Planta
Planta (1988) 175:9%106
9 Springer-Verlag 1988
Regulation of glutathione S-transferases of Zea mays in plants and cell cultures R. Edwards* and W.J. Owen Department of Biochemistry, Royal Holloway and Bedford New College, Egham Hill, Egham, Surrey TW20 OEX, U.K.
Abstract. An antiserum to glutathione S-transferase (EC 2.5.1.18) from maize (Zea mays L.) responsible for herbicide detoxification has been raised in rabbit. The antiserum was specific to the Mr 26 000 subunit of the enzyme from maize seedlings and suspension-cultured cells, and recognized the isoenzymes active toward both atrazine and metolachlor. When plants were treated for 24 h with the herbicide antidote N,N-diallyl-2-2-dichloroacetamide (DDCA), enzyme activities toward metolachlor were doubled in the roots and this was associated with a 70% increase in immunodetectable protein. Translation of polysomal R N A in vitro showed that the increase in the transferase in root tissue was brought about by a ninefold increase in m R N A activity encoding the enzyme. Treatment of suspension-cultured cells with cinnamic acid, metolachlor and D D C A raised enzyme activities but did not increase synthesis of glutathione S-transferase. In cultured maize cells, enzyme synthesis was maximal in mid-logarithmic phase, coinciding with the highest levels of enzyme activity. When callus cultures were established from the shoots of a maize line known to conjugate chloros-triazines, enzyme activity towards atrazine was lost during primary dedifferentiation. However, levels of total immunodetectable enzyme and activity toward metolachlor were increased in cultured cells compared with the parent shoot tissue.
Key words: Detoxification (herbicide) - Enzyme induction - Glutathione S-transferase - Herbicide antidote - Zea (herbicide detoxification). Department of Environmental Science, Schering Agrochemicals Ltd., Chesterford Park Research Station, Saffron Walden, Essex CBI0 IXL, UK * Present address and address' for correspondence:
Abbreviations: CDNB = 1-chloro-2,4-dinitrobenzene;DDCA =
N,N-diallyl-2-2-dichloroacetamide; GST = glutathione S-transferase; SDS-PAGE=sodium dodecyl sulphate-polyacrylamide gel electrophoresis
Introduction
Glutathione S-transferases (GST; EC 2.5.1.18) are a family of enzymes catalysing the conjugation of a range of both pesticides and naturally occurring compounds with glutathione to form polar, lesstoxic metabolites. Crops, such as maize contain GST activities toward chloro-s-triazine and chloracetanilide herbicides and permit such compounds to be used selectively in this species (Cole et al. 1987). Previous studies with plants and cell cultures of maize indicated that the transferases responsible for detoxifying atrazine and metolachlor (Fig. 1) were distinct isoenzymes (Edwards and Owen 1986 a). Subsequently three isoenzymes of GST all of subunit relative molecular mass (Mr) of 26000 but with isoelectric points of pH 4.9, 4.7 and 4.5 were resolved by chromatofocussing of maize leaf extracts which differed in their activities toward the two herbicides (Edwards and Owen 1987). Three isoenzymic forms of GST with component subunits of Mrs 29000, 27000 and 26000 which conjugated alachlor have also been described (Mozer et al. 1983; Moore et al. 1986). The genes encoding the M r 29000 and M r 26000 subunits have been cloned and sequenced and shown to have little homology with mammalian GST (Moore etal. 1986; Shah etal. 1986). Similarly, differences in the properties and specificity of GST isoenzymes involved in atrazine detoxification in maize have also been reported (Timmerman and Tu 1987) with the suggestion that these enzymes represent a supergene family similar to that found in rats (Lai and Tu 1986). The regulation of GST and other pesticidedetoxifying enzymes in crops by herbicide antidotes such as N,N-diallyl-2-2-dichloroacetamide (DDCA, Fig. 1) is of interest, as these compounds can alter the selective action of herbicides (Cole
100
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
CI
--CH 3
ATRAZINE \H
the first seedling node and sliced into 2-ram-thick disks. The disks were then placed on the culture medium described previously (Edwards and Owen 1986a) but containing 1.5 mg. 1 - 1 2,4-dichlorophenoxyacetate. Explants were maintained in the environmental growth chamber and callus tissue subcultured onto fresh medium containing 5 rag- 1 - 1 2,4-dichlorophenoxyacetate. The initiated callus was then maintained in the dark at 25 ~ C. An existing line of F71 cells derived from shoot explants was maintained as described previously on a medium supplemented with coconut water (Edwards and Owen 1986b). \
Me /
Me
Me
I
/CH--CHzOMe
II O
METOLACHLOR /CH2CH'- CHz CHCIzC--N. IoI \CH2CH-- CH 2
DDCA Fig. 1. Herbicide substrates atrazine and metolachlor, and herbicide antidote DDCA. Et, -- C2H~ ; Me, -- CH3
et al. 1987). By raising an antiserum to maize GST, it has been possible to monitor changes in the concentration and rates of synthesis of the enzyme in whole plants and cell cultures following treatment with the herbicide antidote DDCA. We also report the effect of initiating dedifferentiated growth on the levels and enzyme activities of GST in maize (cv. F71) roots and shoots. An established culture of this maize variety was known to be deficient in conjugating atrazine even though F71 shoots actively metabolise the herbicide (Edwards and Owen 1986b). Material and methods Plant material. Caryopses (cv. LGI 1) and suspension cultures (Black Mexican Sweeteorn) of maize (Zea mays L.) were prepared and maintained as described previously (Edwards and Owen 1986a). Seedlings were grown in vermiculite at 23~ under a 16-h photoperiod at an energy fluence rate of 14 W- m - z. Callus cultures derived from shoot tissue were initiated from maize seeds (cv. F71) obtained from Minist+re de l'Agriculture, INRA, Estrtes, Mons F. Ptronne, France. After washing in distilled water, seeds were sterilised by vigorous shaking in a solution of mercuric chloride (0.1%, w/v) for 15 min, followed by a wash in sterile distilled water for 16 h. Individual seeds were then germinated on sterile agar plates containing no nutrients in an environmental growth chamber for 6 d at 23 ~ C with 16 h light at 14 W . m -2. Shoots were excised 1 cm above
Chemicals and radioehemieals. Analytical-grade D D C A was generously donated by Stauffer Chemicals (UK) Ltd., Manningtree, Essex. [14C-Triazinyl]atrazine, [14C-phenyl]metolachlor and reference metabolites were obtained, prepared and analysed as described previously (Edwards and Owen 1986a). Rabbit reticulocyte lysate, [14C]methylated protein molecular-weight standards, L-[35S]methionine (54.4 TBq-mmol-1) and tzSI-labelled protein A (1.1 G B q . m g - 1 ) were purchased from Amersham International, Amersham, Bucks, UK. Protein-A covalently linked to Sepharose CL-4B was obtained from Sigma Chemical Co., Poole, Dorset, UK. Treatment with herbicide antidote. Whole seedlings of LG II maize grown in vermiculite for 7 d in an environmental growth chamber were treated by root drench with 15 ~tM D D C A in distilled water containing 0.01% (v/v) Teepol and harvested after 24 h. Control seedlings were treated with the detergent solution only. Whole plants were washed in distilled water, blotted dry and the excised leaf and root tissue frozen in liquid N2 prior to storage at --70 ~ C. Suspension cultures of Black Mexican Sweetcorn were used 7 d after subculture. Chemical treatments were added aseptically in methanol such that the final proportion of solvent in the medium did not exceed (0.001%, v/v). Control cells were treated with methanol only. After 24 h, cells were harvested by vacuum filtration and washed with distilled water prior to freezing in liquid Nz and storage at - 7 0 ~ C. Enzyme assay. Plant tissue was extracted in 2 (v/w) 0.2 M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1 buffer pH 7.8 with a pestle and mortar and a cell-free supernatant prepared by centrifugation (10000.g, 10 min, 4 ~ C). The spectrophotometric assay of GST with l-chloro-2,4-dinitrobenzene (CDNB) and radioassay with 14C-labelled herbicides has been described (Edwards and Owen 1986 a). To increase the sensitivity of the radioassay the final specific activities of the herbicide substrates were increased to 1.5 kBq per assay for [14C]atrazine and 0.5 kBq per assay for [~4C]metolachlor. Protein content was determined by the method of Bradford (1976) using Fraction-V bovine serum albumin as the reference protein for standard curves. Preparation o f antiserum. The isoenzyme of GST previously purified from maize leaves by bromosulfophthalein-glutathione affinity chromatography (Edwards and Owen 1986a) was used to raise antiserum. This isoenzyme was subsequently shown to have an isoelectric point of pH 4.9 and a greater activity towards metolachlor than atrazine (Edwards and Owen 1987). The pure protein (subunit Mr 26000) was lyophilised and 0.2 mg dissolved in I ml phosphate-buffered saline and dialysed against 2 t of this buffer for 16 h. The solution (1.6 ml) was then subdivided and 0.8 ml mixed with Freund's complete adjuvant (1:1, v/v) and injected subcutaneously into a male New Zealand white rabbit at eight sites. The remaining enzyme was subsequently injected in Freund's incomplete adjuvant at 28 d
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
101
and 42 d in equal lots. Blood was collected from the ear vein at 52 d and the prepared sermn stored at - 70 ~ C.
Table 1. Immunoprecipitation of enzyme activity by anti-GST serum compared with samples treated with control serum
Immotitration of radiolabelled enzyme. Suspension cultures of Black Mexican Sweetcorn cells (0.2-0.4 g) were incubated for 8 h with 0.9 MBq of L-[35S]methionine in fresh medium (4 ml). Cell extracts were prepared as described for enzyme assay and activity determined using 1-chloro-2,4-dinitrobenzene (CDNB) as substrate. Incorporation of radioactivity into protein was determined by liquid scintillation counting of glassfibre filters following filtration of precipitates resulting from treatment with 1 vol. 20% (w/v) trichloracetic acid containing 0.1% (w/v) I.methionine. Cell supernatants (20 gl) were mixed in microcentrifuge tubes with 0-50 lal of either control (pre-immuue) serum or anti-GST-serum made up to 0.25 ml with phosphate-buffer saline (pH 7.2) and incubated at 37 ~ C for 1 h. Samples were then treated for 16 h at 4 ~ C with protein-A-Sepharose (4 rag) which had been pre-swollen in a 1% solution of bovine serum albumin. Following mixing on an end-over-end turntable, antibody-antigen complex was recovered by centrifugation and the supernatant assayed for enzyme activity with CDNB. The Sepharose pellet was washed three times with phosphate-buffered saline (pH 7.2) containing Triton X-100 (1% v/v) and then resuspended in 60 gl of loading buffer composed of 63 mM Tris-HC1 buffer (pH 6.8) containing sodium dodecyl sulphate (4%, w/v), glycerol (5%, v/v), 2-mercaptoethanol (5%, v/v) and bromophenol blue (0.0013%, w/v). After incubation in a boiling-water bath (10 rain) samples were centrifuged and the supernatant (30 gl) analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) on 12% slab gels as described by Laemmli (1970). Following electrophoresis (constant 50 V), gels were washed in acetic acid (8%, v/v) and then fluorographed quantitatively as described by Robbins and Dixon (1984).
Substrate
Immune blotting. Plant extracts were prepared as described for enzyme assay and known amounts of protein (50-100 l~g) analysed by SDS-PAGE after treatment with 3 vol. of loading buffer as described above. Electroblotting and visualisation of antigen by autoradiography following treatment with 125I-labelled protein A was carried out by the method of Robbins and Dixon (1984). Quantification of antigen-antibody complex was achieved by excising radioactive zones from the nitrocellulose blots and radioassaying on a gamma counter. Translation of m R N A in vitro. Total polysomal R N A was isolated from maize plants and cell cultures by the method of Palmiter (1974) as modified by Schr6der et al. (1976). The R N A concentration was determined from the absorbance at 260 nm, and 15 gg of R N A added to an mRNA-dependent rabbit reticulocyte system containing L-[35S]methionine (1.1 MBq) in a total volume of 30 gl. Following incubation for 30 min at 37 ~ C as detailed by Amersham International, UK, the incorporation of radioactivity into protein was determined by acid precipitation. Samples (10 ~1) were then immunoprecipitated and analysed by SDS-PAGE and fluorography as described for GST immunotitration.
Results
Characterisation of the antiserum. Crude enzyme extracts from etiolated LG 11 maize leaves and Black Mexican Sweetcorn suspensions were prepared as described previously (Edwards and Owen 1986 a) and incubated with the antiserum and pro-
Immunoprecipitation of enzyme activity (Nat- gI- 1 serum), Enzyme source
CDNB Metolachlor Atrazine
Maize leaves (cv. L G l l )
Cell cultures (Black Mexican Sweetcorn)
40 600 5 5
36 400 3.3 0b
a Activity expressed as fmol product inhibited- s - x. gl - ~ antiserum u Cell cultures had negligible activity toward atrazine
tein-A-Sepharose as detailed for the immunotitration of the radiolabelled enzyme. The antiserum raised against the isoenzyme from maize leaves with an isoelectric point of pH 4.9 inhibited enzyme activities toward CDNB, metolachlor and atrazine (Table 1). However, no inhibition was observed prior to precipitation of antibody-antigen complex with protein-A-Sepharose indicating that the epitope(s) was not located at the active site of the enzyme. When crude extracts from leaves and cells were immuneblotted following SDSPAGE a single polypeptide of Mr 26000 was recognised (Fig. 2). Similarly, the GST purified by affinity chromatography (Edwards and Owen 1986a) was identified as the same polypeptide by the antiserum following immunoblotting. When immunoprecipitates of extracts from radiolabelled Black Mexican Sweetcorn cells were analysed by SDS-PAGE and fluorography and compared with extracts treated with control serum, several polypeptides were specifically precipitated by the anti-GST serum (Fig. 2). As determined from 14C-labelled protein markers, specifically immunoprecipitated peptides were observed of Mrs 87000, 80000 and 26000. The radioactive bands of Mrs 100000, 46000 and 40000 represent nonspecific binding to protein-A-Sepharose and were present in the control treatments. The peptide of Mr 26000 was of particular interest as this corresponds to the subunit molecular weight of GST detected by immune blotting. To confirm the identity of the peptide as GST, the radiolabelled cell extract was immunotitrated with the antiserum and a correlation established between declining enzyme activity toward CDNB and an increase in the concentration of immunoprecipitated peptide of Mr 26000 (Fig. 3). Control serum did not inhibit enzyme activity and failed to precipitate a peptide
102
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
Fig. 2. Characterisation of antigens recognised by anti-GST serum. Extracts were analysed by SDS-PAGE as follows. Track i, purified GST from maize leaves immuneblotted and treated with 125Ilabelled protein-A. Track 2, crude extract from maize leaves treated as per track 1. Track 3, crude extract from Black Mexican Sweetcorn treated as per track 1. Tracks 4-7 are fluorographs of extracts from suspension-cultured maize cells labelled for 8 h with [35S]methionine. Track 4, total extract from 7-d cells. Track 5, immunoprecipitate from 7-d cells. Track 6, immunoprecipitate from 14-d cells. Track 7, control immunoprecipitate obtained from treating 7-d cells with preimmune serum. Track 8, ~4C-labelled molecular-weight markers
100
~
~
~/__~
of similar molecular weight. Using cell extracts, complete inhibition of CDNB conjugation was observed with 140pl anti-GST serum per nkat enzyme activity.
100
v
50
50
o
~-
-~--m 0 0
30
antiserum
(~JI)
60
Fig. 3. Immunotitration of GST activity (CDNB as substrate) and radioactivity present as Mr-26000 peptide extracted from suspension cultures of Black Mexican Sweetcorn incubated for 8 h with [35S]methionine, as determined from a control extract of 0.35 nkat of activity containing no serum. The percentage decrease in enzyme activity in the supernatant was monitored using CDNB as the substrate after the addition of increasing amounts of pre-immune serum ( n - - D ) and anti-GST serum ( o - - o ) . The radioactivity present as Mr-26000 peptide in the irnmunoprecipitate was determined from quantitative fluorography of SDS-PAGE gels following addition of pre-immune serum (m--m) and anti-GST serum ( e - - e ) . 1 nkat = 1 nmol product formed- s- 1
Regulation of GST in cell-suspension culture. Previous studies with Black Mexican Sweetcorn suspension cultures had shown that GST activity towards CDNB and metolachlor was maximal in the mid-logarithmic phase of growth (Edwards and Owen 1986a). It was therefore of interest to determine the growth stage corresponding to the maximal rate of synthesis of GST in cells. Cultures were incubated with [BSS]methionine for 8 h at intervals up to 14 d after sub-culture, and radiolabelled peptides analysed by SDS-PAGE and fluorography. Synthesis of GST was maximal at 5 d and 7 d (midlogarithmic growth) and negligible by 14 d (Fig. 2). Densitometry of the fluorographs of 5-d immunoprecipitates indicated that 1% of the radioactivity incorporated into protein was present as GST after an 8-h incubation with [35S]methionine. The effect of pretreating cells for 24 h with known substrates of GST was investigated as the exposure of crops to pesticides has been reported subsequently to modify their xenobiotic metabolism (Cole et al. 1987). Both the herbicide metolachlor and cinnamic acid, an endogenous substrate of GST in plants (Diesperger and Sandermann 1979), raised the extractable GST activity toward metolachlor in cells (Table 2). In addition, the herbicide antidote D D C A also increased enzyme activities. In contrast no effect on GST activity was observed when maize cells were exposed to phenobarbitone (0.1 mM) or 3-methylcholanth-
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
103
Table 2. Effect of 24 h chemical treatments on GST activity in suspension cultures of Black Mexican Sweetcorn. Results expressed as mean values _+SD (n = 3) Chemical treatment
Enzyme activitya (pkat-mg- 1 protein)
Control Metolachlor (0.2 mM) Cinnamic acid (0.2 raM) D D C A (0.15 mM)
27 +- 2 36 _+4 34 _+3 39_+2
a Enzyme activity expressed as pmol product formed, s- x
rene (0.2 gM). Both compounds are potent inducers of GST isoenzymes in mammalian and insect systems. The increase in GST activity in maize cells following treatment with metolachlor, cinnamic acid and DDCA was not associated with detectable increases in the de-novo synthesis of the enzyme or in changes in the total amounts of immunodetectable protein. Thus, when polysomal mRNA from cultured cells treated for 24 h with 15 gM DDCA was translated in vitro no increase in the synthesis of the Mr 26000 peptide was observed as compared with controls (Fig. 4). Effect of DDCA on G S T in whole plants. Following a 24-h treatment with 15 gM DDCA, GST activity toward metolachlor was doubled in the roots of maize seedlings (Table 3). However, this increase was not observed when atrazine was assayed as an enzyme substrate. Crude extracts of maize roots exposed to DDCA were compared with controls by immuneblotting of identical amounts of protein following SDS-PAGE. In the DDCA-treated roots, the levels of enzyme were raised by 71% (Table 3). However, DDCA had no appreciable effect on enzyme activities or immunodetectable protein in the leaves. Changes in the 'activity' of mRNA encoding GST were monitored by immunoprecipitation of protein labelled with [3SS]methionine following in vitro translation. The immunoprecipitates were then analysed by SDS-PAGE followed by quantitative densitometry of the resulting fluorographs. In roots the major immunoprecipitated peptide (Mr 26000) appeared to be identical to GST (Fig. 4). In leaves an additional peptide (Mr 28 000) was also observed as a major immunoprecipitated product of translation but was not detected by immuneblotting of whole-leaf extracts (Fig. 2). The additional bands present as immunoprecipitated products of in-vitro translation of leaf and root mRNA represented non-specific binding of radiolabelled peptides to the protein-A-Sepharose and
Fig. 4. Effect of 24-h treatment with 15 gM D D C A on m R N A 'activity' encoding GST in plants and cell cultures of maize. Immunoprecipitates of [35S]methionine-labelled peptides obtained from in-vitro translation of polysomal m R N A were analysed by SDS-PAGE and fluorography. Sources of m R N A were as follows: track 5, control leaves; track 6, leaves from seedlings treated with antidote applied as a root drench; track 3, control roots; track 4, roots treated with antidote; track 1, control suspension-cultured Black Mexican Sweetcorn ceils; track 2, cells treated with antidote; track 7, immunoprecipitate obtained using pre-immune serum Table 3. Effect of 24 h treatment with 15 gM D D C A on GST activities and concentrations of immunodetectable protein in L G l l maize seedlings Enzyme source
Enzyme activity" (pkat.mg- 1 protein)
Immunodetectable b cpm
Atrazine
Metolachlor
0.24+0.06 0.30_+0.08
22___5 45+_3
1118_+55 1563+_43
3.72 +- 1.37 2.96_+0.60
25 • 8 19•
1255_+ 108 1249_+67
Root Control D D C A treated Leaf Control D D C A treated
a Enzyme activity expressed as pmol product formed.s-1 and presented as mean values + SD (n = 3) b Derived from treatment of immune-blotted enzyme with lzsIlabelled protein A
were also observed in samples treated with control serum. In maize leaves, DDCA had no effect on immunoprecipitated peptides derived from translation of total polysomal mRNA. However, in roots
104
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
Table 4. Changes in the levels of extractable GST enzyme activity and immunodetectable protein in germinating shoots and initiated cell cultures of F71 maize
Developmental stage
Enzyme activity (pkat.mg- 1 protein) a Atrazine
Immunodetectableb cpm (%)
Metolachlor
Germinating shoot (day 6)
0.31
7.04
810 (100%)
Initial callus
0.12
14.14
855 (106%)
New cell culture
0.04
16.48
1022 (126%)
Established cell culture
0.02
37.33
1183 (146%)
Enzyme activity expressed as pmol product formed, s -1. rag- 1 protein and presented as mean values of duplicate determinations b Derived from treatment of immune-blotted enzyme with 125Ilabelled protein A
mRNA 'activity' encoding GST was enriched ninefold in antidote-treated tissue as compared with controls. Effect on initiation of undifferentiated growth on G S T in maize shoots. Callus cultures were initiated from germinating shoots of F71 maize seeds and analysed for extractable enzyme activities and immunodetectable protein (Table 4). Following the initiation of callus from 6-d shoots, enzyme activity with atrazine as substrate declined and, after subsequent subculture to give dedifferentiated growth, was reduced to 13% of the level found in the shoots. Similarly, an existing cell line of F71 shoot callus had very low activity toward atrazine (0.02 pkat.mg- ~ protein). In contrast, GST activity toward metolachlor was raised by the initiation and maintenance of cell culture such that the activity in the new cell line was 134% higher than that of 6-d shoots. Total immunodetectable GST was increased during sub-culture and was maximal in the established callus. However, the increase in total enzyme could not account for the much larger proportional increases in GST activity. Discussion
Specificity of the anti-GST serum. The antiserum raised against the isoenzyme of GST from maize leaves responsible for metolachlor conjugation recognized the isoenzymes responsible for atrazine detoxification. Immuneblotting experiments have shown that the three isoenzymes responsible for atrazine and metolachlor conjugation in LG 11 maize all have a subunit Mr of 26000 (Edwards
and Owen 1987). Similarly, when crude extracts from leaves and roots were immuneblotted only the Mr 26000 peptide was recognised, indicating that the antiserum was highly specific. The identity of the antigen as maize GST was subsequently confirmed by immunotitration of the Mr 26000 peptide such that increased immunoprecipitation of the protein corresponded to a directly proportional decline in soluble enzyme activity. When polysomal mRNA from maize leaves was translated in vitro an additional peptide (Mr 28000) was observed which may represent a precursor or related form of the Mr 26 000 subunit. In previous studies, purification of maize leaf GST by bromosulfophthalein-glutathione affinity chromatography showed that an Mr 28 800 peptide was present as a minor component of the purified enzyme (Edwards and Owen 1986a). Recently, it has been reported that an antibody raised against a homodimer of maize GST of subunit Mr 28 000 showed little cross reactivity with a lower-molecular-weight subunit present in a heterodimer form of the enzyme (Timmerman and Tu 1987). At present it remains unclear whether the complexity of GSTsubunit forms in maize arises wholly from gene multiplicity (Moore et al. 1986; Shah et al. 1986; Wiegand et al. 1986) or that post-translational effects give rise to related enzymes of lower molecular weight. Thus an Mr 25 000 peptide of GST has been reported as being derived from an Mr 29000 subunit of the transferase (Wiegand et al. 1986). When immunoprecipitates were prepared from suspension cultures of Black Mexican Sweetcorn labelled with [35S]methionine, two high-molecularweight peptides (Mr 80000, Mr 87000) were specifically recognised. However, these peptides were not observed following immunoprecipitation of products derived from the in-vitro translation of cell polysomes. Similarly no high-molecular-weight forms of GST were reported in extracts of maize leaf mRNA probed with a l.l-kilobase copy cDNA encoding an Mr 29000 subunit of the enzyme (Wiegand et al. 1986). These results indicate that the Mr 80000 and Mr 87000 peptides are not related to GST but share a common epitope with the transferase. Enhancement of G S T in maize seedlings treated with the herbicide antidote DDCA. By using the antiserum to GST in maize it has been possible to study antidote-induced changes in enzyme synthesis in the roots of seedlings 24 h after treatment with 15 gM DDCA. Thus DDCA increased the 'activity' of mRNA encoding GST ninefold in polysomes isolated from roots and this was associated with
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
a 71% increase in immunodetectable Mr 26000 peptide. The elevation in enzyme concentration resulted in a doubling in extractable activity toward the chloracetanilide herbicide metolachlor but had no effect on atrazine conjugation. Previous studies have shown that the isoenzymes in maize leaves responsible for the detoxification of atrazine and metolachlor can be differentiated on the basis of their isoelectric points (Edwards and Owen 1987). The results presented here indicate that DDCA specifically regulates the expression of the gene encoding the isoenzyme of GST active toward metolachlor but has no effect on the isoenzyme responsible for atrazine conjugation. In a recent study a cDNA probe for the mRNA encoding the GST subunit (Mr 29 000) responsible for alachlor conjugation in maize has been prepared and used to study the effect of a seed dressing of DDCA on GST transcription (Wiegand et al. 1986). The mRNA encoding the isoenzyme was increased between three- and fourfold in the leaves of 8-d seedlings grown from antidotedressed seeds as compared with control plants. This increase in transcription corresponded to known increases in the concentration and activity of an isoenzyme of GST in maize leaves which actively conjugated alachlor (Mozer et al. 1983). It is now apparent that a root drench with DDCA can induce an Mr 26 000 subunit of the transferase and that this effect occurs within 24 h in the roots. The failure of the antidote to increase the synthesis of GST in leaves probably results from slow translocation of DDCA to the leaves or detoxification of the compound to inactive metabolites in the roots. This was not confirmed however.
Regulation of GST in cell cultures of Z. mays. When suspension cultures of Black Mexican Sweetcorn were grown in the presence of [35S]methionine, analysis of immunoprecipitates by SDS-PAGE showed that the synthesis of GST was maximal in cells in mid-logarithmic growth phase (5-7 d). Similarly, such actively dividing cultures have been shown to contain higher extractable activities of GST responsible for metolachlor conjugation than was the case in cells in stationary phase (Edwards and Owen 1986a). Treatment of cells with cinnamic acid, metolachlor or DDCA, which are all substrates of GST, enhanced the activity of the enzyme toward metolachlor but this was not accompanied by increased incorporation of [aSS]methionine into the Mr 26000 subunit. In addition, DDCA did not increase the activity of polysomal mRNA in the cells encoding GST and this indicated that the antidote may exert additional
105
post-translational effects in maize. Interestingly DDCA has also been reported to have rapid effects on lipid metabolism in maize cell cultures and to increase cellular glutathione (Ezra and Gressel 1982). Previous studies have shown that cell cultures derived from maize shoots were unable to detoxify atrazine as a consequence of an apparent deficiency in the isoenzyme of GST responsible for chloro-s-triazine conjugation. By initiating a callus culture from the shoot tissue of a maize line (cv. F71) known actively to conjugate atrazine (Edwards and Owen 1986b) it has been possible to show that the loss in isoenzyme activity occurs after primary dedifferentiation. By contrast, metolachlor conjugation was enhanced during the establishment of the callus and this was accompanied by a 46% increase in the total immunodetectable GST. The high concentration of the enzyme active toward metolachlor in suspension cultures of Black Mexican Sweetcorn may explain the very rapid metabolism of the herbicide in this cell line (Cole et al. 1987).
Regulation of pesticide-degrading enzymes in plants. Increasing evidence indicates that enzymes which detoxify herbicides in plants can be influenced by chemicals, thus allowing the selectivity of pesticides to be manipulated (Cole et al. 1987). Studies with the glutathione S-transferases of maize have shown that such regulation may be restricted to specific isoenzymic forms at the level of gene transcription. Such selective regulation of GST isoenzymes may represent a sensitive means of regulating intracellular concentrations of endogenous electrophilic substrates by forming products of greatly increased polarity. In this respect, some herbicide antidotes may mimic the regulatory activity of naturally occurring plant products. In order to modify the activity of additional pesticide-degrading enzymes in plants it will be necessary to study the enzymology and molecular biology of these proteins, which are currently largely uncharacterised. This study was supported by the Agricultural and Food Research Council. The authors would like to thank Sue Robins for the preparation of this manuscript.
References Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248 254 Cole, D.J., Edwards, R., Owen, W.J. (1987) The role of metabo-
106
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
lism in herbicide selectivity. In: Progress in pesticide biochemistry and toxicology, vol. 6, pp. 57-104, Hutson, D.H.. Roberts, T.R. eds. J. Wiley and Sons, Chichester Diesperger, H., Sandermann (Jr), H. (1979) Soluble and rnicrosornal glutathione s-transferase activities in pea seedlings (Fisum sativum L.) Planta 146, 643 648 Edwards, R., Owen, W.J. (1986a) Comparison of glutathione S-transferases of Zea mays responsible for herbicide detoxification in plants and suspension-cultured cells. Planta 169, 208-215 Edwards, R., Owen, W.J. (1986b) Glutathione S-transferases involved in herbicide detoxification in plant cell culture. Aspects Appl. Biol. 77, 121-130 Edwards, R., Owen, W.J. (1987) lsoenzymes of glutathione Stransferase in Zea mays. Biochem. Soc. Trans. 15, 1184 Ezra, G., Gressel, J. (1982) Rapid effects of a thiocarbamate protectant on rnacrornolecular synthesis and glutathione levels in maize cell cultures. Pestic. Biochem. Physiol. 17, 48-58 Laemrnli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Lai, H-C.J., Tu, C-P.D. (1986) Rat glutathione s-transferase supergene family. J. Biol. Chem. 261, 13793-13799 Moore, R.E., Davies, M.S., O'Connell, K.M., Harding, E.I., Wiegand, R.C., Tierneier, D.C. (1986) Cloning and expression of a cDNA encoding a maize glutathione S-transferase in E. coff. Nucl. Acids Res. 14, 7227-7235 Mozer, T.J., Tierneier, D.C., Jaworski, E.G. (1983) Purification
and characterisation of corn glutathione S-transferase. Biochemistry 22, 1068-1072 Palmiter, R.D. (1974) Magnesium precipitation of ribonucleoprotein complexes. Expedient techniques for the isolation of undegraded polysomes and messenger ribonucleic acid. Biochemistry 13, 3606-3615 Robbins, M.P., Dixon, R.A. (1984) Induction of chalcone isomerase in elicitor-treated bean cells : comparison of rates of synthesis and appearance of irnmunodetectable enzyme. Eur. J. Biochem. 145, 195-202 Shah, D.M., Hironaka, C.M., Wiegand, R.C., Harding, E.I., Krivi, G.G., Tiemeier, D.C. (1986) Structural analysis of a maize gene coding for glutathione S-transferase involved in herbicide detoxification. Plant Mol. Biol. 6, 203-211 Schr6der, J., Betz, B., Hahlbrock, K. (1976) Light-induced enzyrne synthesis in cell suspension cultures of Petroselinum hortense. Eur. J. Bioehern. 67, 527 541 Timmerrnan, K.P., Tu, C-P.D. (1987) Genetic evidence ofxenobiotics metabolism by glutathione S-transferases from corn. In: Glutathione S-transferase and carcinogenesis, pp. 47-49, Mantle, T.J., Pickett, C.B., Hayes, J.D. eds. Taylor and Francis, London New York Philadelphia Wiegand, R.C., Shah, D.M., Mozer, T.J., Harding, E.I., DiazCollier, J., Saunders, C., Jaworski, E.G., Tiermeier, D.C. (1986) Messenger RNA encoding a glutathione S-transferase responsible for herbicide tolerance is induced in response to safener treatment. Plant. Mol. Biol. 7, 235-243 Received 20 Novemer 1987; accepted 17 March 1988
Planta (1988) 175:9%106
9 Springer-Verlag 1988
Regulation of glutathione S-transferases of Zea mays in plants and cell cultures R. Edwards* and W.J. Owen Department of Biochemistry, Royal Holloway and Bedford New College, Egham Hill, Egham, Surrey TW20 OEX, U.K.
Abstract. An antiserum to glutathione S-transferase (EC 2.5.1.18) from maize (Zea mays L.) responsible for herbicide detoxification has been raised in rabbit. The antiserum was specific to the Mr 26 000 subunit of the enzyme from maize seedlings and suspension-cultured cells, and recognized the isoenzymes active toward both atrazine and metolachlor. When plants were treated for 24 h with the herbicide antidote N,N-diallyl-2-2-dichloroacetamide (DDCA), enzyme activities toward metolachlor were doubled in the roots and this was associated with a 70% increase in immunodetectable protein. Translation of polysomal R N A in vitro showed that the increase in the transferase in root tissue was brought about by a ninefold increase in m R N A activity encoding the enzyme. Treatment of suspension-cultured cells with cinnamic acid, metolachlor and D D C A raised enzyme activities but did not increase synthesis of glutathione S-transferase. In cultured maize cells, enzyme synthesis was maximal in mid-logarithmic phase, coinciding with the highest levels of enzyme activity. When callus cultures were established from the shoots of a maize line known to conjugate chloros-triazines, enzyme activity towards atrazine was lost during primary dedifferentiation. However, levels of total immunodetectable enzyme and activity toward metolachlor were increased in cultured cells compared with the parent shoot tissue.
Key words: Detoxification (herbicide) - Enzyme induction - Glutathione S-transferase - Herbicide antidote - Zea (herbicide detoxification). Department of Environmental Science, Schering Agrochemicals Ltd., Chesterford Park Research Station, Saffron Walden, Essex CBI0 IXL, UK * Present address and address' for correspondence:
Abbreviations: CDNB = 1-chloro-2,4-dinitrobenzene;DDCA =
N,N-diallyl-2-2-dichloroacetamide; GST = glutathione S-transferase; SDS-PAGE=sodium dodecyl sulphate-polyacrylamide gel electrophoresis
Introduction
Glutathione S-transferases (GST; EC 2.5.1.18) are a family of enzymes catalysing the conjugation of a range of both pesticides and naturally occurring compounds with glutathione to form polar, lesstoxic metabolites. Crops, such as maize contain GST activities toward chloro-s-triazine and chloracetanilide herbicides and permit such compounds to be used selectively in this species (Cole et al. 1987). Previous studies with plants and cell cultures of maize indicated that the transferases responsible for detoxifying atrazine and metolachlor (Fig. 1) were distinct isoenzymes (Edwards and Owen 1986 a). Subsequently three isoenzymes of GST all of subunit relative molecular mass (Mr) of 26000 but with isoelectric points of pH 4.9, 4.7 and 4.5 were resolved by chromatofocussing of maize leaf extracts which differed in their activities toward the two herbicides (Edwards and Owen 1987). Three isoenzymic forms of GST with component subunits of Mrs 29000, 27000 and 26000 which conjugated alachlor have also been described (Mozer et al. 1983; Moore et al. 1986). The genes encoding the M r 29000 and M r 26000 subunits have been cloned and sequenced and shown to have little homology with mammalian GST (Moore etal. 1986; Shah etal. 1986). Similarly, differences in the properties and specificity of GST isoenzymes involved in atrazine detoxification in maize have also been reported (Timmerman and Tu 1987) with the suggestion that these enzymes represent a supergene family similar to that found in rats (Lai and Tu 1986). The regulation of GST and other pesticidedetoxifying enzymes in crops by herbicide antidotes such as N,N-diallyl-2-2-dichloroacetamide (DDCA, Fig. 1) is of interest, as these compounds can alter the selective action of herbicides (Cole
100
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
CI
--CH 3
ATRAZINE \H
the first seedling node and sliced into 2-ram-thick disks. The disks were then placed on the culture medium described previously (Edwards and Owen 1986a) but containing 1.5 mg. 1 - 1 2,4-dichlorophenoxyacetate. Explants were maintained in the environmental growth chamber and callus tissue subcultured onto fresh medium containing 5 rag- 1 - 1 2,4-dichlorophenoxyacetate. The initiated callus was then maintained in the dark at 25 ~ C. An existing line of F71 cells derived from shoot explants was maintained as described previously on a medium supplemented with coconut water (Edwards and Owen 1986b). \
Me /
Me
Me
I
/CH--CHzOMe
II O
METOLACHLOR /CH2CH'- CHz CHCIzC--N. IoI \CH2CH-- CH 2
DDCA Fig. 1. Herbicide substrates atrazine and metolachlor, and herbicide antidote DDCA. Et, -- C2H~ ; Me, -- CH3
et al. 1987). By raising an antiserum to maize GST, it has been possible to monitor changes in the concentration and rates of synthesis of the enzyme in whole plants and cell cultures following treatment with the herbicide antidote DDCA. We also report the effect of initiating dedifferentiated growth on the levels and enzyme activities of GST in maize (cv. F71) roots and shoots. An established culture of this maize variety was known to be deficient in conjugating atrazine even though F71 shoots actively metabolise the herbicide (Edwards and Owen 1986b). Material and methods Plant material. Caryopses (cv. LGI 1) and suspension cultures (Black Mexican Sweeteorn) of maize (Zea mays L.) were prepared and maintained as described previously (Edwards and Owen 1986a). Seedlings were grown in vermiculite at 23~ under a 16-h photoperiod at an energy fluence rate of 14 W- m - z. Callus cultures derived from shoot tissue were initiated from maize seeds (cv. F71) obtained from Minist+re de l'Agriculture, INRA, Estrtes, Mons F. Ptronne, France. After washing in distilled water, seeds were sterilised by vigorous shaking in a solution of mercuric chloride (0.1%, w/v) for 15 min, followed by a wash in sterile distilled water for 16 h. Individual seeds were then germinated on sterile agar plates containing no nutrients in an environmental growth chamber for 6 d at 23 ~ C with 16 h light at 14 W . m -2. Shoots were excised 1 cm above
Chemicals and radioehemieals. Analytical-grade D D C A was generously donated by Stauffer Chemicals (UK) Ltd., Manningtree, Essex. [14C-Triazinyl]atrazine, [14C-phenyl]metolachlor and reference metabolites were obtained, prepared and analysed as described previously (Edwards and Owen 1986a). Rabbit reticulocyte lysate, [14C]methylated protein molecular-weight standards, L-[35S]methionine (54.4 TBq-mmol-1) and tzSI-labelled protein A (1.1 G B q . m g - 1 ) were purchased from Amersham International, Amersham, Bucks, UK. Protein-A covalently linked to Sepharose CL-4B was obtained from Sigma Chemical Co., Poole, Dorset, UK. Treatment with herbicide antidote. Whole seedlings of LG II maize grown in vermiculite for 7 d in an environmental growth chamber were treated by root drench with 15 ~tM D D C A in distilled water containing 0.01% (v/v) Teepol and harvested after 24 h. Control seedlings were treated with the detergent solution only. Whole plants were washed in distilled water, blotted dry and the excised leaf and root tissue frozen in liquid N2 prior to storage at --70 ~ C. Suspension cultures of Black Mexican Sweetcorn were used 7 d after subculture. Chemical treatments were added aseptically in methanol such that the final proportion of solvent in the medium did not exceed (0.001%, v/v). Control cells were treated with methanol only. After 24 h, cells were harvested by vacuum filtration and washed with distilled water prior to freezing in liquid Nz and storage at - 7 0 ~ C. Enzyme assay. Plant tissue was extracted in 2 (v/w) 0.2 M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1 buffer pH 7.8 with a pestle and mortar and a cell-free supernatant prepared by centrifugation (10000.g, 10 min, 4 ~ C). The spectrophotometric assay of GST with l-chloro-2,4-dinitrobenzene (CDNB) and radioassay with 14C-labelled herbicides has been described (Edwards and Owen 1986 a). To increase the sensitivity of the radioassay the final specific activities of the herbicide substrates were increased to 1.5 kBq per assay for [14C]atrazine and 0.5 kBq per assay for [~4C]metolachlor. Protein content was determined by the method of Bradford (1976) using Fraction-V bovine serum albumin as the reference protein for standard curves. Preparation o f antiserum. The isoenzyme of GST previously purified from maize leaves by bromosulfophthalein-glutathione affinity chromatography (Edwards and Owen 1986a) was used to raise antiserum. This isoenzyme was subsequently shown to have an isoelectric point of pH 4.9 and a greater activity towards metolachlor than atrazine (Edwards and Owen 1987). The pure protein (subunit Mr 26000) was lyophilised and 0.2 mg dissolved in I ml phosphate-buffered saline and dialysed against 2 t of this buffer for 16 h. The solution (1.6 ml) was then subdivided and 0.8 ml mixed with Freund's complete adjuvant (1:1, v/v) and injected subcutaneously into a male New Zealand white rabbit at eight sites. The remaining enzyme was subsequently injected in Freund's incomplete adjuvant at 28 d
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
101
and 42 d in equal lots. Blood was collected from the ear vein at 52 d and the prepared sermn stored at - 70 ~ C.
Table 1. Immunoprecipitation of enzyme activity by anti-GST serum compared with samples treated with control serum
Immotitration of radiolabelled enzyme. Suspension cultures of Black Mexican Sweetcorn cells (0.2-0.4 g) were incubated for 8 h with 0.9 MBq of L-[35S]methionine in fresh medium (4 ml). Cell extracts were prepared as described for enzyme assay and activity determined using 1-chloro-2,4-dinitrobenzene (CDNB) as substrate. Incorporation of radioactivity into protein was determined by liquid scintillation counting of glassfibre filters following filtration of precipitates resulting from treatment with 1 vol. 20% (w/v) trichloracetic acid containing 0.1% (w/v) I.methionine. Cell supernatants (20 gl) were mixed in microcentrifuge tubes with 0-50 lal of either control (pre-immuue) serum or anti-GST-serum made up to 0.25 ml with phosphate-buffer saline (pH 7.2) and incubated at 37 ~ C for 1 h. Samples were then treated for 16 h at 4 ~ C with protein-A-Sepharose (4 rag) which had been pre-swollen in a 1% solution of bovine serum albumin. Following mixing on an end-over-end turntable, antibody-antigen complex was recovered by centrifugation and the supernatant assayed for enzyme activity with CDNB. The Sepharose pellet was washed three times with phosphate-buffered saline (pH 7.2) containing Triton X-100 (1% v/v) and then resuspended in 60 gl of loading buffer composed of 63 mM Tris-HC1 buffer (pH 6.8) containing sodium dodecyl sulphate (4%, w/v), glycerol (5%, v/v), 2-mercaptoethanol (5%, v/v) and bromophenol blue (0.0013%, w/v). After incubation in a boiling-water bath (10 rain) samples were centrifuged and the supernatant (30 gl) analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) on 12% slab gels as described by Laemmli (1970). Following electrophoresis (constant 50 V), gels were washed in acetic acid (8%, v/v) and then fluorographed quantitatively as described by Robbins and Dixon (1984).
Substrate
Immune blotting. Plant extracts were prepared as described for enzyme assay and known amounts of protein (50-100 l~g) analysed by SDS-PAGE after treatment with 3 vol. of loading buffer as described above. Electroblotting and visualisation of antigen by autoradiography following treatment with 125I-labelled protein A was carried out by the method of Robbins and Dixon (1984). Quantification of antigen-antibody complex was achieved by excising radioactive zones from the nitrocellulose blots and radioassaying on a gamma counter. Translation of m R N A in vitro. Total polysomal R N A was isolated from maize plants and cell cultures by the method of Palmiter (1974) as modified by Schr6der et al. (1976). The R N A concentration was determined from the absorbance at 260 nm, and 15 gg of R N A added to an mRNA-dependent rabbit reticulocyte system containing L-[35S]methionine (1.1 MBq) in a total volume of 30 gl. Following incubation for 30 min at 37 ~ C as detailed by Amersham International, UK, the incorporation of radioactivity into protein was determined by acid precipitation. Samples (10 ~1) were then immunoprecipitated and analysed by SDS-PAGE and fluorography as described for GST immunotitration.
Results
Characterisation of the antiserum. Crude enzyme extracts from etiolated LG 11 maize leaves and Black Mexican Sweetcorn suspensions were prepared as described previously (Edwards and Owen 1986 a) and incubated with the antiserum and pro-
Immunoprecipitation of enzyme activity (Nat- gI- 1 serum), Enzyme source
CDNB Metolachlor Atrazine
Maize leaves (cv. L G l l )
Cell cultures (Black Mexican Sweetcorn)
40 600 5 5
36 400 3.3 0b
a Activity expressed as fmol product inhibited- s - x. gl - ~ antiserum u Cell cultures had negligible activity toward atrazine
tein-A-Sepharose as detailed for the immunotitration of the radiolabelled enzyme. The antiserum raised against the isoenzyme from maize leaves with an isoelectric point of pH 4.9 inhibited enzyme activities toward CDNB, metolachlor and atrazine (Table 1). However, no inhibition was observed prior to precipitation of antibody-antigen complex with protein-A-Sepharose indicating that the epitope(s) was not located at the active site of the enzyme. When crude extracts from leaves and cells were immuneblotted following SDSPAGE a single polypeptide of Mr 26000 was recognised (Fig. 2). Similarly, the GST purified by affinity chromatography (Edwards and Owen 1986a) was identified as the same polypeptide by the antiserum following immunoblotting. When immunoprecipitates of extracts from radiolabelled Black Mexican Sweetcorn cells were analysed by SDS-PAGE and fluorography and compared with extracts treated with control serum, several polypeptides were specifically precipitated by the anti-GST serum (Fig. 2). As determined from 14C-labelled protein markers, specifically immunoprecipitated peptides were observed of Mrs 87000, 80000 and 26000. The radioactive bands of Mrs 100000, 46000 and 40000 represent nonspecific binding to protein-A-Sepharose and were present in the control treatments. The peptide of Mr 26000 was of particular interest as this corresponds to the subunit molecular weight of GST detected by immune blotting. To confirm the identity of the peptide as GST, the radiolabelled cell extract was immunotitrated with the antiserum and a correlation established between declining enzyme activity toward CDNB and an increase in the concentration of immunoprecipitated peptide of Mr 26000 (Fig. 3). Control serum did not inhibit enzyme activity and failed to precipitate a peptide
102
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
Fig. 2. Characterisation of antigens recognised by anti-GST serum. Extracts were analysed by SDS-PAGE as follows. Track i, purified GST from maize leaves immuneblotted and treated with 125Ilabelled protein-A. Track 2, crude extract from maize leaves treated as per track 1. Track 3, crude extract from Black Mexican Sweetcorn treated as per track 1. Tracks 4-7 are fluorographs of extracts from suspension-cultured maize cells labelled for 8 h with [35S]methionine. Track 4, total extract from 7-d cells. Track 5, immunoprecipitate from 7-d cells. Track 6, immunoprecipitate from 14-d cells. Track 7, control immunoprecipitate obtained from treating 7-d cells with preimmune serum. Track 8, ~4C-labelled molecular-weight markers
100
~
~
~/__~
of similar molecular weight. Using cell extracts, complete inhibition of CDNB conjugation was observed with 140pl anti-GST serum per nkat enzyme activity.
100
v
50
50
o
~-
-~--m 0 0
30
antiserum
(~JI)
60
Fig. 3. Immunotitration of GST activity (CDNB as substrate) and radioactivity present as Mr-26000 peptide extracted from suspension cultures of Black Mexican Sweetcorn incubated for 8 h with [35S]methionine, as determined from a control extract of 0.35 nkat of activity containing no serum. The percentage decrease in enzyme activity in the supernatant was monitored using CDNB as the substrate after the addition of increasing amounts of pre-immune serum ( n - - D ) and anti-GST serum ( o - - o ) . The radioactivity present as Mr-26000 peptide in the irnmunoprecipitate was determined from quantitative fluorography of SDS-PAGE gels following addition of pre-immune serum (m--m) and anti-GST serum ( e - - e ) . 1 nkat = 1 nmol product formed- s- 1
Regulation of GST in cell-suspension culture. Previous studies with Black Mexican Sweetcorn suspension cultures had shown that GST activity towards CDNB and metolachlor was maximal in the mid-logarithmic phase of growth (Edwards and Owen 1986a). It was therefore of interest to determine the growth stage corresponding to the maximal rate of synthesis of GST in cells. Cultures were incubated with [BSS]methionine for 8 h at intervals up to 14 d after sub-culture, and radiolabelled peptides analysed by SDS-PAGE and fluorography. Synthesis of GST was maximal at 5 d and 7 d (midlogarithmic growth) and negligible by 14 d (Fig. 2). Densitometry of the fluorographs of 5-d immunoprecipitates indicated that 1% of the radioactivity incorporated into protein was present as GST after an 8-h incubation with [35S]methionine. The effect of pretreating cells for 24 h with known substrates of GST was investigated as the exposure of crops to pesticides has been reported subsequently to modify their xenobiotic metabolism (Cole et al. 1987). Both the herbicide metolachlor and cinnamic acid, an endogenous substrate of GST in plants (Diesperger and Sandermann 1979), raised the extractable GST activity toward metolachlor in cells (Table 2). In addition, the herbicide antidote D D C A also increased enzyme activities. In contrast no effect on GST activity was observed when maize cells were exposed to phenobarbitone (0.1 mM) or 3-methylcholanth-
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
103
Table 2. Effect of 24 h chemical treatments on GST activity in suspension cultures of Black Mexican Sweetcorn. Results expressed as mean values _+SD (n = 3) Chemical treatment
Enzyme activitya (pkat-mg- 1 protein)
Control Metolachlor (0.2 mM) Cinnamic acid (0.2 raM) D D C A (0.15 mM)
27 +- 2 36 _+4 34 _+3 39_+2
a Enzyme activity expressed as pmol product formed, s- x
rene (0.2 gM). Both compounds are potent inducers of GST isoenzymes in mammalian and insect systems. The increase in GST activity in maize cells following treatment with metolachlor, cinnamic acid and DDCA was not associated with detectable increases in the de-novo synthesis of the enzyme or in changes in the total amounts of immunodetectable protein. Thus, when polysomal mRNA from cultured cells treated for 24 h with 15 gM DDCA was translated in vitro no increase in the synthesis of the Mr 26000 peptide was observed as compared with controls (Fig. 4). Effect of DDCA on G S T in whole plants. Following a 24-h treatment with 15 gM DDCA, GST activity toward metolachlor was doubled in the roots of maize seedlings (Table 3). However, this increase was not observed when atrazine was assayed as an enzyme substrate. Crude extracts of maize roots exposed to DDCA were compared with controls by immuneblotting of identical amounts of protein following SDS-PAGE. In the DDCA-treated roots, the levels of enzyme were raised by 71% (Table 3). However, DDCA had no appreciable effect on enzyme activities or immunodetectable protein in the leaves. Changes in the 'activity' of mRNA encoding GST were monitored by immunoprecipitation of protein labelled with [3SS]methionine following in vitro translation. The immunoprecipitates were then analysed by SDS-PAGE followed by quantitative densitometry of the resulting fluorographs. In roots the major immunoprecipitated peptide (Mr 26000) appeared to be identical to GST (Fig. 4). In leaves an additional peptide (Mr 28 000) was also observed as a major immunoprecipitated product of translation but was not detected by immuneblotting of whole-leaf extracts (Fig. 2). The additional bands present as immunoprecipitated products of in-vitro translation of leaf and root mRNA represented non-specific binding of radiolabelled peptides to the protein-A-Sepharose and
Fig. 4. Effect of 24-h treatment with 15 gM D D C A on m R N A 'activity' encoding GST in plants and cell cultures of maize. Immunoprecipitates of [35S]methionine-labelled peptides obtained from in-vitro translation of polysomal m R N A were analysed by SDS-PAGE and fluorography. Sources of m R N A were as follows: track 5, control leaves; track 6, leaves from seedlings treated with antidote applied as a root drench; track 3, control roots; track 4, roots treated with antidote; track 1, control suspension-cultured Black Mexican Sweetcorn ceils; track 2, cells treated with antidote; track 7, immunoprecipitate obtained using pre-immune serum Table 3. Effect of 24 h treatment with 15 gM D D C A on GST activities and concentrations of immunodetectable protein in L G l l maize seedlings Enzyme source
Enzyme activity" (pkat.mg- 1 protein)
Immunodetectable b cpm
Atrazine
Metolachlor
0.24+0.06 0.30_+0.08
22___5 45+_3
1118_+55 1563+_43
3.72 +- 1.37 2.96_+0.60
25 • 8 19•
1255_+ 108 1249_+67
Root Control D D C A treated Leaf Control D D C A treated
a Enzyme activity expressed as pmol product formed.s-1 and presented as mean values + SD (n = 3) b Derived from treatment of immune-blotted enzyme with lzsIlabelled protein A
were also observed in samples treated with control serum. In maize leaves, DDCA had no effect on immunoprecipitated peptides derived from translation of total polysomal mRNA. However, in roots
104
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
Table 4. Changes in the levels of extractable GST enzyme activity and immunodetectable protein in germinating shoots and initiated cell cultures of F71 maize
Developmental stage
Enzyme activity (pkat.mg- 1 protein) a Atrazine
Immunodetectableb cpm (%)
Metolachlor
Germinating shoot (day 6)
0.31
7.04
810 (100%)
Initial callus
0.12
14.14
855 (106%)
New cell culture
0.04
16.48
1022 (126%)
Established cell culture
0.02
37.33
1183 (146%)
Enzyme activity expressed as pmol product formed, s -1. rag- 1 protein and presented as mean values of duplicate determinations b Derived from treatment of immune-blotted enzyme with 125Ilabelled protein A
mRNA 'activity' encoding GST was enriched ninefold in antidote-treated tissue as compared with controls. Effect on initiation of undifferentiated growth on G S T in maize shoots. Callus cultures were initiated from germinating shoots of F71 maize seeds and analysed for extractable enzyme activities and immunodetectable protein (Table 4). Following the initiation of callus from 6-d shoots, enzyme activity with atrazine as substrate declined and, after subsequent subculture to give dedifferentiated growth, was reduced to 13% of the level found in the shoots. Similarly, an existing cell line of F71 shoot callus had very low activity toward atrazine (0.02 pkat.mg- ~ protein). In contrast, GST activity toward metolachlor was raised by the initiation and maintenance of cell culture such that the activity in the new cell line was 134% higher than that of 6-d shoots. Total immunodetectable GST was increased during sub-culture and was maximal in the established callus. However, the increase in total enzyme could not account for the much larger proportional increases in GST activity. Discussion
Specificity of the anti-GST serum. The antiserum raised against the isoenzyme of GST from maize leaves responsible for metolachlor conjugation recognized the isoenzymes responsible for atrazine detoxification. Immuneblotting experiments have shown that the three isoenzymes responsible for atrazine and metolachlor conjugation in LG 11 maize all have a subunit Mr of 26000 (Edwards
and Owen 1987). Similarly, when crude extracts from leaves and roots were immuneblotted only the Mr 26000 peptide was recognised, indicating that the antiserum was highly specific. The identity of the antigen as maize GST was subsequently confirmed by immunotitration of the Mr 26000 peptide such that increased immunoprecipitation of the protein corresponded to a directly proportional decline in soluble enzyme activity. When polysomal mRNA from maize leaves was translated in vitro an additional peptide (Mr 28000) was observed which may represent a precursor or related form of the Mr 26 000 subunit. In previous studies, purification of maize leaf GST by bromosulfophthalein-glutathione affinity chromatography showed that an Mr 28 800 peptide was present as a minor component of the purified enzyme (Edwards and Owen 1986a). Recently, it has been reported that an antibody raised against a homodimer of maize GST of subunit Mr 28 000 showed little cross reactivity with a lower-molecular-weight subunit present in a heterodimer form of the enzyme (Timmerman and Tu 1987). At present it remains unclear whether the complexity of GSTsubunit forms in maize arises wholly from gene multiplicity (Moore et al. 1986; Shah et al. 1986; Wiegand et al. 1986) or that post-translational effects give rise to related enzymes of lower molecular weight. Thus an Mr 25 000 peptide of GST has been reported as being derived from an Mr 29000 subunit of the transferase (Wiegand et al. 1986). When immunoprecipitates were prepared from suspension cultures of Black Mexican Sweetcorn labelled with [35S]methionine, two high-molecularweight peptides (Mr 80000, Mr 87000) were specifically recognised. However, these peptides were not observed following immunoprecipitation of products derived from the in-vitro translation of cell polysomes. Similarly no high-molecular-weight forms of GST were reported in extracts of maize leaf mRNA probed with a l.l-kilobase copy cDNA encoding an Mr 29000 subunit of the enzyme (Wiegand et al. 1986). These results indicate that the Mr 80000 and Mr 87000 peptides are not related to GST but share a common epitope with the transferase. Enhancement of G S T in maize seedlings treated with the herbicide antidote DDCA. By using the antiserum to GST in maize it has been possible to study antidote-induced changes in enzyme synthesis in the roots of seedlings 24 h after treatment with 15 gM DDCA. Thus DDCA increased the 'activity' of mRNA encoding GST ninefold in polysomes isolated from roots and this was associated with
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
a 71% increase in immunodetectable Mr 26000 peptide. The elevation in enzyme concentration resulted in a doubling in extractable activity toward the chloracetanilide herbicide metolachlor but had no effect on atrazine conjugation. Previous studies have shown that the isoenzymes in maize leaves responsible for the detoxification of atrazine and metolachlor can be differentiated on the basis of their isoelectric points (Edwards and Owen 1987). The results presented here indicate that DDCA specifically regulates the expression of the gene encoding the isoenzyme of GST active toward metolachlor but has no effect on the isoenzyme responsible for atrazine conjugation. In a recent study a cDNA probe for the mRNA encoding the GST subunit (Mr 29 000) responsible for alachlor conjugation in maize has been prepared and used to study the effect of a seed dressing of DDCA on GST transcription (Wiegand et al. 1986). The mRNA encoding the isoenzyme was increased between three- and fourfold in the leaves of 8-d seedlings grown from antidotedressed seeds as compared with control plants. This increase in transcription corresponded to known increases in the concentration and activity of an isoenzyme of GST in maize leaves which actively conjugated alachlor (Mozer et al. 1983). It is now apparent that a root drench with DDCA can induce an Mr 26 000 subunit of the transferase and that this effect occurs within 24 h in the roots. The failure of the antidote to increase the synthesis of GST in leaves probably results from slow translocation of DDCA to the leaves or detoxification of the compound to inactive metabolites in the roots. This was not confirmed however.
Regulation of GST in cell cultures of Z. mays. When suspension cultures of Black Mexican Sweetcorn were grown in the presence of [35S]methionine, analysis of immunoprecipitates by SDS-PAGE showed that the synthesis of GST was maximal in cells in mid-logarithmic growth phase (5-7 d). Similarly, such actively dividing cultures have been shown to contain higher extractable activities of GST responsible for metolachlor conjugation than was the case in cells in stationary phase (Edwards and Owen 1986a). Treatment of cells with cinnamic acid, metolachlor or DDCA, which are all substrates of GST, enhanced the activity of the enzyme toward metolachlor but this was not accompanied by increased incorporation of [aSS]methionine into the Mr 26000 subunit. In addition, DDCA did not increase the activity of polysomal mRNA in the cells encoding GST and this indicated that the antidote may exert additional
105
post-translational effects in maize. Interestingly DDCA has also been reported to have rapid effects on lipid metabolism in maize cell cultures and to increase cellular glutathione (Ezra and Gressel 1982). Previous studies have shown that cell cultures derived from maize shoots were unable to detoxify atrazine as a consequence of an apparent deficiency in the isoenzyme of GST responsible for chloro-s-triazine conjugation. By initiating a callus culture from the shoot tissue of a maize line (cv. F71) known actively to conjugate atrazine (Edwards and Owen 1986b) it has been possible to show that the loss in isoenzyme activity occurs after primary dedifferentiation. By contrast, metolachlor conjugation was enhanced during the establishment of the callus and this was accompanied by a 46% increase in the total immunodetectable GST. The high concentration of the enzyme active toward metolachlor in suspension cultures of Black Mexican Sweetcorn may explain the very rapid metabolism of the herbicide in this cell line (Cole et al. 1987).
Regulation of pesticide-degrading enzymes in plants. Increasing evidence indicates that enzymes which detoxify herbicides in plants can be influenced by chemicals, thus allowing the selectivity of pesticides to be manipulated (Cole et al. 1987). Studies with the glutathione S-transferases of maize have shown that such regulation may be restricted to specific isoenzymic forms at the level of gene transcription. Such selective regulation of GST isoenzymes may represent a sensitive means of regulating intracellular concentrations of endogenous electrophilic substrates by forming products of greatly increased polarity. In this respect, some herbicide antidotes may mimic the regulatory activity of naturally occurring plant products. In order to modify the activity of additional pesticide-degrading enzymes in plants it will be necessary to study the enzymology and molecular biology of these proteins, which are currently largely uncharacterised. This study was supported by the Agricultural and Food Research Council. The authors would like to thank Sue Robins for the preparation of this manuscript.
References Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248 254 Cole, D.J., Edwards, R., Owen, W.J. (1987) The role of metabo-
106
R. Edwards and W.J. Owen: Regulation of glutathione S-transferases of maize
lism in herbicide selectivity. In: Progress in pesticide biochemistry and toxicology, vol. 6, pp. 57-104, Hutson, D.H.. Roberts, T.R. eds. J. Wiley and Sons, Chichester Diesperger, H., Sandermann (Jr), H. (1979) Soluble and rnicrosornal glutathione s-transferase activities in pea seedlings (Fisum sativum L.) Planta 146, 643 648 Edwards, R., Owen, W.J. (1986a) Comparison of glutathione S-transferases of Zea mays responsible for herbicide detoxification in plants and suspension-cultured cells. Planta 169, 208-215 Edwards, R., Owen, W.J. (1986b) Glutathione S-transferases involved in herbicide detoxification in plant cell culture. Aspects Appl. Biol. 77, 121-130 Edwards, R., Owen, W.J. (1987) lsoenzymes of glutathione Stransferase in Zea mays. Biochem. Soc. Trans. 15, 1184 Ezra, G., Gressel, J. (1982) Rapid effects of a thiocarbamate protectant on rnacrornolecular synthesis and glutathione levels in maize cell cultures. Pestic. Biochem. Physiol. 17, 48-58 Laemrnli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Lai, H-C.J., Tu, C-P.D. (1986) Rat glutathione s-transferase supergene family. J. Biol. Chem. 261, 13793-13799 Moore, R.E., Davies, M.S., O'Connell, K.M., Harding, E.I., Wiegand, R.C., Tierneier, D.C. (1986) Cloning and expression of a cDNA encoding a maize glutathione S-transferase in E. coff. Nucl. Acids Res. 14, 7227-7235 Mozer, T.J., Tierneier, D.C., Jaworski, E.G. (1983) Purification
and characterisation of corn glutathione S-transferase. Biochemistry 22, 1068-1072 Palmiter, R.D. (1974) Magnesium precipitation of ribonucleoprotein complexes. Expedient techniques for the isolation of undegraded polysomes and messenger ribonucleic acid. Biochemistry 13, 3606-3615 Robbins, M.P., Dixon, R.A. (1984) Induction of chalcone isomerase in elicitor-treated bean cells : comparison of rates of synthesis and appearance of irnmunodetectable enzyme. Eur. J. Biochem. 145, 195-202 Shah, D.M., Hironaka, C.M., Wiegand, R.C., Harding, E.I., Krivi, G.G., Tiemeier, D.C. (1986) Structural analysis of a maize gene coding for glutathione S-transferase involved in herbicide detoxification. Plant Mol. Biol. 6, 203-211 Schr6der, J., Betz, B., Hahlbrock, K. (1976) Light-induced enzyrne synthesis in cell suspension cultures of Petroselinum hortense. Eur. J. Bioehern. 67, 527 541 Timmerrnan, K.P., Tu, C-P.D. (1987) Genetic evidence ofxenobiotics metabolism by glutathione S-transferases from corn. In: Glutathione S-transferase and carcinogenesis, pp. 47-49, Mantle, T.J., Pickett, C.B., Hayes, J.D. eds. Taylor and Francis, London New York Philadelphia Wiegand, R.C., Shah, D.M., Mozer, T.J., Harding, E.I., DiazCollier, J., Saunders, C., Jaworski, E.G., Tiermeier, D.C. (1986) Messenger RNA encoding a glutathione S-transferase responsible for herbicide tolerance is induced in response to safener treatment. Plant. Mol. Biol. 7, 235-243 Received 20 Novemer 1987; accepted 17 March 1988
