Volume 45, Issue 5

International Workshop at Monheim , Germany, August 13 -15 , 1989, in honor of the 60th birthday of Achim Trebst

Studies on structure/activity relationships of phloroglucinol derivatives that had been designed based on the structures of grandinol and homograndinol. potent photosystem II (PS II) inhibitors in Eucalyptus grandis, revealed that two electron-withdrawing groups which differ by their electron-withdrawing power on a phloroglucinol nucleus were essential for activity. A larger difference in the electron-withdrawing power between the two groups enhanced the activity, and 3-nitro-phloroglucinecarboxamides and the corresponding thioamides were the most active ones when they had proper lipophilic side chains. Their binding domain seems to overlap with those of DCMU and atrazine, whereas they may approach to the site in a similar manner to that of phenol type inhibitors. Accordingly, the phloroglucinol derivatives represent a new type of PS II inhibitors.

A third group of antibiotics, the aurachins, have been isolated from the myxobacterium Stigmatella aurantiaca. The aurachins chemically are quinolones. and four of them (aurachins A-D) have been tested for their inhibitory activity in photosystem II and cytochrome b/c-complexes. Aurachin C is the best inhibitor in photosystem II (pI 50 -value 7.2); its biochemical behaviour being the same like that of other photosystem II herbicides. Both aurachins C and D are also excellent inhibitors in the cytochrome b 6 /f-complex (pI 50 -values of 7.00 and 7.49. respectively). In its mechanism of action, aurachin C resembl|es antimycin, whereas aurachin D is different from either antimycin or myxothiazol.

Herbicides inhibiting the photosynthetic electron transport (PET) system may be an important research target to introduce a biorational approach, since accumulated knowledge of the biological system has already reached a satisfactory level for detail explanation of their behaviors. Besides structural modification of the known herbicides, an effective molecular design of potent PET inhibitors can be arrived at through consideration of the function of PQ at D 1 protein. New PET-inhibitory structures may also be discovered in natural products. Those compounds should be modified for the installation of herbicidal potency, which might require chemical parameters additional to those of the optimized structures at the binding site. Binding studies of the PET inhibitors are essential in this type of research since it is very difficult to understand their structure-activity correlation in cases when a minor structural modification causes a dramatic change in their attacking sites. It is also necessary to devise quick and reliable assay methods for herbicide screening in order to get accurate information about the structural effects and with this strategy the bio-assay using cultured cells has been proven to be a useful tool of structure-activity studies to detect the herbicidal requisites in the structures of PET inhibitors.

Three series of phenyl-substituted 2-cyanoacrylates were evaluated using simple quantitative structure activity relationships (QSAR) in an attempt to elucidate the nature of the regions of the binding site occupied by different parts of the molecules. Inhibition of the Hill reaction by substituted 3-phenylamino-2-cyanoacrylates correlated well with the lipophilicity of the substituent. The hydrophobic effect was also dominant when the Hill activity of a series of 3-benzylamino-2-cyanoacrylates was analyzed, although potency was considerably higher in the latter series. Lipophilicity and the electronic nature of the substituents were not major determinants in the Hill inhibitory activity of a series of substituted phcnoxycthyl 2-cyanoacrylic esters. In this case, a significant correlation was found with the molar rcfractivity (MR) of meta substituents, a parameter reflecting substituent size. The results indicate that the phenyl moiety of substituted 3-phenylamino- and 3-bcnzyl- amino-2-cyanoacrylates interacts with an essentially lipophilic binding domain, though it is likely that the two series are oriented differently with the 3-bcnzylamino series able to bind with greater affinity. In the phcnoxycthyl ester series, the substituted phenyl group interacts with a different environment, wherein ortho- and we7tf-substitution is tolerated, dependent on the bulk of the substituent, but /wra-substitution is detrimental to affinity for this region of the site.

Cyanoacrylate thiolate salts such as the sodium salt of ethoxyethyl-3-p-chlorobenzylthio-2-cyano-3-mercapto acrylate have been shown to be relatively slow binding, potent inhibitors of photosynthetic electron transport, that are equally active against thylakoids isolated from atrazine-susceptible and atrazine-resistant Brassica napus seedlings. It has been suggested that their mode of binding involves the substituted benzyl group interacting with the membrane lipids and the thiolate ion interacting in the region of the histidine 215 residue of the D 1 peptide, close to the non-heme Fe centre. This, together with their slow rate of binding, has led to the thiolate salts being classified as phenol type inhibitors.

The kinctics of inhibition of photosystem II electron transfer by different diuron-like herbicides (ureas, triazines, triazinoncs, biscarbamates. uraciles) were studied, mainly by chlorophyll fluorescence measurements. Uracil derivatives and cyanazine, a particular triazinc. were the slowest acting compounds. The half-times of action were strongly temperature-dependent and were of the order of tens of seconds at 5 °C for urea or triazine inhibitors. The role of different limiting steps in the binding process is discussed.

Thylakoids of genetically engineered species of Synechococcus PCC7942 were investigated by thermoluminescence (TL), polarographic and herbicide-binding studies. The results were as follows: 1) Deletion of either copy I or copy II/III of the psbA gene family resulted in modified oscillation patterns of flash induced TL and O 2 evolution compared to wild type cells. 2) Replacement of Ser 264 in psbA I by Ala or Gly inducing strong herbicide resistance leads to a downshift of the TL peak temperature of B-band by 10 C and alters the oscillation pattern of flash induced TL and O 2 evolution significantly to exhibit a pattern with comparatively high O 2 yield after the 2nd flash. Additional deletion of psbA II III does not lead to significant changes compared to the single mutated strains. 3) Contrary to Ser, M mutants, replacement of Phe 255 by Tyr in psbA I alone, which also induces herbicide resistance, is not accompanied by- corresponding changes in TL peak temperature and/or oscillation pattern of flash-induced TL and O 2 evolution. However, strongly altered properties are observed after additional deletion of psbA II/III. Based on these data it is inferred that expression of either psbA I or psbA II III genes only gives rise to PS II complexes slightly different in their function. Site-directed muta- genesis of psbA I gene modulates the expression ratio between copies I and II/III. The possible implications are discussed.

Two D2 mutants were created with a site-directed mutation near the presumable binding site of Q A . In one of the mutants, in which Trp-253, the aromatic residue potentially involved in facilitating electron transport from pheophytin to Q A and/or in binding of Q A, had been replaced by Leu, PS II was undetectable in thylakoids. This mutant is an obligate photoheterotroph. In another mutant the Gly-215 residue, located next to the His residue that is proposed to bind Q A and Fe 2+ , was mutated to Trp. This mutation leads to a rapid inactivation of oxygen evolution capacity in the light, and to a virtual elimination of the potential to grow photoautotrophically, but does not greatly affect the number of photosystem II reaction centers on a chlorophyll basis. We propose that proper binding of Q A to the photosystem II reaction center complex is a prerequisite for stability of the photosystem II complex. Impairment of Q a binding leads to rapid inactivation of photosystem II, which may be followed by a structural disintegration of the complex.

Isoelectrofocusing has been used to separate various chlorophyll-protein complexes of photosystem two (PS II). Light-harvesting complexes containing chlorophyll a and chlorophyll b (LHC II) were located in bands having p/s in the region of 4.5. At slightly higher pH other light-harvesting complexes containing little or no chlorophyll b were found. In the most basic region of the isoelectrofocusing gel, were located PS II core complexes characterized by con­taining the proteins of CP47, CP43, D 1, D 2 and α-subunit of cytochrome b 559 . The number of PS II core bands depended on the particular conditions employed for the separation procedure and in some cases were contaminated by CP 29. It is suggested that this heterogeneity resulting from different protonation states of the PS II. The least-acidic PS II core complex (pI 5.5) was found to bind the herbicides atrazine, diuron and dinoseb. In contrast, a PS II core complex with a p / of 4.9 bound only diuron. Its inability to bind atrazine was shown to be due to the low pH but no such explanation could be found for dinoseb. When atrazine-resistant mutant Senecio vulgaris was used, no binding of radioactive atra­ zine was observed with the PS II cores having a p i of 5.5. It is therefore suggested that the normal atrazine binding observed with PS II cores involves the high affinity site detected with intact membranes. With the PS II cores, however, this site has a reduced affinity probably due to structural modification in the D 1-polypeptide resulting from the isolation procedures.

The mechanism of atrazine binding and its modification by Chelex-100-induced Ca 2+ depletion and proteolytic degradation by trypsin, was analyzed in PS II membrane fragments from spinach. It was found: 1) Chelex-100 treatment leads in a comparatively slow process (t 1/2 = 5 - 10 min) to Ca 2+ re­ moval from a site that is characterized by a high affinity as reflected by K D values of the order of 10 -7 M . The number of these binding sites was found to be almost one per PS II in samples washed twice with Ca 2+ -free buffer. 2) Chelex-100 treatment does not affect the affinity of atrazine binding but increases the susceptibility to proteolytic attack by trypsin. 3) The electron transport activity is only slightly affected by Chelex-100 treatment. 4) The atrazine binding exhibits a rather small T-dependence within the physiological range of 7 °C to 27 °C. The implications of these findings for herbicide binding are discussed.

A hypothetical molecular model of part of the D 1 protein of photosystem II, based on the analogous portion of the L subunit of the Rhodopseudomonas viridis reaction centre, has been used to study the binding of an extended hydrophobic phenylurea inhibitor (N,N-dimethyl-carbamoyl)4 -amino-4 ′-chloro-trans-stilbene) (I) to the Q B site. The inhibitor was fitted by eye into a cleft in the site, and a limited part of the inhibitor/D 1 complex was energy minimized. The gross orientation of the inhibitor placed the dimethylurea moiety towards the predicted binding domain of the plastoquinone head group, and the stilbene moiety directed along the quinone isoprenoid side chain binding domain, suggesting a similar pathway of approach of the two molecules from the membrane into the binding site. Binding interactions of the inhibitor included hydrogen bonds to the side chain hydroxyl of ser 264 and the peptide carbonyl group of ala 251, with the side chain hydroxyl of ser 268 as an alternative ligand. Numerous hydrophobic contacts were also possible. Although phenylureas do not bind to reaction centres of Rp. viridis, many of the binding interactions to D1 could also be detected in Rp. viridis. However, the β-CH 2 and δ-CO 2 - groups of glu 212 in Rp. viridis are located in the corresponding region of D1 occupied by the dimethylurea moiety of the inhibitor in our model of its binding to D 1. This may explain why diuron (DCMU) does not bind to Rp. viridis reaction centres.

Proteolytic mapping of the D 1 subunit of photosystem two and a degradation product which arises during its rapid catabolism shows that the latter is a result of proteolysis within the peptide motif QEEET. This motif is located in a portion of the D1 protein thought to form a stroma-exposed connection between fourth and fifth transmembrane segments. This connection domain also contains a “PEST”-like sequence and forms part of the Q B /herbicide binding niche. The QEEET motif seems to provide a major epitope in immunological studies, as judged from reaction of D1 and its fragments with polyclonal antibodies. Antibodies against D1 were found to react with other animal and plant proteins which contain similar sequence motifs.

The effect of unoccupancy of the Q B site by plastoquinone on the photoinactivation of reaction center II in a Cyt b 6 /f-less mutant of Chlamydomonas reinhardtii, B 6 , was investigated. In these cells the oxidation of plastoquinol generated by electron flow via RC II to plastoquinone and thus the turnover of PQH 2 /PQ via the Q B site are drastically reduced. Reaction center II of the mutant cells was resistant to photoinactivation relative to the control cells as demonstrated by measurements of light-induced destabilization of S 2- Q B charge recombination, rise in in­ trinsic fluorescence and loss of variable fluorescence. These parameters relate to functions in­ volving the reaction center II D1 protein. The light-induced degradation of D1 in the mutant cells was also considerably reduced, with a t 1/2 value of 7 h as compared, under similar conditions, to about 1.5 h for the control cells. These results indicate that the photoinactivation of RC II and turnover of the D1 protein are related and require occupancy of the Q B site by PQ and its light-driven reduction.

Isogenic strains of Synechococcus PCC 7942 were genetically engineered so that copy I of the gene psbA was mutated at specific sites. These mutations resulted in replacements of Ser 264 by Gly or Ala and of Phe 255 by Tyr or Leu in the D1 protein. The mutants were resistant to herbicides inhibiting electron transfer in photosystem II. All mutants exhibited alterations in the stability of Q B - as demonstrated by a temperature downshift, to various extents, of the in vivo thermoluminescence emission. Measurements of the light-dependent turnover of D1 showed a marked decrease in the t 1/2 of this protein in the mutants as compared to wild-type, under low to medium light intensities. A correlation was found between the degree of pertur­ bation in the Q B - stability and the rate of acceleration in the turnover of D1. These data pro­ vide a direct evidence for the overlapping binding sites for the plastoquinone B and herbicides in the D1 protein. In addition these data indicate a close link between Q B - destabilization in reaction center II and the mechanism controlling the light-dependent turnover of D1. Based on these results and previous work we suggest that destabilization of the semireduced quinone, facilitates a light-induced damage in D1 which triggers its degradation.

Degradation of the 32 kDa-D1 protein, a photosystem II reaction centre component, was studied as a function of linear electron flow in visible light in the presence of various photosys­ tem II herbicides. Under these conditions, herbicide specific effects on protein degradation were clearly evident. 32 kDa-D 1 protein degradation and electron flow between Q a and Q b proved to be only partially correlated. We conclude that inhibition of protein degradation by PS II herbicides in visible light is not simply correlated to displacement of Q b .

Involvement of oxygen-free radicals in the rapid, light-dependent degradation of the 32 kDa photosystem II reaction center protein was investigated. The free radical scavengers propyl-gallate and uric acid inhibited 32 kDa protein degradation without affecting linear electron flow. The involvement of singlet oxygen was excluded. Protection from degradation was also afforded under ultra-violet and far-red radiations. These data implicate free-radical damage as a common step in the degradation process, and emphasize the oxygen environment as a causative factor in destabilization of the 32 kDa protein.

Single amino acid substitutions in the D1 protein of photosystem II may cause resistance to various herbicides. In all organisms studied these substitutions are located in or between helices IV and V of the protein. The increasing number of herbicide-resistant organisms necessitates development of a rapid methodology to characterize deviations from the wildtype se­ quence. Here, two procedures are described to identify mutations in the psbA gene, which is coding for D1. These procedures involve the isolation and amplification of DNA and R A and subsequent sequencing reactions without the need to clone the psbA gene. A triazine-resistant and a -susceptible biotype of Chenopodium album were used as model species. An A to G transition, giving rise to a serine to glycine mutation at position 264 in the D1 protein, is found in the resistant plant.

Oligonucleotide-directed mutagenesis techniques were used to delete the psbF gene, encoding the β subunit of the cytochrom e b 559 protein of the photosystem II complex in the cyano­ bacterium, Synechocystis 6803. Cyt b 559 is an integral com ponent of PS II complex. However, its precise functional role in PS II remains to be determined. Previously, we created a mutant in which the psbF gene as well as three of its neighbouring genes, psbE , psbL and p sb i were simultaneously deleted from the chrom osom e of Synechocystis 6803 (Pakrasi, Williams and Arntzen, EMBO J. 7, 325 -332 , 1988). This mutant had no PS II activity. However, the role of any one of the four individual gene products could not be determined by studying this mutant. The newly generated mutant, T 256, had only one gene, p sbF , deleted from the genome. This mutant was also impaired in its PS II activities. In addition, it had barely detectable levels of two other protein com ponents, D1 (herbicide binding protein) and D2, of the reaction center of PS II, in its thylakoid membranes. In contrast, two other proteins of PS II, CP47 and CP43 were present in appreciable amounts. Fluorescence spectra (77 K) of the mutant showed the absence of a peak at 695 nm that was previously believed to originate from CP47. In addition, phycobilisomes, the light-harvesting antenna system of PS II, were found to be assembled normally in this mutant. We conclude that the presence of the β subunit of Cyt b 559 in the thylakoid membranes is critically important for the assembly of PS II reaction center.

A new method of analysis of the factor contributing to solvation of small molecules is described. Approximations of the energetic contributions to the entry of a molecule into water have been derived from partition coefficient data of solutes and selected derivatives from a multiplicity of solvents. These include taking separate account energy of the cost of introduction of a molecular cavity in water, the strength of solute-water dipolar interactions, and the strengths of hydrogen bond formation with water of the lone pairs and hydroxylic hydrogens associated with the molecule. In this report the solvation-free energies of benzoquinone and hydroquinone in water are described. We also consider the solvation of the sem iquinone anion and show it to be fundamentally different from that of either the quinone or hydroquinone; this is discussed as a potentiality for selective binding (“solvation”) of quinone, sem iquinone and hydroquinone in sites of redox catalysis such as those found in the photosynthetic reaction center.

A new Synechocystis 6714 mutant, loxII A , resistant to the phenol-type herbicide ioxynil was isolated and characterized. The mutation found in the psbA gene (encoding the D1 photosystem II protein) is at the same codon 266 as for the first ioxynil-resistant mutant IoxI A previously selected [G. Ajlani. I. Meyer, C. Vernotte. and C. Astier, FEBS Lett. 246, 207-210 (1989)]. In IoxII A , the change of Asn 266 to Asp gives a 3 × resistance, whereas in IoxI A , the change of the same amino acid to Thr gives a 10 × resistance. The effect of these different amino acid substitutions on the ioxynil resistance phenotype has allowed us to construct molecular models and calculate the hydrogen-bonding energies between the hydroxyl group of ioxynil and the respective amino acids at position 266.

The decay of chlorophyll variable fluorescence after a “single turnover” flash is generally assumed to represent the reoxidation of the reduced quinone Q a . We have observed that the kinetics of this decay are very similar in a wide variety of species. Comparing 28 different species, we found an average half decay time of 314 ± 46 μsec. No systematic correlations were found between the decay rate and biochemical or physiological specializations such as C2, C4 or CAM. This indicates that structural as well as functional factors controlling photosystem II electron transfer between Q a and Q b are highly conserved. Apparently, the freedom for natural structural variations in this region is very limited. Triazine resistant plants, characterized by an altered amino acid sequence of the D1 protein, have clearly decreased rates of Q a/ Q b electron transfer. We found an average half decay time of 946 ± 100 (isec (5 species). However, this three-fold decrease is much less than previously reported. Therefore, if alterations of photosystem II electron transfer efficiency contributes to an often reported reduction of “ecological fitness”, this contribution is smaller than was hitherto assumed.

Mutants of Anacystis R2 with different amino acid exchanges in positions 255 and/or 264 in copy I of the psbA gene, leading to different tolerances to DCMU-type herbicides, are com- pared with the respective wild type concerning pigmentation and incorporation of 35 S into the D1 protein upon growth in the presence of [ 35 S]methionine. All mutants have shade-type appearance compared to the wild type, although to different extents depending on site and mode of the amino acid exchange in the D1 protein. Except for 3 mutants, there is no correlation between shade-type appearance on one hand and resistance towards a certain inhibitor on the other hand. Not only the molar ratio of phycocyanin (PC) to chlorophyll (Chi) is higher in all mutants compared to the respective wild type, but also the rate of synthesis of the D1 protein. On the background of different levels of total 35 S incorporation within 18 min, D1 synthesis can be related to shade adaptation. Degradation of the D1 protein remains to be thoroughly studied in this context. No reproducible differences in whole chain electron transport were observed between mutants and wild type.

Fluorescence induction characteristics have been studied in wild-type and atrazine-resistant mutant of Rliodobacter capsulatus. Fluorescence induction was found to be a useful technique to monitor the altered electron transfer in the atrazine-resistant mutants as well as in the different membrane fractions of wild-type R. capsulatus. In both cases, the proportion of the fast rise of variable fluorescence was increased indicating the enhancement of Q A . I`n the mutant strain, the I 50 value of triazine herbicide terbutryn was increased by 100-fold whereas the natural resistance of R. capsulatus against diuron was abolished by the mutation.

A first model of the three-dimensional structure of the photosynthetic reaction center of the mutant T1 (SerL 223 → Ala, ArgL 217 → His) from Rhodopseudomonas viridis, resistant toward the triazine herbicide terbutryn (2-methylthio-4-ethylamino-6-f-butylamino-5-triazine), has been developed from X-ray data measured to a resolution of 2.5 Å. The secondary quinone, Q B , which in T1 binds better than in the wild type, is present in the crystals. Both substituted residues are clearly visible in the difference fourier map. The replacement of these two residues in the Q B site causes only minor changes in the overall structure of the protein.

Four atrazine-resistant mutants from the purple bacterium Rhodopseudom onas viridis were isolated. Sequence analysis revealed three different mutant strains carrying mutations in the herbicide-binding pocket: i) M AV 2: L 212-Glu → Lys, ii) M AV 3: L216-Phe → Ser and iii) MAV 4 = MAV 5: L217-Arg → His, L220-Val → Leu. Except M AV 3 all Rps. viridis mutants are different from those selected by their resistance towards the closely related triazine terbutryn.

A few species have evolved resistance to paraquat after repeated selection. As paraquat still inhibited NADP reduction, we hypothesized that resistance might be due to (a) detoxification of the paraquat-generated active oxygen species and (b) that resistant plants would have some cross resistance to other xenobiotic oxidants as well as to photoinhibition, which we subsequently demonstrated. The levels of plastid isozymes of the oxygen detoxification pathway: (CuZn) superoxide dismutase, ascorbate peroxidase and glutathione reductase were genetically higher in the resistant than in the sensitive biotype of Conyza bonariensis through the F 2 generation. Resistance was suppressed by chelators of copper and/or zinc. Intact chloroplasts from resistant plants had less membrane damage with and without paraquat, than those from sensitive plants. Resistant Conyza plants recover from paraquat inhibition of photosynthesis in 3-4 h in high light, whereas sensitive plants died. Both resistant and sensitive plants recovered from paraquat in 3-4 h in low light intensities. Paraquat-resistant Conyza plants were cross-tolerant to SO 2 , atrazine, acifluorfen and to photoinhibition. Drought-tolerant maize inbreds were cross-tolerant to paraquat, SO 2 and acifluorfen (compared to sensitive lines) and they also possessed higher levels of (Cu/Zn) superoxide dismutase and glutathione reductase. The tolerance to oxidant stresses in Conyza and maize increases with plant age, suggesting that the shift to resistance is a constitutive, earlier expression of the genes normally expressed later in development.

Many photosystem II inhibiting herbicides still inhibit this process in triazine-resistant plants; i.e. they have no cross resistance with atrazine. Five- to twenty-fold lower concentrations of phenolic type herbicidcs and 5-fold less of the active ingredient of pyridate and half as much ioxynil are required to inhibit thylakoid PS II in atrazine-resistant biotypes than in sensitive biotypes; i.e., they even show “negative cross resistance”. Negative cross resistance may be the major reason that atrazine resistance did not evolve where herbicide mixtures were used, when the mixed herbicide (usually a non-PS II inhibiting acetanilide) also controlled triazine-sensitivc weeds. Mathematical modeling in principle allows quantification of the very low field levels of herbicides possessing negative cross resistance that could be mixed with atrazine that would stop or delay the evolution of resistant populations without affecting the maize crop. There are few available actual dose response curves of atrazine-resistant vs. susceptible weeds at the whole plant level for herbicidcs exerting negative cross resistance. Thus, “real situation” modeling cannot be done. Data acquisition is called for so that the model can be extrapolated from the thylakoid to the field.

Studies were conducted to test the hypothesis that the mutation to the psbA. plastid gene that confers s-triazine resistance (R) also results in an altered diurnal pattern of photosynthetic carbon reduction (CER) relative to that of the susceptible (S) wild-type. In all experiments CER approximated the increasing and decreasing light levels during the diurnal. S C ER exceeded that of R during the midday period, but R CER was greater early and late in the diurnal at 25 °C. R CER exceeded that of S for most of the diurnal at 35 °C and in older, crowded, nitrogen-starved plants. These studies support the stated hypothesis and indicate a more complex model of photosynthetic productivity than previously observed. An assessment of the photosynthetic competence of either biotype may be a function of the time of day or the environmental conditions the plants are exposed to, especially temperature.

Triazine-resistant chloroplasts of the Canadian spring oilseed rape variety OAC Triton were transferred into four German winter oilseed rape lines and two cultivars of double-low quality by means of protoplast fusion. X-irradiation has been used to reduce the amount of nuclear D N A of the spring type cultivar and to promote cybrid formation. RFLP-analysis showed that some regenerants and their progeny carried both types of chloroplasts. In some instances regenerants and progeny containing mixtures of both chloroplasts not kept under selective conditions lost their triazine-resistant chloroplasts completely during further plant growth. Preliminary results of greenhouse and field experiments indicate that volunteer plants can be eliminated by application of 150-300 g/ha metribuzin (Sencor R , Bayer AG) in a stand of triazine-resistant oilseed rape of double-low quality.

The herbicide norflurazon inhibits carotene biosynthesis in photosynthetic organisms by blocking the enzyme phytoene dehydrogenase (= phytoene desaturase). We have isolated norflurazon-resistant mutants of the cyanobacterium Synechococcus PCC7942. The herbicide- resistance gene from the mutant NFZ4 has been cloned by genetic complementation of the resistance trait in wild type cells. The experiment described here illustrates the usefulness of employing cyanobacteria to clone herbicide-resistance genes in a quick and simple way.

Colorless carotenoids were accumulated, analyzed and quantified in heterotrophically- grown Scenedesmus cultures treated with the bleaching herbicide norflurazon. By optical, IR and mass spectroscopy 15-cis as well as all-trans phytofluene, 15-cis phytoenes and traces of its all-trans isomer were identified. Furthermore, a phytoene epoxyde and several hydroxyphy- toenes were assigned. Comparing the concentrations of these phytoene derivatives in hetero-trophic cultures grown in darkness with those in light-grown autotrophic cells, an exchange of phytoene derivatives was observed. Levels of phytoene were much higher in the dark whereas concentrations of hydroxyphytoenes dominated in the light. Other herbicides like fluridone, diflufenican, and difunon accumulate the same colorless carotenoids including phytoene epoxyde and various hydroxyphytoenes. For all herbicides an almost constant hydroxyphytoene to phytoene epoxyde ratio was observed indicating an interrelated formation of both oxygenated phytoene derivatives which can be influenced by light conditions of the cultures.

The target sites of three herbicides which inhibit carotenoid biosynthesis have been characterized using HPLC analysis of pigment extracts from two higher plant systems, carrot cell suspension cultures and barley seedlings. Diflufenican causes an accumulation of phytoene and phytofluene. Dichlormate causes accumulation of phytoene, phytofluene, ξ-carotene, neurosporene and β-zeacarotene. Amitrole causes accumulation of phytoene, phytofluene, β- γ- and δ-carotenes and lycopene. Significant differences in the geometric and hydroxylated natures of the accumulated precursors occurred between the carrot cell and dark- and light-grown barley. These differences are discussed with respect to both the target sites of the three carotenogenic herbicides and the biosynthetic pathway leading to carotenoid biosynthesis in higher plants.

The reactivity of the bleaching herbicide clomazone with reduced glutathione (GSH) was studied under in vitro conditions. Clomazone at 1.3 mM and GSH at 100 mM were reacted in a phosphate buffer (pH 8.8) at 25 °C for 3 days. The resulting synthetic product was identified as a GS-conjugate of clomazone by means of TLC, HPLC, and mass spectrometry (chemical ionization) procedures. Exogenous application of 0.5 mM of GSH and cysteine reduced significantly the bleaching effect of clomazone on leaf disks of velvetleaf (Abutilon theophrasti Medik.). Mercaptoethanol was not effective in antagonizing the bleaching effect of clomazone on velvetleaf leaf disks. These results show that clomazone can react with selected thiol compounds under in vitro conditions. The conjugation of clomazone with GSH may be involved in the metabolic detoxication of this herbicide in plants.

Several hypotheses have been proposed to explain the light-dependent phytotoxicity of diphenyl ethers and related herbicides: inhibition of photosynthesis; activation of herbicides by light, by the photosynthetic electron transfer chain or by excited forms of carotenoids; and interaction with the biosynthesis of tetrapyrrole pigments. It is shown that the most likely mode of action consists in inhibiting the enzyme protoporphyrinogen oxidase. As a consequence, protoporphyrinogen is oxidized non-enzymatically to protoporphyrin IX. The latter molecule is a powerful photosensitizer, able to generate singlet oxygen in the light and thus to induce peroxidative destruction of membrane lipids.

Peroxidation is indicated by several physiological parameters: (1) leakage of cell membranes and production of short-chain hydrocarbons, (2) degradation of cell constituents, (3) inhibition of chlorophyll biosynthesis, (4) accumulation of tetrapyrroles (protoporphyrin IX). (5) inhibition of protoporphyrinogen oxidase by peroxidizers, (6) alleviation of peroxidation by inhibitors of chlorophyll biosynthesis. (7) alleviation of peroxidation by photosynthesis inhibitors (“diuron effect”), (8) counteraction of the diuron effect by glucose (heterotrophic conditions). Such parameters, described and elaborated with microalgae and higher plants, arc verified in this study with soybean cells, duckweed and mung beans. Preliminary data indicate lack of quantitative relationship between inhibition of protoporphyrinogen oxidase, protoporphyrin (IX) formation and peroxidative consequences.

Thiolactomycin was shown to be a potent inhibitor of de novo fatty acid biosynthesis in intact isolated chloroplasts (measured as [ 14 C]acetate incorporation into total fatty acids). In our attempt to further localize the inhibition site we confirmed the inhibition with a fatty acid synthetase preparation, measuring the incorporation of [ 14 C]malonyl-CoA into total fatty acids. From the two proposed enzymic targets of the fatty acid synthetase by thiolactomycin we could exclude the acetyl-CoA: ACP transacetylase. It appears that the inhibition by thiolactomycin occurs on the level of the condensing enzymes, i.e. the 3-oxoacyl-ACP synthases. We also demonstrated that the two starting enzymes of de novo fatty acid biosynthesis, the acetyl-CoA synthetase and the acetyl-CoA carboxylase, are not affected by thiolactomycin.

The mode of action of cyclohexane-l,3-dione-type (cycloxydim, clethodim, sethoxydim, tralkoxydim) and aryloxyphenoxypropanoate-type herbicides (diclofop, fenoxaprop, haloxyfop, fluazifop) is summarized in this review. Both herbicide classes, though structurally completely different, specifically block the same target enzyme i.e. the plastid acetyl-CoA carboxylase (ACC) (EC 6.4.1.2). Most members of the Poaceae are sensitive towards both herbicide groups, whereas other monocotyledonous plants as well as the dicotyledonous plants appear to be resistant. This resistance, which can be found on the level of whole plants, in isolated chloroplasts and also on the level of ACC-enzyme preparations, is apparently due to a modification of the target enzyme ACC. Within the sensitive grass family some members (Festuca and Poa species) are partially tolerant against both graminicide groups. In the case of cyclohexanedione herbicides the tolerance seems to be due to a reduced sensitivity of the target enzyme. In the case of aryloxyphenoxypropionic acid herbicides the tolerance is apparently based on a combined action of cytoplasmic factors (metabolization?) and a slightly reduced sensitivity of the target enzyme. From differences in the sensitivity of certain grasses against the two herbicide classes it is concluded that both graminicide groups bind to the same binding domaine of the ACC enzyme but possess different subsites. The consequences of the block of de novo fatty acid biosynthesis in the plastids of sensitive plants is the lack of glycerolipid and biomembrane formation which finally causes cell death in the meristematic tissues.

Glyphosate (N[phosphonomethyl]glycine) applied to an assimilate-exporting leaf of tomato or spinach plants is translocated via the phloem to the young growing areas of the shoot apex and root where it causes the rapid accumulation of shikimate and shikimate 3-phosphate to up to 16% of the dry weight or these tissues. Using the technique of non-aqueous fractionation of spinach leaves it was shown that the herbicidal target of glyphosate, the shikimate pathway enzyme 5-enolpyruvlyshikimate 3-phosphate synthase, is located within the chloroplast and that, in the presence of glyphosate. shikimate 3-phosphate accumulates in the same organelle, while shikimate was preferentially localized in the vacuolc. The major part of [ l4 C]glyphosatc imported into the leaf via the phloem was detected in the cytosolic fraction of the mesophyll cells.

Phosphinothricin (PPT) causes a rapid inhibition of photosynthesis under atmospheric conditions (400 ppm CO 2 , 21% O 2 ). However, under conditions (1000 ppm CO 2 , 2% O 2 ) under which photorespiration cannot occur, there is no or only a very low rate of photosynthesis inhibition by phosphinothricin. Under both conditions, a strong NH 4 + -accumulation is apparent caused through the inhibition of glutamine synthetase by phosphinothricin. This indicates, that NH 4 + -accumulation cannot be the primary cause for photosynthesis inhibition by phosphinothricin, but a process in connexion with photorespiration plays a central role. Through the lack of amino donors, the transamination of glyoxylate to glycine in photorespiration cannot take place. PPT causes a great decrease in glutamine, glutamate, aspartate, serine, and glycine. Following addition of these amino acids to PPT, there is a decrease in photosynthesis inhibition by PPT. With the addition of glutamine or glutamate to PPT no decrease in serine and glycine is detected, because the transamination of glyoxylate to glycine in photorespiration can occur.

We present here the isolation and molecular characterization of acetolactate synthase (ALS) genes from the cyanobacterium Synechococcus PCC7942 which specify a sulfonylurea-sensitive enzyme and from the sulfonylurea-resistant mutant SM3/20, which specify resistance to sulfonylurea herbicides. The ALS gene was cloned and mapped by complementation of an Escherichia coli ilv auxotroph that requires branched-chain amino acids for growth and lacks ALS activity. The cyanobacterial gene is efficiently expressed in this heterologous host. The ALS gene codes for 612 amino acids and shows high sequence homology (46%) at the amino acid level with ALS III of E. coli and with the tobacco ALS. The resistant phenotype is a consequence of proline to serine substitution in residue 115 of the deduced amino acid sequence. Functional expression of the mutant gene in wild-type Synechococcus and in E. coli confirmed that this amino-acid substitution is responsible for the resistance. Yet the deduced amino-acid sequence as compared with othjer ALS proteins supports the notion that the amino-acid context of the substitution is important for the resistance.

Several major classes of herbicides, discovered by conventional screening techniques, have been found to inhibit the first common enzyme of branched-chain amino acid biosynthesis, acetolactate synthase, as their mode of action. These herbicides seem to bind to an evolutionary vestige of a quinone-binding site, extraneous to the active site, that is present due to the evolutionary history of this enzyme. Besides their herbicidal effect on sensitive plants, these compounds can effect stasis in the growth of bacteria and yeast. Recently is has been reported that an experimental herbicide from Hoechst. Hoe 704. that was discovered by conventional screening techniques, inhibits the second common enzyme of branched-chain amino acid bio- synthesis [Schultz etal., FEBS Lett. 238, 375-378 (1988)]. We have also recently designed novel reaction-intermediate analogs (e.g. N-isopropyl oxalylhydroxamate) that arc exceptionally potent (K i = 22 p M : half-time for release approximately six days) and selective inhibitors of the second common enzyme, ketol-acid reductoisomerase. Both of these selective inhibitors of the second common enzyme will kill sensitive plants, but will only inhibit the growth (without killing) of bacteria. The effects in bacteria parallel those obtained by mutations in the relevant genes, where loss of either the first or second common enzyme in the pathway gives an organ- ism that is auxotrophic for branched-chain amino acids, but does not result in a conditionally lethal phenotype. Higher plant mutants have only been obtained to date that arc deficient in functional leucine-specific gene products (as yet uncharacterized), threonine deaminase (isoleucine specific), and dihydroxyacid dehydratase (common). The phenotypes of these mutants. at least at the level of cell culture, are similar to those of their bacterial counterparts, in that auxotrophy, but not conditional lethality, is obtained. These results highlight the potential non-equality of the enzymes of branched-chain amino acid biosynthesis as targets in herbicide design.

At least two different classes of ADP binding sites on chloroplast coupling factor are described in the literature. High-affinity sites are assumed to entail regulatory functions while low-affinity sites are involved in catalysis. Diphenyl ether herbicides, acting as energy transfer inhibitors, interfere with nucleotide exchange on both categories of ADP binding sites. Their inhibitory potency varies depending on their substitution. We found that each diphenyl ether assayed revealed identical I 50 values for inhibition of nucleotide interaction with both classes of binding sites. We show here that diphenyl ether inhibition of energytransfer is primary based upon the interference with ADP binding to high-affinity binding sites. Thereby the control of proton permeability through the coupling factor complex is affected. Moreover, we found that the three β-subunits are not absolutely fixed in one conformational state: After covalently blocking the high-affinity site by an azido-label. a new high affinity state appeared on another β-subunit.

A microsomal fraction isolated from the shoots of 3- to 4-day-old, dark-grown, grain sorghum (Sorghum bicolor cv. Funk G 522 D R ) seedlings was characterized. The preparations had a cytochrome P-450 content that varied from approximately 90 to 150 pmol P-450/mg protein with cytochrome P-420 varying from 0 to 3% of the P-450 content. Type I difference spectra were formed with cinnamic acid and metolachlor, and a type II spectrum was formed with tetcyclacis. In short-term assays with [ 14 C]metolachlor as substrate, the preparations produced a single time-dependent product that separated on silica gel TLC plates developed in benzene/acetone (2:1, v/v). R F values for metolachlor and the metabolite were approximately 0.70 and 0.48, respectively. The microsomal reaction required N A D P H and oxygen, and was inhibited by carbon monoxide, with the inhibition being partially reversed by actinic light. Compounds known to inhibit the activity of cytochrome P-450 monooxygenases (piperonyl butoxide, tetcyclacis, and tridiphane) also prevented formation of the metabolite. Identity of the metabolite was confirmed by TLC and positive ion thermospray LC/MS to be 2-chloro-N-(2-ethyl-6- methylphenyl)-N-(2-hydroxy-l-methylethyl)acetamide. Hence, the reaction catalyzed by the sorghum microsomes involved O-demethylation of the methoxypropyl side chain of metolachlor.

Dichlormid, a safener for thiolcarbamate herbicides, was tank-mixed with several herbicidal inhibitors of photosystem II, or with the herbicide acifluorfen, and applied postemergence to Ipomoea hederacea plants. Dichlormid had no visible effects on the plants when applied alone, but interacted synergistically with the herbicides in the combination treatments. Dichlormid strongly decreased the ascorbic acid levels in the Ipomoea hederacea cotyledons. Ascorbate is known to protect plant tissue from photooxidative damage. The herbicides which interacted synergistically with dichlormid are believed to generate their phytotoxic action via the production of excess singlet oxygen. It is suggested that the decreased ascorbate levels in the Ipomoea hederacea cotyledons after dichlormid treatment result in an impaired singlet oxygen scavenging system and consequently lead to increased plant damage in the presence of singlet oxygen generating herbicides.