a cell-freesystem. Phytochemistry 13 131-136

16. Parry AD, Babiano MJ, Horgan R (1990) The role of cis-carotenoids in abscisic acid biosynthesis. Planta 182: 118-128

17. Pierce M, Raschke K (1981) Synthesis and metabolism of ABAin detached leaves of Phaseolus vulgaris L. after loss andrecovery of turgor. Planta 153: 156-165

18. Rock CD, Zeevaart JAD (1990) Abscisic (ABA)-aldehyde is aprecursor to, and 1 ',4'-trans-ABA-diol a catabolite of, ABA inapple. Plant Physiol 93: 915-923

19. Sadka A, Lers A, Zamir A, Avron M (1988) A critical examina-tion of the role of de novo protein synthesis in the osmoticadaption of the halotolerant alga Dunaliella. FEBS Lett 244:93-98

20. Tietz A, Ruttkowski V, Kohler R, Kasprik W (1989) Furtherinvestigations on the occurrence and the effects of abscisic acidin algae. Biochem Physiol Pflanzen 184: 259-266

21. Uknes SJ, Ho T-HD (1984) Mode of action of abscisic acid inbarley aleurone layers. (Abscisic acid induces its own conver-sion to phaseic acid). Plant Physiol 75: 1126-1132

22. Zeevaart JAD, Creelman RA (1989) Metabolism and physiologyof abscisic acid. Annu Rev Plant Physiol Plant Mol Biol 39:439-473

23. Zeevaart JAD, Milborrow BV (1976) Metabolism ofabscisic acidin Phaseolus vulgaris. Phytochemistry 15: 493-500

24. Zeevaart JAD, Heath TG, Gage DA (1989) Evidence for auniversal pathway of abscisic acid in higher plants from 18Oincorporation patterns. Plant Physiol 91: 1594-1601

803

www.plantphysiol.orgon July 26, 2019 - Published by Downloaded from Copyright © 1991 American Society of Plant Biologists. All rights reserved.