Planta (1985)165:311-321

P l a n t a 9 Springer-Verlag1985

Tagetitoxin affects plastid development in seedling leaves of wheat J.H. Lukens* and R.D. Durbin Department of Plant Pathology, University of Wisconsin, and Agricultural Research Service, U.S. Department of Agriculture, Madison, WI 53706, USA

Abstract. Ultrastructural and biochemical approaches were used to investigate the mode of action of tagetitoxin, a nonhost-specific phytotoxin produced by Pseudomonas syringae pv. tagetis (Hellmers) Young, Dye and Wilkie, which causes chlorosis in developing - but not mature - leaves. Tagetitoxin has no effect on the growth rate or morphology of developing leaves of wheat (Triticure aestivum L.) seedlings. Its cytological effects are limited to plastid aberrations; in both lightand dark-grown leaves treated with toxin, internal plastid membranes fail to develop normally and plastid ribosomes are absent, whereas mitochondrial and cytoplasmic ribosomes are unaffected. The activity of a plastid stromal enzyme, ribulose1,5-bisphosphate carboxylase (RuBPCase, EC 4.1.1.39), which is co-coded by nuclear and chloroplast genes, is markedly lower in extracts of both light- and dark-grown toxin-treated leaves, whereas the activity of another stromal enzyme, NADP-glyceraldehyde-3-phosphate dehydrogenase (NADP-G-3P-DH, EC 1.2.1.13), which is coded only by the nuclear genome, is significantly lower in extracts of light-grown, but not of darkgrown, treated leaves. The mitochondrial enzymes fumarase (EC 4.2.1.2) and cytochrome-c oxidase (EC 1.9.3.1) are unaffected by toxin in dark-grown leaves, but fumarase activity is reduced in lightgrown ones. Four peroxisomal enzyme activities are lowered by toxin treatment in both light- and dark-grown leaves. Light- and dark-grown, toxintreated leaves contain about 50% and 75%, respectively, of the total protein of untreated leaves. Biological Laboratories, Harvard University, Cambridge, MA 02138, USA Abbreviations. NADP-G-3P-DH = NADP-glyceraldehyde-3phosphate dehydrogenase; PLB=prolamellar body; RuBPCase=ribulose-l,5-bisphosphate carboxylase; SADH=shikimic acid dehydrogenase * Present address:

There are threefold and twofold increases in free amino acids in light-grown and dark-grown treated leaves, respectively. In general, the effects of tagetitoxin are more extensive and exaggerated in lightgrown than in dark-grown leaves. We conclude that tagetitoxin interferes primarily with a lightindependent aspect of chloroplast-specific metabolism which is important in plastid biogenesis. Key words: Chloroplast development - Chloroplast ribosomes - Etioplast (prolamellar body) Pseudomonas (toxin) - T a g e t i t o x i n - Triticum (chloroplast development).

Introduction

Pseudomonas syringae pv. tagetis (Hellmers) Young, Dye and Wilkie causes a leaf spot and apical chlorosis in certain Compositae (Styer and Durbin 1982a, b). The apical chlorosis results from the action of a nonhost-specific exotoxin (Trimboli et al. 1978) which has been given the trivial name tagetitoxin (Mitchell and Durbin 1981). Tagetitoxin differs from the phytotoxins produced by other chlorosis-inducing pseudomonads both in chemical structure and in the nature of the chlorosis it induces. Whereas most known pseudomonad phytotoxins are modified peptides, tagetitoxin consists of an eight-membered ring containing a sulfur atom (Fig. 1) (Mitchell and Durbin 1981; Mitchell and Hart 1983). Also, unlike other pseudomonad toxins, tagetitoxin does not affect pre-existing chlorophyll, but instead prevents the accumulation of chlorophyll in developing leaf tissue (Lukens 1983). The mode of action of tagetitoxin in plants is not known. An ultrastructural study of the effects

312

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat

HO HO2G CH3C-O. ]

HO.

S

shaken in the dark for 12-16 h, and the total chlorophyll (a + b) concentration in the ethanol extracts was calculated according to the method of Wintermans and De Mots (1965).

I

aj+ OH Fig. 1. Proposed structure of tagetitoxin. From Mitchell and Hart (1983)

of tagetitoxin on light-grown Zinnia leaves indicated that abnormalities were restricted to chloroplasts (Jutte and Durbin 1979), which showed severe internal disorganization, including the absence of a stacked thylakoid membrane system and of identifiable plastid ribosomes. Chlorosis and chloroplast ultrastructural aberrations are not always symptomatic of specific disruptions in uniquely plastidial metabolism, but often represent indirect effects on chloroplast development and morphology resulting from perturbation of extra-plastidial cellular processes. We report here the results of a combination of ultrastructural and biochemical approaches designed to ascertain the effects of tagetitoxin on both plastid and non-plastid components of wheat seedling leaf cells. Material and methods Plant material and growth conditions. Seedlings of spring wheat (Triticum aestivum L. cv. Lathrop, University of Wisconsin seed lots, harvested 1978-1982) were grown on filter paper at 22~ C in darkness or 12-h photoperiods (cool-white fluorescent lamps; Sylvania, Danvers, Mass., U S A ; 281 _+ 10 gmol m -2 s-l). The root systems of the seedlings were moistened with either distilled water or 5-10-5 M tagetitoxin. The seedlings were harvested between 6 and 10 d after germination, when the first leaves were 8-12 cm long. Toxin purification. Tagetitoxin was purified by a method modified from Mitchell and Durbin (1981). The initial organic solvent partitioning steps were replaced with a single step in which the culture supernatant was applied to a column of diethylaminoethyl (DEAE) Sephadex A-25 (4.6 cm diameter, 30 cm long) equilibrated with 0.05 M KNO 3 . Toxin was eluted with 0.4 M NH4HCO 3 at approximately twice void volume in three 50-ml fractions. Subsequent purification steps followed the published protocol. Based on its specific activity in a Zinnia apical chlorosis bioassay, the isolated toxin was estimated to be 90% pure. Effect of tagetitoxin on chlorophyll accumulation in developing wheat seedling leaves. Wheat seeds were germinated on filter paper disks kept moist with 10 4-10-8 M tagetitoxin. After 6 d growth in the light, eight to nine leaves were harvested from each of three replicate groups of seedlings and were washed briefly with deionized water, cut into 1-cm pieces, and added to vials containing 95% ethanol. The vials were gently

Electron microscopy. Samples (1-2 mm long) were taken from about 6 cm above the base of first leaves of 7-d-old seedlings grown in the light or dark with water or 5.10- 5 M tagetitoxin. The tissues were fixed with 5% glutaraldehyde in 0.08 M Nacacodylate buffer (pH 7.4) for 45 rain at 4 ~ after vacuum infiltration of the glutaraldehyde. Tissue from dark-grown leaves was fixed under green light. Fixed tissue pieces were rinsed for i h in two changes of 0.08 M cacodylate buffer (pH 7.4) in 1.5% sucrose, post-fixed overnight at 4 ~ C in Palade's fixative containing 2% osmium, and then dehydrated in a graded acetone series at 22~ which included staining in saturated uranyl acetate overnight (4 ~ C) at the 70%-acetone step. Samples were embedded in Spurr's epoxy resin, sectioned, stained with lead citrate, and viewed in a JEM 7 transmission electron microscope (Japanese Electron Optics Co., Tokyo, Japan). Preparation of leaf extracts of enzyme assays. The apical 5 mm and basal 10-15 mm of the seedling leaf blades (including the coleoptile) were removed; coleoptiles were not removed from dark-grown leaves. Portions of 10-20 first leaves of 6-d-old seedlings were ground at 4 ~ in a motar with 5-10 ml of 50 mM N-[2-hydroxy-/-l-bis (hydroxymethyl)ethyl]glycine (Tricine)-NaOH (pH 7.5), 5 mM dithioerythritol, and I mM ethylene diaminetetraacetic acid (EDTA). Insoluble polyvinylpyrrolidone was added to a final concentration of 2% (w/v). Extracts of dark-grown leaves were prepared under green light. Extracts were stirred gently at 4~ C for 1 h, squeezed through one layer of Miracloth (Calbiochem-Behring, San Diego, Cal., USA), and centrifuged at 120 g for 5 min. Supernatants were kept at 4 ~ C and used for the determination of enzyme activities within 72 h. Total protein per first leaf was determined according to Bradford (1976). Enzyme assays. Marker-enzyme activities in leaf extracts were assayed spectrophotometrically at 24 ~ C. Fumarase (EC 4.2.1.2), catalase (EC 1.11.1.6), glycolate oxidase (EC1.1.3.1) and shikimic acid dehydrogenase (SADH; EC 1.1.1.25) were assayed according to protocols used by Feierabend (1975) and Feierabend et al. (1972a, 1976, 1978) for wheat leaf extracts. The empirical extinction coefficient, E 2' ~4 0 = 2.10, was used to calculate the specific activity of fumarate formation by fumarase. Catalase activity was calculated with mM E24o=36.0 for hydrogen peroxide (Lfick 1965). The rate of oxidation of reduced cytochrome c by cytochrome-c oxidase (EC 1.9.3.1) was monitored at 550 nm according to a modification of the method of Tolbert et al. (1968). Between 25 and 50 gl of leaf extract were incubated for 8-10 min at 22 ~ C with an equal amount of 4% digitonin before adding the mixture to 0.1 M potassium phosphate (pH 7.0) containing 28 ~tM cytochrome c reduced with sodium dithionite. The activity of hydroxypyruvate reductase (EC 1.1.1.81) was assayed as described by de Boer and Feierabend (1974). Serine-glyoxylate aminotransferase (EC 2.6.1.45) activity was assayed using a coupled system (D. Hondred, University of Wisconsin, Madison, USA, personal communication. The reaction mixture contained 70raM 4-(2-hydroxyethyl)-l-piperazine ethanesulfonic acid (Hepes) (pH 8.2), 0.3 mM fl-NADH, 4 mM glyoxylate, 0.1 mM pyridoxal-5-phosphate, 0.03 units hydroxypyruvate reductase (glyoxylate reductase No. G-5259; Sigma Chemical Co., St. Louis, Mo., USA), and 20 mM serine. Serine-glyoxylate aminotransferase activity was monitored by determining the rate of fl-NADH oxidation during the conversion of hydroxypyruvate

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat II0

313

I

I

100 - -

-tJ

3

O

tY

90

I-

Z 0

80

v

70--

I-

60

O

Z

hl I-Z O to

J .2

>-r

13_ 0

S --r to

50

/

40 30

20 10-0

~176

4

5

6

7

8

(-LOGI0 M) Fig. 3. Chlorophyll content of the first leaves of 6-d-old wheat seedlings as a function of the tagetitoxin concentration in which the seedlings were grown. Chlorophyll content is expressed as a percentage of the chlorophyll in leaves of control seedlings grown in HaO. Plotted values represent the means _+SE of three replicates of eight or nine seedlings TAGETITOXI N

CONCENTRATION

Fig. 2. Six-day-old wheat seedlings grown in 12-h photoperiods

and treated with water (/eft) or 5.10 5 M tagetitoxin (right). Arrow points to section of leaf blade which was sampled for electron microscopy to glyceric acid. Ribulose-l,5-bisphosphate carboxylase activity was assayed according to the coupled enzyme procedure of Racker (1962). Assay of NADP-glyceraldehyde-3-phosphate dehydrogenase was according to Mfiller et al. (1969) after incubating extract aliquots with 2.75 mM N A D P H and 27.5 mM dithioerythritol at 24~ for 30 min. Peroxidase (EC 1.11.1.7) was assayed at 470 nm by the formation of tetraguaiacol from the peroxidation of guaiacol (Nadolny 1978 ; Lee 1973). Specific activity was calculated with E~vo-6.67 for tetraguaiacol (George 1953). m M

__

Extraction and determination of free amino acids. Amino acids were extracted by a modification of the methods of Bieleski and Turner (1966) and Mitchell and Bieleski (1977). Leaf tissue fi-om 6-d-old seedlings (200 mg fresh weight) was ground in a Duall C homogenizer (Kontes, Vineland, N.J., USA) with 5 ml methanol: chloroform : water (3 : 1 : 1, by vol.). The homogenate was filtered and the filter cake was washed with 10 ml additional solvent. The combined filtrate and washings were evaporated to dryness on a rotary evaporator, and the residue was taken up in 1 2 ml water and extracted with 8 ml chloroform. After centrifugation (1-2 min, 4000 g), the aqueous phase was removed and the chloroform phase was re-extracted with 2 ml water. The aqueous phases were pooled and evaporated to dryness. The residue was taken up in 1-2 ml water and analyzed with an amino acid analyzer.

Results

Characteristics of seedlings grown in tagetitoxin. Tagetitoxin has little or no effect on the morphology, growth rate or fresh weight of developing wheat seedlings (Fig. 2). The only visual effect of toxin treatment is on leaf pigmentation. Leaves of light-grown, toxin-treated seedlings lack chlorophyll and are ivory-white or pale yellow (with a slight pink hue caused by other pigments). Leaves from dark-grown, untreated seedlings are bright yellow, but toxin-treated leaves are white with pale-yellow tips. Both light- and dark-grown, toxin-treated leaves have less protein than the corresponding untreated leaves. The depletion of protein as a result of toxin treatment is approx. 60% in light-grown and 25% in dark-grown leaves. Effect of tagetitoxin on chlorophyll accumulation in developing wheat seedling leaves. Leaf chlorophyll content is dependent on the concentration of tagetitoxin supplied during the growth of wheat seedlings (Fig. 3). A salient characteristic of the response curve is its steep slope: nearly the full

Fig. 4A, B. Sections of leaves of 7-d-old wheat seedlings grown under 12-h photoperiods either in A HzO (x 38160) or in B 5-10 -5 M tagetitoxin ( x 35280). CP, chloroplast; E, envelope; G, osmiophilic plastoglobule; M, mitochondrion; P, peroxisome; T, thylakoid; V, vacuole; VE, vesicular aggregation; W, cell wall. Slender arrowheads point to ribosomes in the chloroplast and mitochondria; arrows point to cytoplasmic ribosomes. Thick arrowhead (B) points to a DNA-like fibril. Bars = 0.5 gm

J.H. Lukensand R.D. Durbin: Tagetitoxineffectson plastid developmentin wheat response range (i.e., 0-100% of control) occurs within only one order of magnitude change in toxin concentration. Based on these results, solutions of from 2.10 s to 5 - 1 0 - 5 M tagetitoxin were used to determine the ultrastructural and physiological effects of tagetitoxin on seedling leaves. Effects of tagetitoxin on ultrastructure of lightgrown leaves. Figure 4A illustrates a portion of a normal oblong chloroplast, typically associated with mitochondria and peroxisomes and located near the mesophyll cell wall in cytoplasm peripheral to the large central vacuole. Although, in general, the cellular components of toxin-treated, lightgrown wheat leaves stain less darkly overall than those of untreated leaves, the specific ultrastructural aberrations caused by tagetitoxin are essentially limited to the chloroplast (Fig. 4 B). Although the aberrant chloroplast has an intact envelope, it lacks any semblance of the normal internal plastid membrane system and contains instead only a circular string of beaded vesicles surrounding an aggregation of plastoglobuli. The stroma lacks most of the dark-staining, largely proteinaceous ground substance normally present in plastids, and there are no ribosomes. Frequent "windows" in plastid profiles, through which cytoplasm and cytoplasmic ribosomes can be seen (Fig. 4B), are evidence of invagination or collapse of the "bag-like" organelles, which have lost their three-dimensional integrity. There are not noticeably fewer plastids in toxin-treated than in control cell sections, but the rudimentary plastids are approximately one-third the size of normal plastids. Other cellular components of toxin-treated, light-grown leaves are either essentially normal or display far more subtle effects of the toxin. Mitochondria appear to be unaffected by toxin in size, shape, internal structure, and number per cell; peroxisomes, however, appear to be smaller, perhaps fewer in number, and less commonly associated with plastids than their counterparts in untreated cells. Both mitochondria and peroxisomes appear to have somewhat less darkly-stained matrices than normal. Mitochondrial and cytoplasmic ribosomes are both present in apparently normal numbers in toxin-treated cells. Effects of tagetitoxin on ultrastructure of darkgrown leaves. The mesophyll cells of dark-grown, untreated leaves contain many large, rounded etioplasts which have an internal membrane system consisting of several highly-organized prolamellar bodies (PLBs), from which radiate numerous prothylakoids (Fig. 5A). There are ribosomes in the

315

plastid stroma and within the PLB lattices, where they appear to be arranged at regular intervals inside the patterned arrays of membranous tubules. As in cells of light-grown leaves, mitochondria and peroxisomes are located close to plastids in the peripheral cytoplasm. The effects of tagetitoxin on cells of darkgrown leaves are limited to changes in the plastids: both ribosomes and normal, fully-developed PLBs are absent from etioplasts of treated leaves (Figs. 5 B, 6 B). As in light-grown, treated leaves, the depletion of plastid ribosomes is a markedly specific effect; mitochondrial and cytoplasmic ribosomes are abundant (Fig. 5 B). The rudimentary PLB of toxin-treated etioplasts consists of a largely unorganized aggregate of vesiculated membranes associated with a group of plastoglobuli. Although some of the more organized instances of these PLB-like aggregates (Fig. 6B) resemble the authentic PLB of a normal etioplast (Fig. 6A), there are important differences between the two membrane structures. The PLB-like structure of treated etioplasts is not organized in a regular three-dimensional array of subunits, it lacks radiating prothylakoids, it is usually present in only one " c o p y " per etioplast, and plastid ribosomes, which appear prominently within the tight lattice of the normal PLB (Fig. 6A), are absent from the loosely-organized, rudimentary PLB (Fig. 6 B). Apart from its marked effect on ribosomes and PLB structure, tagetitoxin disrupts etioplasts much less severely than it does chloroplasts of lightgrown leaves. The etioplast stroma is not cleared of staining ground material as is the stroma of affected chloroplasts; phytoferritin is also present in apparently normal amounts in the stroma of toxin-treated etioplasts (Fig. 5B). The disruption of the internal membrane is notably less severe in etioplasts than in chloroplasts: in both the extent of their elaboration and the complexity of their structure, the rudimentary PLBs resemble normal PLBs far more closely than the few beaded vesicles of the toxin-treated chloroplast resemble the elaborate granal stacks of the normal chloroplast. All other components of toxin-treated, etiolated cells, including mitochondria and peroxisomes, stain normally and appear to be unaffected by the toxin. Peroxisomes, however, are more difficult to locate in sections of toxin-treated cells and may therefore be reduced in number. Mitochondrial enzymes. The activities of fumarase (a soluble matrix enzyme) and cytochrome c oxidase (membrane-bound) are not significantly affected by tagetitoxin in extracts of dark-grown

Fig. 5A, B. Sections of leaves of 7-d-old wheat seedlings grown in the dark either in A H20 (x 34560) or in B 5.10-5 M tagetitoxin ( x 43 200). E, envelope; EP, etioplast; G, osmiophilic globule; M, mitochondrion; PF, phytoferritin; PLB, prolamellar body; PT, prothylakoid; V, vacuole; VE, vesicular aggregation; W, cell wall. Arrowheads point to ribosomes in etioplasts and mitochondria; arrows point to cytoplasmic ribosomes. Bars--0.5 lam

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat

317

Fig. 6. A Prolamellar body (PLB) and B PLB-like aggregation of membranes of etioplasts of 7-d-old wheat seedling leaves grown either in A H 2 0 ( x 56700) or B 5. l0 -5 M tagetitoxin ( x 65200). Arrowheads point to plastid ribosomes. Bars=0.5 gm Table 1. Activities of marker enzymes in cell-free extracts of the first leaves" of 6-d-old wheat seedlings grown in water or in 5 . 1 0 - 5 M tagetitoxin, either in the light or dark. Enzyme activities are expressed as nmol min - 1, except for catalase and peroxidase activities, which are expressed as ~tmol min - 1. Results are means _+SD of two to four determinations from separate extractions Enzyme

1. Mitochondrial a. Fumarase b. Cytochrome c oxidase 2. Chloroplastic a. Ribulose-bisphosphate carboxylase b. NADP-glyceraldehyde-3-P dehydrogenase c. Shikimicacid dehydrogenase b 3. Peroxisomal a. Catalase b. Hydroxypyruvate reductase c. Glycolateoxidase d. Serine-glyoxylate aminotransferase c 4. General (and vacuolar) a. Peroxidase a b ~ d

EC

Light

Dark

H20

Toxin

Toxin/ H 2 0 (%)

4,2.1.2 1.9.3.1

108.8_+_31.0 56.0_+ 11.4

54.5_+ 2.2 ~ 50 44.0_+11.5 79

4.1.1.39

26.6+ 3.5

2.0_+ 0.7 d

7

8.6_+ 2.0

1.2.1.13

119.5+17.7

22.5_+ 9.2 a

19

21.0+ 1.4

19.2_+ 1.1

91

1.1,1.25

21.9_+ 1.4

25.4_+ 3.8

116

22.3•

26.6_+ 2.8

119

1.11.1.6 1.1.1.81 1,1.3.1 2.6.1.45

74.3• 9.8 193.6+10.7 24,8+_ 6.7 24.6

1,11.1.7

1.3+ 0.0

11.1_+ 3,9 d 15 38.8_+ 7.4 a 20 4.1-+ 2.7 d 17 2.8 d 11

1,3_+ 0.0

100

HzO

Toxin

Toxin/ HzO (%)

83,5_+ 9.1 78.7+ 6.4

69.1_+18.1 65.3_+ 6.8

83 83

2.3

34.3_+12.0 72.2_+ 0.2 3.4• 1.6 1.1

3,9_+ 0.2

1.8+ 0.2 a

21

23.2+ 7.0 d 68 14.2_+ 1.6 d 20 0.6_+ 0.4 ~ 19 0.3 n 27

3.9_+ 0.1

100

First leaf minus 5-mm tip and basal 10-15 m m Localized in both chloroplasts and cytoplasm Data from one experiment Activities are significantly different from controls at P_< 0.05 (t-test)

leaves (Table 1). In light-grown leaves, however, the toxin causes a 50% reduction in fumarase activity.

Chloroplastic enzymes. Tagetitoxin greatly lowers the specific activity of RuBPCase in extracts of dark-grown leaves, but does not reduce the activity of N A D P - G - 3 P - D H (Table 1). Thus, the toxin af-

fects a chloroplast enzyme which is partially made on plastid ribosomes (RuBPCase) (Blair and Ellis 1973), but does not affect another chloroplast enzyme (NADP-G-3P-DH) which is synthesized outside the plastid (Ellis and Hartley 1971). Whereas RuBPCase and NADP-G-3P-DH activities exhibit light induction (about three- and sixfold, respectively) in untreated leaves, there is

318

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat

no similar light enhancement of these activities in toxin-treated leaves. As a result, the levels of both RuBPCase and NADP-G-3P-DH activities in light-grown, toxin-treated leaf extracts are substantially lower than in extracts of untreated leaves (Table 1). The activity of the nuclear-encoded and partially chloroplast-localized SADH (Feierabend and Brassel 1977) is slightly elevated in extracts of both light- and dark-grown, toxin-treated leaves.

Peroxisomal enzymes. Tagetitoxin treatment significantly reduces the activities of all four peroxisomal enzymes in extracts of both light- and darkgrown leaves (Table 1). The activities of hydroxypyruvate reductase, glycolate oxidase, and serineglyoxylate aminotransferase in extracts of toxintreated leaves are only about 20% of those in untreated leaf extracts. However, these activities all of which are normally light-induced during peroxisomal biogenesis (Feierabend 1975) - are greater in extracts of light-grown than of darkgrown, toxin-treated leaves, and the light-induced increases are of about the same magnitude in both treated and untreated leaves. Therefore, although tagetitoxin significantly reduces the absolute levels of these three peroxisomal enzyme activities, it does not appear to interfere with the developmental mechanism of their light induction. Tagetitoxin treatment affects catalase, the fourth peroxisomal marker enzyme, somewhat differently. Catalase activity in dark-grown leaves is not decreased by tagetitoxin treatment to the extent that the activities of other peroxisomal enzymes are. Moreover, while there is a significant light induction of catalase activity (2.2-fold) in untreated leaves, there is no apparent light induction of enzyme activity in toxin-treated leaves; therefore, catalase activity in light-grown, treated leaves is only 15% of that in untreated leaves. Peroxidase activity. Peroxidase was assayed as a general reference enzyme which is not confined to a single cell compartment, although 50-70% of cellular peroxidase activity has been reported to be localized in vacuoles (Boller and Kende 1979; Grob and Matile 1980). Tagetitoxin has no effect on extractable peroxidase activity in leaves grown either in the light or dark. Effects of tagetitoxin on free amino acids in leaves. The pool sizes of total extractable free amino acids are significantly greater in both dark- and lightgrown, toxin-treated leaves than in the respective untreated leaves (approximately two- and three-

Table 2. Effects of tagetitoxin on the free amino acid pools of first leaves of 6- to 7-d-old wheat seedlings grown either in the dark or light. Seedlings were grown either in water or 5 - 1 0 - 5 M tagetitoxin. Values are ~tmol amino acids/200mg FW leaf tissue and are means + SD of three experiments"

Treatment

Dark

Light

H20 Tagetitoxin

8.3_-4-2.0 13.8_+0.4 b

%6_+0.5 23.4_+1.9 b

a Fourteen amino acids identified by comparison with a standard calibration mixture (Pierce): alanine, arginine, aspartic acid, glutamic acid, glycine, histidine, isoleucine, lysine, methionine, phenylalanine, serine, tyrosine and valine b Toxin effect on pool size significant at P_> 0.05 (t-test)

fold, respectively; Table 2). In dark-grown leaves, tagetitoxin causes increases in the levels of all fourteen individual amino acids assayed, with average increases ranging from 1.2- to 2.8-fold. Although most free amino acids also increase in light-grown, treated leaves, there is a significant depletion of tyrosine and phenylalanine to 60% and 50% of the controls, respectively. Discussion

The evidence presented here indicates that tagetitoxin interferes primarily with a light-independent aspect of chloroplast-specific metabolism. Furthermore, earlier experiments have shown that the toxin affects developing, rather than mature, plastids (Lukens 1983). Toxin interference with plastid development in both light- and dark-grown leaves includes: 1)an abnormal plastid internal membrane system; 2)the lack of plastid ribosomes in electron micrographs; and 3) the absence or significant depletion of the activity of a major plastid enzyme, RuBPCase. The physical characteristics of wheat seedlings grown in tagetitoxin support the notion of a chloroplast-specific mode of action of the toxin. The normal morphology and growth rate of toxintreated seedlings bear witness to the absence of a general interference with cellular metabolism. Furthermore, the fact that chlorotic leaf tissues are capable of regreening when toxin is removed from the growing medium (Lukens 1983) indicates that the cells are not irreversibly debilitated by the effects of toxin. Tagetitoxin, therefore, apparently exerts a specific and reversible effect on the chloroplasts of leaf cells. The effects of tagetitoxin on leaf cell ultrastructure - as well as on enzymes and amino acid levels - are more pronounced and less chloroplast-specif-

J.H. Lukens and R,D. Durbin: Tagetitoxin effects on plastid development in wheat

ic in light-grown than in dark-grown wheat seedlings. It is reasonable, therefore, to conclude that effects on dark-grown leaf cells represent primary results of tagetitoxin action, whereas the additional or more exaggerated - effects on light-grown leaves are secondary. Although internal membranes of plastids in light-grown, toxin-treated tissues are extremely degenerate, micrographs of treated etioplasts show that tagetitoxin does not totally prevent membrane synthesis, but rather selectively inhibits the production, function, or stability of some membrane component(s) required for normal PLB symmetry and prothylakoid development. The differential effect of tagetitoxin on the activities of plastid stromal enzymes (RuBPCase and NADP-G-3P-DH) indicates that, whereas plastidcontrolled processes are inhibited, at least some nuclear-directed processes which contribute to plastid development are unaffected by the toxin. Additionally, the activity of the cytoplasmically synthesized SADH is not reduced by the toxin, indicating that the plastid component of the cellular activity is probably unaffected. Therefore, the treated etioplast must be competent for uptake of cytoplasmically synthesized polypeptides, or, alternatively, NADP-G-3P-DH and SADH must be stable in the cytoplasm in the absence of their transport into the etioplast. Further indirect evidence for the normal synthesis and-or functioning of nuclear-encoded chloroplast components can be drawn from experiments which demonstrated that toxin-treated cucumber cotyledons can synthesize protochlorophyllide from exogenous 6-aminolevulinic acid and subsequently photoreduce it to chlorophyllide (Lukens 1983). The chlorophyll biosynthetic enzymes are known to be nuclear-encoded, and the terminal enzyme responsible for photoreduction, NADPH-protochlorophyllide oxidoreductase (EC 1.6.99.1) is a major component of the etioplast internal membrane system (Castelfranco and Beale 1981; Griffiths and Beer 1982). The reduction of NADP-G-3P-DH activity in lightgrown treated leaves probably represents general secondary damage to constituent proteins of the severely disrupted chloroplasts, as indicated by the stromal clearing (Fig. 4B) and loss of 57% of the total leaf protein. Alternatively, cytoplasmically synthesized chloroplast components may not be light-induced in the absence of functional chloroplasts. The effects of tagetitoxin on peroxisomes are difficult to explain within the context of a chloroplast-specific mode of action of the toxin. The size reduction of peroxisomes in light-grown tissue, as

319

noted in electron micrographs, and the great decreases in extractable peroxisomal enzyme activities in both light- and dark-grown toxin-treated leaves could be explained as secondary effects of a primary disruption of plastid development and function in view of the close metabolic relationship between the two organelles (see review by Beevers 1979). Although normal peroxisomal development has been reported in cereal leaves when chloroplast development is inhibited (Gruber etal. 1972; Feierabend and Beevers 1972b; Feierabend and Schrader-Reichhardt 1976; Feierabend and Mikus 1977), sharp decreases in catalase and other peroxisomal enzyme activities are often symptomatic of chlorotic tissues with aberrant chloroplast morphology (Eyster 1950; Feierabend and Schubert 1978; Feierabend and Kemmerich 1983; Reiss et al. 1983). Until more is known about the metabolic and developmental interdependence ofperoxisomes and chloroplasts, it will be difficult to determine whether tagetitoxin primarily affects plastid development and only secondarily affects peroxisoreal development, or vice-versa, or whether both organelles are simultaneously affected by toxin interference with some developmental signal which coordinates their joint biogenesis. There is no major increase or decrease of individual amino acids in tagetitoxin-treated leaves which would indicate that the toxin induces chlorosis through a primary effect on some aspect of nitrogen metabolism. The general increase in all free amino acids in toxin-treated tissues is similar to that reported in leaves treated with herbicides known to inhibit carotenoid biosynthesis (for review, see Fedtke 1982). The depletion of tyrosine and phenylanine by tagetitoxin resembles the effect of the chlorosis-inducing herbicide glyphosate (Nilsson 1978) which is known to inhibit aromatic amino acid biosynthesis (Steinrficken and Amrhein 1980); however, because tyrosine and phenylalanine levels are not affected by tagetitoxin in darkgrown leaves, where there are other notable effects of the toxin on plastid development, an interference with the biosynthesis of aromatic amino acids cannot be considered a primary effect of tagetitoxin. Our results show that plastids whose biogenesis is disrupted by tagetitoxin do, in fact, develop in both size and complexity beyond the proplastid stage. In addition, the numbers of plastids per cell, as estimated from electron micrographs and observations of protoplasts (not shown), are similar in cells from untreated and toxin-treated leaves, indicating that tagetitoxin does not extensively interfere with chloroplast division. We cannot yet dis-

320

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat

tinguish between the possibilities of direct toxin action on chloroplast metabolism and indirect toxin action through interference with nuclear-encoded chloroplast components such as constituent molecules of the PLB or plastid ribosomes. Clearly, aspects of chloroplast (or nuclear) metabolism involving the biogenesis, stability, or function of plastid ribosomes are excellent candidates for the primary site of action of tagetitoxin: it is evident that ribosomes are not merely secondarily destroyed by photooxidation, and their absence could of course explain a loss of chloroplast-made proteins, such as RuBPCase. Indeed, in many of their physical and biochemical characteristics, tagetitoxin-treated wheat seedlings resemble plants with heat-induced plastid ribosome deficiencies. These plants similarly exhibit both inhibition of plastid protein synthesis and a failure to green (heat-bleaching) (Feierabend and Schrader-Reichhardt 1976; Feierabend 1977; Feierabend and Mikus 1977; Griffiths and Beer 1982). Furthermore, because plastid ribosomes undergo a large increase early in the development of a wheat leaf cell and subsequently demonstrate little or no turnover of ribosomal RNA in mature cells (Dean and Leech 1982), the inhibition by tagetitoxin of an initial or early step in plastid ribosome biosynthesis would be consonant with both the specific effect of tagetitoxin on an early event in leaf development. from the meristem and the apparent inability of the toxin to affect plastids in mature cells (Lukens 1983). We gratefully acknowledge the assistance of Gary Gaard in the use of the electron microscope and the preparation of micrographs. We also thank T.F. Uchytil for assistance in amino acid analysis and D. Schaefer for typing the manuscript.

References Beevers, H. (1979) Mierobodies in higher plants. Annu. Rev. Plant. Physiot. 30, 159-193 Bieleski, R.L., Turner, N.A. (1966) Separation and estimation of amino acids in crude plant extracts by thin-layer electrophoresis and chromatography. Anal. Biochem.17, 278593 Blair, G.E., Ellis, R.J. (1973) Protein synthesis in chloroplasts. I. Light-driven synthesis of the large subunit of Fraction I protein by isolated pea chloroplasts. Biochim. Biophys. Acta 319, 223-234 Boller, T., Kende, H. (1979) Hydrolytic enzymes in the central vacuole of plant cells. Plant Physiol. 63, 1123-1132 Bradford, M.M. (1976) A rapid and sensitive method for quantitation of gg quantities of protein using the principle of protein-dye binding. Anal. Biochem. 72, 248-254 Castelfranco, P.A., Beale, S.I. (1981) Chlorophyll biosynthesis. In: Stumpf, P.K., Corm, E.E., eds. The biochemistry of plants, vol. 8, pp. 375-421, Academic Press, New York London Dean, C., Leech, R.M. (1982) Genome expression during normal leaf development. I. Cellular and chloroplast numbers

and DNA, RNA, and protein levels in tissues of different ages within a seven day old wheat leaf. Plant Physiol. 69, 904-910 De Boer, J., Feierabend, J. (1974) Comparison of the effects of cytokinins on enzyme development in different cell compartments of the shoot organs of rye seedlings. Z. Pflanzenphysiol. 71, 261-270 Ellis, R.J., Hartley, M.R. (1971) Sites of synthesis of chloroplast proteins. Nature New Biol. 233, 193 196 Eyster, H.C. (1950) Catalase activity in pigment-deficient types of corn. Plant Physiol. 25, 630-638 Fedtke, C. (1982) Biochemistry and physiology of herbicide action. Springer, New York Heidelberg Berlin Feierabend, J. (1975) Developmental studies on microbodies in wheat leaves. III. On the photocontrol of microbody development. Planta 123, 63-77 Feierabend, J. (1977) Capacity for chlorophyll synthesis in heatbleached 70S ribosome-deficient rye leaves. Planta 135, 83-88 Feierabend, J., Beevers, H. (1972a) Developmental studies on microbodies in wheat leaves. I. Conditions influencing enzyme development. Plant Physiol. 49, 28-32 Feierabend, J., Beevers, H. (1972b) Developmental studies on microbodies in wheat leaves. II. Ontogeny of particulate enzyme associations. Plant Physiol. 49, 33-39 Feierabend, J., Brassel, D. (1977) Subcellular localization of shikimate dehydrogenase in higher plants. Z. Pflanzenphysiol. 82, 334-346 Feierabend, J., Kemmerich, P. (1983) Mode of interference of chlorosis-inducing herbicides with peroxisomal enzyme activities. Physiol. Plant. 57, 346-351 Feierabend, J., Mikus, M. (1977) Occurrence of a high temperature sensitivity of chloroplast ribosome formation in several higher plants. Plant Physiol. 59, 863 867 Feierabend, J., Schrader-Reichhardt, U. (1976) Biochemical differentiation of plastids and other organelles in rye leaves with a high temperature-induced deficiency of plastid ribosomes. Planta 129, 133-145 Feierabend, J., Schubert, B (1978) Comparative investigation of the action of several chlorosis-inducing herbicides on the biogenesis of chloroplasts and leaf microbodies. Plant Physiol. 61, 10121022 George, P. (1953) Intermediate compound formation with peroxidase and strong oxidizing agents. J. Biol. Chem. 201, 413-426 Griffiths, W.T., Beer, N.S. (1982) Site of synthesis of NADPH protochlorophyllide oxidoreductase in rye (Secale cereale). Plant Physiol. 70, 1014-1018 Grob, K., Matile P (1980) Compartmentation of ascorbic acid in vacuoles of horseradish root cells. Z. Pflanzenphysiol. 98, 235-243 Gruber, P.J., Becker, W.M., Newcomb, E. (1972) The occurrence of microbodies and peroxisomal enzymes in achlorophyllous leaves. Planta 105, 114-138 Jutte, S.M., Durbin, R.D. (1979) Ultrastructural effects in zinnia leaves of a chlorosis-inducing toxin from Pseudomonas tagetis. Phytopathology 69, 839-843 Lee, T.T. (1973) On extraction and qnantitation of plant peroxidase isoenzymes. Physiol. Plant. 29, 198-203 Lfick, H. (1965) Catalase. In: Bergmeyer, H.U., ed. Methods of enzymatic analysis, pp. 885-894, Academic Press, New York London Lukens, J.H. (1983) Investigations into the mode of action of tagetitoxin in plants. Ph.D. thesis, University of Wisconsin, Madison, USA Mitchell, R.E., Bieleski, R.L. (1977) Involvement of phaseolotoxin in the halo blight of beans. Plant Physiol. 60, 723-729

J.H. Lukens and R.D. Durbin: Tagetitoxin effects on plastid development in wheat Mitchell, R.E., Durbin, R.E. (1981) Tagetitoxin, a toxin produced by Pseudomonas syringae pv. tagetis: purification and partial characterization. Physiol. Plant Pathol. 18, 157-168 Mitchell, R.E., Hart, P.A. (1983) The structure of tagetitoxin, a phytotoxin of Pseudornonas syringae pv. tagetis. Phytochemistry 22, 1425-1428 Mfiller, B., Ziegler, I. Ziegler, H. (1969) Lichtinduzierte, reversible Aktivitfitssteigerung der NADP-abh/ingigen Glycerinaldehyd-3-Phosphatdehydrogenase in Chloroplasten. Zum Mechanismus der Reaktion. Eur. J. Biochem. 9, 101-106 Nadolny, L. (1978) The involvement of peroxidase in protection of tobacco against the hypersensitive reaction induced by the Pseudomonas solanacearurn B1 isolate. M.S. thesis, University of Wisconsin, Madison, USA Nilsson, G. (1978) Effects of glyphosate on the amino acid content in spring wheat plants. Weed Abstr. 27, 407 Racker, E. (1962) Ribulose diphosphate carboxylase from spinach leaves. Methods Enzymol. g, 226-270 Reiss, T., Bergfeld, R., Link, G., Thien, W., Mohr, H. (1983) Photooxidative destructive of chloroplasts and its consequences for cytosolic enzyme levels and plant development. Planta 159, 518-528

321

Steinriicken, H.C., Amrhein, N. (1980) The herbicide glyphosate is a potent inhibitor of 5-enolpyruvyl-shikimic acid-3phosphate synthase. Biochem. Biophys. Res. Commun. 94, 1207 Styer, D.J., Durbin, R.D. (1982a) Common ragweed: a new host for Pseudomonas syringae pv. tagetis. Plant Dis. 66, 71 Styer, D.J., Durbin, R.D. (1982b) Isolation of Pseudomonas synringae pv. tagetis from sunflower in Wisconsin. Plant Dis. 66, 701 Tolbert, N.E., Oeser, A., Kisacki, T., Hageman, R.H., Yamazaki, R.K. (1968) Peroxisomes from spinach leaves containing enzymes related to glycolate metabolism. J. Biol. Chem. 243, 5179-5184 Trimboli, D., Fahy, P.C., Baker, K.F. (1978) Apical chlorosis and leaf spot of Tagetes spp. caused by Pseudomonas tagetis Hellmers. Aust. J. Agric. Res. 29, 831-839 Wintermans, J.F.G.M., De Mots, A. (1965) Spectrophotometric characteristics of chlorophylls a and b and their pheophytins in ethanol. Biochim. Biophys. Acta 109, 448-453 Received 2 October 1984; accepted 11 March 1985