Stress Responses in Alfalfa (Medicago sativa L.)

0 downloads 0 Views 1MB Size Report
Stress Responses in Alfalfa (Medicago sativa L.) VI. Differential Responsiveness of ..... Proc Natl Acad Sci USA 82: 5824-5828. 10. Kessmann H, Choudhary AD, ...
Received for publication May 29, 1990 Accepted August 23, 1990

Plant Physiol. (1990) 94, 1802-1807 0032-0889/90/94/1 802/06/$01 .00/0

Stress Responses in Alfalfa (Medicago sativa L.) VI. Differential Responsiveness of Chalcone Synthase Induction to Fungal Elicitor or Glutathione in Electroporated Protoplasts Arvind D. Choudhary', Christopher J. Lamb, and Richard A. Dixon* Plant Biology Division, The Samuel Roberts Noble Foundation, P. 0. Box 2180, Ardmore, Oklahoma 73402, (A.D.C., R.A.D.); and Plant Biology Laboratory, Salk Institute for Biological Studies, La Jolla, California 92037 (C.J.L.)

quantitatively major set of transcripts induced by GSH in cultured bean cells (13) and pea epicotyl tissue (14) encodes CHS2, a key regulatory enzyme in the biosynthesis of isoflavonoid phytoalexins in legume cells. CHS enzyme activity is also strongly induced by GSH in suspension cultured cells of the cactus Cephalocereus senilis (P. Paray, personal communication). CHS is encoded by a family of at least six genes in bean (12). Cis-acting elements involved in regulating the quantitative expression of a bean CHS promoter in response to GSH have recently been identified by means of transient expression assays in electroporated soybean protoplasts (7). We have shown that expression of a chimeric construct containing the promoter of the bean CHS 15 gene (12) linked to a CAT reporter gene in electroporated alfalfa protoplasts is induced on exposure of the protoplasts to GSH (1). Using this system and measurement of endogenous alfalfa CHS activity, we now report differences associated with the induction of CHS expression by GSH and fungal elicitor related to (a) culture age and (b) the presence of cis-acting promoter elements co-electroporated into the alfalfa protoplasts. Our data suggest that the signal transduction pathways for elicitation by fungal elicitor and GSH differ, both in terms of initial sites of action of the two elicitors and in relation to subsequent events associated with transcriptional activation.

ABSTRACT Protoplasts derived from cell suspensions of alfalfa (Medcago sativa L.) responded to treatment with fungal elicitor (FE) by an increase in endogenous chalcone synthase (CHS) activity but were unresponsive to reduced glutathione (GSH). Preexposure of protoplasts to polyethylene glycol and electroporation resulted in strong responsiveness to GSH but little change in responsiveness to FE. Protoplasts from suspension cultures which had been subcultured more than 12 times lost responsiveness to GSH, but not FE, as assessed by measuring expression of a chimeric gene containing a bean CHS promoter linked to a bacterial chloramphenicol acetyltransferase (CAT) reporter gene. In protoplasts in which putative cis-acting CHS promoter sequences had been coelectroporated in trans with the intact CHS promoter-CAT construct, the extent of CAT expression depended upon the elicitor used (FE or GSH), the age (number of times subcultured) of the cells from which the protoplasts were isolated, and the nature of the coelectroporated CHS promoter sequence. For example, a region of the CHS promoter from -326 to -141 behaved as a trans-activator when coelectroporated with the CAT construct into unelicited protoplasts isolated from newly initiated cell suspensions, but the same region acted as a trans-silencer in the same protoplasts in the presence of FE. This silencer activity was much reduced in GSH-treated protoplasts. The results suggest that there are differences in the signal transduction pathways for elicitation of CHS transcription by FE and GSH, which involve previously identified cis-elements in the CHS promoter.

MATERIALS AND METHODS DNA Constructs

The structures of pCHCl, pMaeI-1, and pCHPl are outlined in Figure 1. Details of functional elements within the CHS 15 promoter are given in "Results" and "Discussion." Construction of the plasmids will be described elsewhere (MJ Harrison, AD Choudhary, I Dubery, CJ Lamb, RA Dixon, manuscript in preparation).

A wide range of structurally diverse biotic and abiotic agents (elicitors) have been shown to induce the expression ofdefense metabolism in plants (3, 4). This raises the obvious question of whether they act by a common mechanism. Of particular interest is the observation that reduced GSH acts as a strong elicitor of defense gene activation in intact suspension cells and protoplasts of bean and soybean (7, 13), selectively inducing gene expression in a manner similar to that observed in response to fungal elicitor preparations from the cell walls of the bean pathogen Colletotrichum lindemuthianum. A 'A.D.C. acknowledges financial support from

a

Isolation and Electroporation of Protoplasts Rapidly growing cell suspension cultures of alfalfa (Medicago sativa L. cv Calwest 475) were maintained as described

2Abbreviations: CHS, chalcone synthase (EC 2.3.1.74); CAT, chloramphenicol acetyl transferase; FE, fungal elicitor from the cell walls of C. lindemuthianum.

Government of

India Scholarship for Study Abroad. Present address: Department of

Botany, Nagpur University Campus, Nagpur 440010, India.

1802

CHS INDUCTION BY GLUTATHIONE AND FUNGAL ELICITOR

A.

CHS15 promoter Mael

-332 Hinfl -326

B.

1

CHS15promoter

1

pUC19

50 bp

-141

B R -

T FAi

-326

-29

measuring

.

NOS3'. Y

_

.

D.

pMael-

plBI

11

(Mael)

III

r_-,

-24

I

I7BI

-

24

GATCACCMTTATTGGTTACTAAATTTAACAGT

TGGTTAATAACCAATGATTTAAATGTCACTAG B

E.

pCHP I

the incorporation of [2-14C]malonyl as

coenzyme

A

previously de-

Statistical Analysis of Results

.

+1

1

were

R KR

B R CAT

+120

-326

at

into naringenin in crude buffered extracts scribed (2).

HfX;

1 Hf pCHC I

+179

~ Mael

1I

11

H3

C.

incubation in the dark

-141

Hindlill IMael-l~ |g -332

25°C, protoplasts harvested by centrifugation, frozen in liquid N2, and stored at -70°C. Assay of CAT activity in cell-free extracts from the protoplasts was as described by Fromm et al. (9). CHS was assayed by

ATG

TATA

HindlIl

1 803

pSP65

pSP65

Figure 1. Bean CHS1 5 promoter (A), the Mael-1 fragment of the promoter (B), and the constructs used in the present work (C-E). I, II, and IlIl are the three nuclear protein binding sites within the Mael1 fragment (6, 9). pCHP1 (E) contains four copies of the box III sequence in head-to-tail orientation in plasmid pSP65. B, BamHI; H3, HindlIl; Hf, Hinfl; K, Kpnl; X, Xbal. Note that only A and B are to scale.

previously (1). They were harvested on the fifth day after subculture for preparation of protoplasts and electroporation of DNA constructs using a BTX Transfector 300 unit (BTX, San Diego, CA) as described (1). Electroporation cuvettes contained 500 ,tL of protoplasts (2 x 107 mL-') and, where necessary, 50 Atg pCHC 1 plus 50 ,g carrier calf thymus DNA. In coelectroporation experiments, calf thymus DNA was omitted, and 50 ,g pMaeI-l or pCHPl were added to the cuvette.

Elicitation and Analysis of Protoplasts

Colletotrichum lindemuthianum cell wall elicitor (50 Ag glucose equivalents mL-' final concentration) (5) or GSH (adjusted to pH 5.8, 0.5 mm final concentration) were added to protoplasts 2 h after electroporation. After a further 9 h

The variability of CAT values from replicate electroporations of the same construct into aliquots from the same batch of protoplasts was shown to be less than ±5% (1). However, absolute values from the same electroporated construct could vary by up to a factor of 10 on electroporation of protoplast batches prepared at different times. To correct for this and to allow independent replications of experiments, pCHCI was always electroporated alone into each protoplast batch as an internal control. Its value (in unelicited protoplasts) was then taken as 100%, and results with all other treatments were expressed relative to this value. At minimum, two independent electroporations with different protoplast batches were performed. In cases where the changes observed relative to expression of pCHC1 were small, three or four independent electroporations were performed. Data are expressed as the mean ± the standard deviation of the values relative to expression of pCHCI in the unelicited protoplasts included in each experiment. Significance of differences between treatments was determined by the Friedman two-way analysis of variance for matched samples (8). RESULTS AND DISCUSSION Responsiveness of Alfalfa Protoplasts to GSH

Suspension cultured alfalfa cells are unexpectedly not elicited on exposure to GSH (2), unlike bean, soybean, and Cephalocereus senilis cells. Thus, endogenous CHS activity was not induced when suspension cell protoplasts were exposed to 0.5 mm GSH (Table I), although the protoplasts responded, with a twofold increase in CHS activity, to FE. The concentrations of FE and GSH used were optimal for elicitation of bean cell suspension cultures and alfalfa protoplasts, as assessed by measurement of endogenous CHS activity. Treatment of the protoplasts with polyethylene glycol followed by electroporation in the presence of carrier calf

Table I. Elicitation of Alfalfa Suspension Cell Protoplasts by GSH Occurs after Pretreatment with Polyethylene Glycol and is Further Enhanced by Electroporation CHS Activitya Treatment of Protoplasts

+GSH

-GSH

Exp. 1

Exp. 2

Exp. 1

+FE

Exp. 2

Exp. 2

nkat/kg protein 2.2 2.7 7.11 None (Control) 6.88 12.94 NDc 9.4 +PEGb 4.3 ND ND +PEG + electroporationd 4.3 11.0 18.4 57.05 15.94 a b Final concentration 11.5% Reproducibility of the CHS assay in protoplasts was ±17% (n = 6). d BTX Transfector 300, 150 V, 50 I,F, 26 ms. (w/v). c Not determined.

CHOUDHARY ET AL.

1 804

thymus DNA and 50 ,g of plasmid pCHC 1 (Fig. lA) led to an approximately twofold increase in endogenous CHS activity in the absence of elicitor, no significant increase in responsiveness to FE, but a seven- to eightfold increase in response to GSH (Table I). The nucleotide sequence of the electroporated DNA was not critical for allowing responsiveness to GSH as identical effects were observed with plasmid pSP65 instead of pCHC 1. This effect of electroporation explains why we had previously observed the GSH-dependent induction of CAT expression from the bean CHS promoter-CAT-NOS 3' construct pCHC 1 (Fig. 1) in electroporated alfalfa protoplasts (1). The requirement for treatments that affect membrane permeability to obtain elicitation of alfalfa cells by GSH suggests the possibilities that either (a) internalization or metabolism of exogenously supplied GSH is a prerequisite for its elicitor activity, or (b) a secondary factor, produced or released as a result of membrane disruption or permeability changes, acts as a synergist for the action of GSH, but not FE, in alfalfa cells. The former interpretation is supported by recent work which has shown that the differential responsiveness of bean and alfalfa cell suspension cultures to GSH correlates with a much slower rate of uptake into soluble pools and a different metabolic fate of the exogenously applied tripeptide in alfalfa as compared with bean (R Edwards, JW Blount, R Dixon unpublished results). In experiments in which induction of CHS was monitored by measuring expression of the CAT reporter gene, driven by a bean CHS promoter in the electroporated construct pCHC 1, the responsiveness of alfalfa protoplasts to GSH declined as a function of the age of the cell suspension cultures from which the' protoplasts were isolated (Table II). When different batches of protoplasts from cultures grown for up to approximately eight serial subcultures after initiation from callus ("new" cultures) were electroporated with pCHC 1, basal CAT expression in the absence of elicitor 9 h after electroporation varied from around 300 to 3000 dpm in CAT reaction products per 107 protoplasts in the standard assay. Irrespective of the absolute level of basal expression, expression from the chimeric construct was weakly induced in response to FE, by a factor of 1.35, but GSH induced nearly a 20-fold increase in CAT expression. In contrast, protoplasts from older suspension cultures (greater than 12 passages after initiation, "old" cultures), while exhibiting similar variation in basal CAT expression and retaining the same weak inducibility by FE, supported only a 1.9-fold induction of CAT expression by GSH. A similar differential responsiveness to GSH between

Table II. Alfalfa Protoplasts Loose Responsiveness to GSH as a Function of the Number of Passages of the Suspension Culture from which They were Isolated CAT Induction from pCHC1 Relative to Unelicited Elicitor

None Fungal elicitor

GSH aSee text.

Control in Protoplasts from New culturesa Old culturesa

1

1

1.35 ± 0.134 (n = 3) 18.45 ± 2.98 (n = 3)

1.50 ± 0.10 (n = 4) 1.90 ± 0.50 (n = 3)

Plant Physiol. Vol. 94, 1990

new and old cultures was observed when preelectroporated protoplasts were assayed for endogenous CHS activity (data not shown). The loss of responsiveness to GSH was observed reproducibly on repeated subculture of various lines of the same alfalfa cultivar. In view of the retention of normal inducibility of the CHS promoter by FE, this effect must be the result of a change affecting an event peculiar to the action of GSH, perhaps relating to the way in which the tripeptide is initially metabolized on contact with the protoplasts. It should be noted that the inducibility of the protoplasts by FE, although not changing as a function of culture age, is low compared to the eightfold induction of endogenous CHS activity observed in the suspension cultured cells from which the protoplasts are isolated (10). The loss of responsiveness to GSH cannot be associated with changes in elicitor-responsive elements within the CHS promoter itself, as this was introduced on an electroporated plasmid. We cannot rule out the possibility that it involves a loss of GSH uptake or internalization, as we did not test whether different electroporation conditions might result in responsiveness of the protoplasts from old cultures. Whatever the nature of the loss of responsiveness to GSH, our data indicate that the initial signals occurring in response to GSH and FE are different, a conclusion not surprising in view of their very different chemical natures. Effects of Coelectroporation of CHS Promoter Elements on Inducibility of a Chimeric Gene by GSH and FE The promoter region in pCHC1 extends from -326 bp upstream of the transcription initiation site of the bean CHS 15 gene to a transcriptional fusion with the CAT reporter gene at around + 120 bp (Fig. 1). A number of cis-acting regulatory elements lie between positions + 1 and -326. The region between -326 and the MaeI site at -141 contains three binding sites (boxes 1-111) for a bean nuclear protein(s), as identified by in vitro footprinting (6, 1 1). Collectively, these binding sites have been functionally defined as controlling quantitatively the expression of the promoter in both elicited and unelicited protoplasts of soybean (7) and alfalfa and being involved in determining the tissue-specific level of expression of the promoter in transgenic tobacco (J Kooter, unpublished results). Deletion to -173 results in decreased activity of the CHS 15 promoter in alfalfa protoplasts, suggesting an overall enhancer function for the cis-acting elements (MJ Harrison, AD Choudhary, I Dubery, CJ Lamb, RA Dixon, manuscript in preparation). In vitro DNase I footprinting with alfalfa nuclear extracts has revealed four protein binding sites within the -326 to -140 region, three corresponding to boxes I-III (as defined by footprinting with bean extracts), the other 3' of the box III element (MJ Harrison, AD Choudhary, I Dubery, CJ Lamb, RA Dixon, manuscript in preparation). In order to examine whether the signal transduction pathways involved in response to GSH or FE are similar in relation to the involvement of the previously identified cis-acting elements in the bean CHS promoter, we studied the effects on expression of the electroporated pCHC 1 construct of coelectroporation of plasmids containing boxes I-III, or a multimer of box III alone, in unelicited protoplasts and in protoplasts exposed to FE or GSH (Fig. 2). Plasmid pMael- 1

CHS INDUCTION BY GLUTATHIONE AND FUNGAL ELICITOR

IN

B (plus fungal elicitor) ~I ~O

22

1 805

T

C (plus glutathione) O

N

20

18

L2.0_

16

2.0

Xr,X 0

14

12

0

c)

10

~~~(4)

Er

1.0

(4

8

(4)

6 4

1.t

2

pCHC1 pMaeI-1

pCHP1

+

-

+ + -

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+ -

-

-

+ -

-

+

+

+

-

-

+ -

-

+

Figure 2. Effects of coelectroporation of pMael-1 or pCHP1 on the expression of pCHC1 in electroporated alfalfa protoplasts isolated from new (N) or old (0) suspension cultures (see text). Protoplasts were unelicited (A), exposed to cell wall elicitor from C. lindemuthianum (50 igg glucose equivalents mL-1) (B), or exposed to 0.5 mm GSH (C). Prior to elicitation, protoplasts were electroporated with 50 gg pCHC1 in the presence or absence of 50 4g of pMael-1 or pCHP1. Protoplasts were harvested for determination of CAT activity 9 h after exposure to elicitor. Data are expressed relative to an internal control (CAT activity from pCHC1 in unelicited protoplasts from the same batch) (arbitrary value 1.0, arrowhead in [C]). Bars represent standard deviations of values from the number of replicate independent electroporations shown in brackets on each bar graph. The values for pCHC1 in unelicited cells have no error bars, representing the normalized internal standard for four independent replicate experiments. In A (0), B (N and 0), and G (N), the sets of three treatments being compared show statistically significant differences at P = 0.028. Each replicate electroporation used a separate protoplast batch. Reproducibility for the same treatment within the same protoplast batch was ±5%.

contains the region from -326 to -141 in the vector pIBI.24; pCHPI contains a 4-mer of the box III sequence (-242 to -214) in head-to-tail orientation in the vector pSP65 (Fig. 1). Coelectroporation experiments were performed in protoplasts isolated from the old and new suspension cultures, which exhibited differential responsiveness to GSH. A striking difference was observed in the effects of coelectroporation of pMaeI- 1 in unelicited protoplasts from the two types of cell culture (Fig. 2A). Coelectroporation of pMael- 1 increased by twofold the expression from the CHS promoter-CAT construct in protoplasts from newly initiated cultures but depressed it by twofold in protoplasts from cultures which had been through greater than 12 serial subcultures. Thus, in the new protoplasts in the absence of exogenous elicitors, the coelectroporated region from -326 to -141 acted as a transactivator, whereas in the old cultures it acted as a transsilencer. These data suggest that the corresponding sequences in the intact promoter act as a cis-silencer in the new cultures and as a cis-activator in the old cultures, in the absence of elicitor. This most likely reflects different activities of transacting factors in protoplasts from cells of different ages. In contrast to the effects of pMael- 1, coelectroporation of pCHPl had no significant effect on CAT expression from

pCHC 1. pCHPI may exhibit little or no activity in this assay, because boxes I, II, and III together may be necessary for maximum binding affinity for trans-factors(s). Alternatively, the spacing of the multiple copies of the box III binding site in pCHP 1 may not be optimal (but see effects in GSH-treated protoplasts below). Whether the differences in behavior of the two cell lines on coelectroporation of the CHS-elements in the absence of elicitors is in any way related to their responsiveness to GSH is unclear. In the presence of FE (which induced a statistically significant 1.3- to 1.5-fold increase in CAT expression), coelectroporation of pMaeI- 1 decreased CAT expression in both types of protoplasts to below 50% of unelicited control values (Fig. 2B). In the protoplasts from the new cultures, the twofold increase in CAT expression in unelicited protoplasts as a result of coelectroporation of pMael- 1 was strikingly reversed to a twofold decrease in the presence of FE. A similar response, with pCHPI again having considerably less effect than pMael1, was seen with protoplasts from old cultures. Because of the experimental variability, further replications would be necessary to prove unequivocally that coelectroporation of pCHP 1 significantly inhibited expression from pCHC 1 in FE-treated cells, although analysis of the data in Figure 2A (old cultures)

1 806

CHOUDHARY ET AL.

and 2B (old and new cultures) by the Friedman two-way analysis of variance for matched samples (8) indicated that differences between pCHCl and pCHCl plus pCHPl were significant at the P = 0.028 level. The results in Figure 2, A and B, suggest that exposure to FE alters the function of the cis-elements within the MaeI-1 region of the promoter. The most likely explanation is that a factor(s) binding to these regions is modified as a result of exposure of the protoplasts to elicitor. The same factor(s) may therefore have a complex function as both repressor(s) and activator(s), dependent upon the status of the cells (e.g. elicited or unelicited). Very different responses were recorded in GSH-treated protoplasts (Fig. 2C). In GSH-responsive protoplasts, a 20fold induction of CAT expression from pCHC 1 by GSH was only partly inhibited by coelectroporation of pMaeI- 1; in these protoplasts, pCHP 1 appeared to act as a more effective transsilencer of CAT expression in the presence of GSH. In poorly responsive protoplasts from old suspension cultures, coelectroporation ofeither pMaeI- I or pCHP I resulted in a decrease of CAT expression to, but not below, levels observed in unelicited protoplasts. This response was again different from that seen with fungal elicitor. The loss of responsiveness to GSH in old cultures appears to have little effect on the behavior of pMael- 1 and pCHPl in FE-treated protoplasts (Fig. 2B), in contrast to the differences observed on coelectroporation of these plasmids into protoplasts from cultures of different age in the absence of elicitor. The above data are characterized by three striking observations. First, the presence of fungal elicitor reversed the effects of coelectroporation of pMael- 1 in GSH-responsive protoplasts, but this phenomenon was not observed with GSH as elicitor. Second, in protoplasts poorly responsive to GSH, (but with approximately equal responsiveness to GSH and FE), coelectroporation ofpMael-l reduced expression of CAT from pCHC 1 to less than 50% of unelicited control levels, if the cells were exposed to FE, but only reduced it to control levels in GSH-treated cells. Third, in GSH-responsive protoplasts in the presence of GSH, coelectroporation of a plasmid containing only the box III sequence of the MaeI- 1 fragment of the CHS 15 promoter (pCHPl) resulted in significantly greater inhibition of CAT expression than coelectroporation of the whole MaeI- 1 sequence. These effects contrasted strongly with the relative effects of pCHPl and pMael- 1 in FE-treated protoplasts; even if the effects of pCHPl were not significantly different from the effects of coelectroporation of pMael-1, pMael- 1 was considerably more effective in inhibiting CAT expression than is pCHP I in the presence of fungal elicitor. It could be argued that the data in Figure 2C for the GSHresponsive protoplasts are different from the corresponding data set in Figure 2B because of the much greater induction of CAT activity by GSH compared to FE, implying that the signal transduction pathway is "fully on" in these protoplasts in response to GSH, but only "partly on" (e.g. by a factor of about 15-fold less) in response to FE. This argument is complicated, however, by the up to 10-fold differences in absolute expression between the same treatments in different protoplast batches. The normalized data in Figure 2 (which record values relative to CAT expression from pCHC 1 in unelicited protoplasts) conceal the fact that absolute expression in re-

Plant Physiol. Vol. 94, 1990

sponse to FE in one experiment can be as great as 50% of the absolute expression in response to GSH in a separate experiment. The only safe conclusion that can be made is that GSH and FE differentially affect relative changes in CHS promoter expression in the presence of the co-electroporated sequences. Because pMael- 1 contains binding sites for alfalfa nuclear proteins, including the box III binding site contained in pCHPl (6; Fig. 1), the differences observed between FE and GSH may involve these factors. Indeed, recent studies indicate that the box III region protected from DNase I digestion by alfalfa nuclear proteins is somewhat larger (i.e. spans more nucleotides) than the region protected by bean nuclear proteins and contains both positive and negatively acting ciselements (MJ Harrison, AD Choudhary, I Dubery, CJ Lamb, RA Dixon, in preparation). This complexity might explain the different behavior of pCHPl when coelectroporated with pCHC 1 into protoplasts from new cultures treated with either FE or GSH. Results to date do not support a model whereby activation of the CHS gene by different elicitors occurs through different and independent cis-acting sequences peculiar to a particular class of elicitor (7, 9; MJ Harrison, AD Choudhary, I Dubery, CJ Lamb, RA Dixon, manuscript in preparation). The above data therefore provide evidence for the operation of different pathways, which nevertheless involve common trans-acting factors and cis-elements, for transducing the signals from GSH or FE.

CONCLUSION We conclude that the action of GSH as an elicitor of defense gene activation in alfalfa cells occurs via a signal pathway which is different from that involved in elicitation by a macromolecular fungal elicitor. Thus, GSH is a more effective elicitor than FE, but only with electroporated or PEG-treated protoplasts. In contrast, electroporation is not necessary for GSH to be active as an elicitor in cell cultures of a number of other plant species. Cells lose potential responsiveness to GSH, but not to FE, on repeated subculture in suspension. The presence of FE affects trans-modulation of a bean promoter by coelectroporated cis-acting sequences in a different manner from GSH. In spite of these differences, however, GSH and FE appear to induce very similar qualitative and quantitative changes in defense gene expression (1 1). These results suggest that different but convergent regulatory loops may have evolved for the activation of plant defense responses by different elicitor-active agents. ACKNOWLEDGMENTS We thank Maria Harrison for the construction of pCHPl, Robert Edwards for helpful discussions, and Scotty McGill for preparation of the manuscript. LITERATURE CITED 1. Choudhary AD, Kessmann H, Lamb CJ, Dixon RA (1990) Stress responses in alfalfa (Medicago sativa L.) IV. Expression of defense gene constructs in electroporated suspension cell protoplasts. Plant Cell Rep 9: 42-46 2. Dalkin K, Edwards R, Edington B, Dixon RA (1990) Stress responses in alfalfa (Medicago sativa L.) I. Induction of phenylpropanoid biosynthesis and hydrolytic enzymes in elicitortreated cell suspension cultures. Plant Physiol 92: 440-446

CHS INDUCTION BY GLUTATHIONE AND FUNGAL ELICITOR 3. Darvill AG, Albersheim P (1984) Phytoalexins and their elicitors-a defense against microbial infection in plants. Annu Rev Plant Physiol 35: 243-275 4. Dixon RA (1986) The phytoalexin response: elicitation, signalling and control of host gene expression. Biol Rev 61: 239-291 5. Dixon RA, Lamb CJ (1979) Stimulation of de novo synthesis of L-phenylalanine ammonia-lyase in relation to phytoalexin accumulation in Colletotrichum lindemuthianum elicitor-treated cell suspension cultures of French bean (Phaseolus vulgaris). Biochim Biophys Acta 586: 453-463 6. Dixon RA, Harrison MJ, Jenkins SM, Lamb CJ. Lawton MA, Yu L (1990) Cis-elements and trans-acting factors for regulation of the plant defense gene chalcone synthase. In CJ Lamb, R Beachy, eds, Plant Gene Transfer, Alan R Liss Inc, New York, pp 10 1-109 7. Dron M, Clouse SD, Lawton MA, Dixon RA, Lamb CJ (1988) Glutathione and fungal elicitor regulation of a plant defense gene promoter in electroporated protoplasts. Proc Natl Acad Sci USA 85: 6738-6742 8. Friedman H (1937) The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J Am Stat Assoc 32: 688-689

1807

9. Fromm M, Taylor LP, Walbot V (1985) Expression of genes transferred into monocot and dicot plant cells by electroporation. Proc Natl Acad Sci USA 82: 5824-5828 10. Kessmann H, Choudhary AD, Dixon RA (1990) Stress responses in alfalfa (Medicago sativa L.) IV. Induction of medicarpin and cytochrome P450 enzyme activities in elicitor-treated cell suspension cultures and protoplasts. Plant Cell Rep 9: 38-41 11. Lawton MA, Jenkins SM, Dron M, Kooter JM, Kragh K, Harrison MJ, Yu L, Tanguay L, Dixon RA, Lamb CJ (1990) Silencer region of a chalcone synthase promoter contains multiple binding sites for GT- 1. Plant Mol Biol (in press) 12. Ryder TB, Hedrick, SA, Bell JN, Liang XL, Clouse SD, Lamb CJ (1987) Organization and differential activation of a gene family encoding the plant defense enzyme chalcone synthase. Mol Gen Genet 210: 219-233 13. Wingate VPM, Lawton MA, Lamb CJ (1988) Glutathione causes a massive and selective induction of plant defense genes. Plant Physiol 87: 206-2 10 14. Yamada T, Hashimoto H, Shiraishi T, Oku H (1989) Suppression of pisatin, phenylalanine ammonia-lyase mRNA and chalcone synthase mRNA accumulation by a putative pathogenicity factor from the fungus Mycosphaerella pinodes. Mol Plant Microbe Interact 2: 256-261