Sulphur supply and infection with Pyrenopeziza brassicae influence L-cysteine desulphydrase activity in Brassica napus L.

Abstract

Different field surveys have shown that sulphur (S) fertilization can increase the resistance of agricultural crops against fungal pathogens. The mechanisms of this sulphur-induced resistance (SIR) are, however, not yet known. Volatile S compounds are thought to play an important role because H₂S is toxic to fungi. A field experiment was conducted to analyse the influence of S fertilization and the activity of H₂S-releasing enzymes on fungal infections. Two levels of N and S fertilizers and two varieties of oilseed rape were investigated with respect to their potential to release H₂S by the enzymatic activity of L-cysteine desulphydrase (LCD) and O-acetyl-L-serine(thiol)lyase (OAS-TL). LCD releases H₂S during cysteine degradation, while OAS-TL consumes H₂S during cysteine synthesis and free H₂S is only released in a side reaction. All plots of the field trial showed an infection with Pyrenopeziza brassicae and leaf disc samples were taken from visibly infected leaf areas and apparently uninfected areas to investigate the reaction to the infection in relation to the treatments. Different S fractions and the activities of LCD and OAS-TL were measured to evaluate the potential to release H₂S in relation to S nutrition and fungal infection. S fertilization significantly increased the contents of total S, sulphate, organic S, cysteine, and glutathione in the plants, but decreased LCD activity. Infection with P. brassicae increased cysteine and glutathione contents, as well as the activity of LCD. Therefore crops were able to react to a fungal infection with a greater potential to release H₂S, which is reflected by an increasing LCD activity with fungal infection.

Brassica napus L., H₂S, L-cysteine desulphydrase, O-acetyl-L-serine(thiol)lyase, Pyrenopeziza brassicae, sulphur-induced resistance

Introduction

The role of sulphur (S) in the resistance of crops against diseases became obvious at the end of the 1980s when atmospheric S depositions were so much reduced by clean air acts that S deficiency became a widespread nutrient disorder in European agriculture (Booth et al., 1991; Kjellquist and Gruvaeus, 1995; Knudsen and Pedersen, 1993; Richards, 1990) and the infection of crops with certain diseases became increasingly obvious (Paul, 1992; Schnug and Ceynowa, 1990; Schnug et al., 1995a).

It has been long known that foliar applied elemental S has a fungicidal impact but only recently could it be shown that soil-applied S in the form of sulphate also had a significant effect on the health status of crops. A significant repressive effect of soil-applied S on the infection of oilseed rape with Pyrenopeziza brassicae, grapes with Uncinula necator, and potato tubers with Rhizoctonia solani was found (Bourbos et al., 2000; Klikocka et al., 2004; Schnug et al., 1995a). The results of these experiments indicate that S metabolites are involved in disease resistance and support the concept of sulphur-induced resistance (SIR) (Schnug et al., 1995a).

The S metabolism of plants offers several possibilities to combat fungal attacks and different metabolites were investigated with respect to their role in SIR. For instance elemental S depositions in the vascular tissue of resistant cocoa (Theobroma cacao) in response to infection with Verticillium dahliae were attributed to the toxicity of elemental S (Cooper et al., 1996; Resende et al., 1996; Williams et al., 2002). Other mechanisms to combat biotic stress, which are provided by S metabolism, involve glutathione (GSH), phytoalexins, glucosinolates, and the release of S-containing volatiles (Haneklaus et al., 2004).

H₂S is cytotoxic and therefore a relationship between increasing H₂S emissions and the resistance of crops against pest and diseases is possible (Beauchamp et al., 1984; Schroeder, 1993; Sekiya et al., 1982a). Atmospheric H₂S concentrations that are higher than 46 μg m⁻³ (De Kok, 1990) are reported to be phytotoxic, but there are no data available about fungitoxic H₂S concentrations. Haneklaus et al. (2004) calculated that a minimum uptake of 10 μM H₂S h⁻¹ by the pathogen would be necessary to yield a fungicidal effect.

The release of several volatile reduced-S compounds (hydrogen sulphide, carbonyl sulphide, dimethyl sulphide, carbon disulphide, and methylmercaptan) from various plant species have been identified (Schroeder, 1993). Such measurements were mostly conducted with cut plant parts that were fed with concentrated S-solutions or with living plants under experimental conditions. No field data exist where the emission of gaseous reduced-S was measured from living plants under different nutritional conditions or in relation to a fungal infection. The mechanisms by which H₂S is released, the extent of the H₂S emission under natural conditions, and the relation to fungal diseases are not fully understood. Also the influence of the plant S supply on the amount of H₂S released by crops is not yet known, but in the case of other secondary S compounds, such as GSH and glucosinolates, significant positive relationships were found (Haneklaus et al., 1999; Schnug et al., 1995b). The emission of H₂S is a light-dependent process and can be induced by feeding plants with an excess of S in the form of sulphur dioxide (Wilson et al., 1978), sulphate (Sekiya et al., 1982b; Winner et al., 1981) or L-cysteine (Rennenberg, 1989; Sekiya et al., 1982a). This indicates that the release of H₂S also depends on the S nutritional status of the crop.

H₂S may be released prior to or after cysteine formation (Giovanelli, 1990), but the question is still open as to which enzymes could be responsible for H₂S release. Two candidates are L-cysteine desulphydrase (LCD) and O-acetyl-L-serine(thiol)lyase (OAS-TL). LCD catalyses the decomposition of cysteine to pyruvate, ammonia, and H₂S. OAS-TL is responsible for the incorporation of inorganic S into cysteine, which can be subsequently converted into other S-containing compounds. H₂S is evolved in a side reaction (Tai and Cook, 2000) and in a molar ratio the enzyme formed about 25 times more cysteine than H₂S mg⁻¹ protein during the same incubation time (Burandt et al., 2001). Therefore in vitro the reaction of OAS-TL is a net H₂S-consuming reaction. The studies of Burandt et al. (2001) gave the first indications of a relationship between the activity of potentially H₂S-releasing enzymes, the S status of the crop, and an infection with fungal diseases. An increasing total S content in different genotypes of oilseed rape was associated with a decreasing LCD and an increasing OAS-TL activity and lower infection rates with Verticillium dahliae.

According to Giovanelli et al. (1980) and Schuetz et al. (1991) the OAS-TL activity was insensitive to changes in the S supply and the LCD activity was not coupled to the emission of H₂S. Presumably, the evolution of H₂S in the presence of high concentrations of sulphate or sulphite results from a transitory state in which the incorporation of sulphide into cysteine does not keep pace with light-coupled sulphate/sulphite reduction (Anderson, 1990), as the addition of the physiological sulphide acceptor OAS inhibits the evolution of H₂S and enhances the production of cysteine (Filner et al., 1984; Rennenberg, 1983). On the other hand, compounds which inhibit the incorporation of cysteine into GSH, promote the evolution of H₂S, suggesting that when cysteine-consuming processes are inhibited and the concentration of cysteine increases, sulphide is emitted as H₂S (Rennenberg and Filner, 1982). Therefore, it is still controversial if the enzymes LCD and OAS-TL, respectively, are responsible for the release of H₂S and if there is a relationship between the enzyme activities and the S nutritional status of the crops or fungal infections.

In this experiment the influence of S and N nutrition on different S fractions and enzyme activities were investigated, because N and S show strong interactions in their nutritional effects on crop growth and quality due to their mutual occurrence in amino acids and proteins. The enzyme OAS-TL links the S and the N assimilatory pathways as the precursor, OAS, is derived from the C and N assimilation pathways and the reaction product cysteine, may be regarded as the primary organic compound containing reduced S (Warrilow and Hawkesford, 1998).

It was the aim of this work to investigate the relationship between the S and N nutritional status of the crops, the activity of the enzymes LCD and OAS-TL, to analyse the relationship between a fungal infection, and to assess the activity of these enzymes in relation to the nutritional status of the crops.

Materials and methods

Experimental design of the field experiment

A multi-factorial field experiment was conducted in 2002 in Braunschweig (52°18′ N, 10°27′ E in Lower Saxony, Germany) on a loamy sand (dystric Cambisol/orthic Luvisol). The following factors were investigated in a completely randomized block design: (i) two varieties of winter oilseed rape (Brassica napus L.), one of which was susceptible to the fungus P. brassicae (Bristol) and a resistant variety (Lipton); (ii) two levels of S fertilization (0 and 150 kg ha⁻¹ S); and (iii) two levels of N fertilization (100 and 200 kg ha⁻¹ N).

Each treatment had four replicates and the plot size was 60 m². For defining the growth stages (GS) of oilseed rape, the BBCH scale was used (Strauss et al., 1994). The S fertilization was applied as potassium sulphate (K₂SO₄) with potassium balance by KCl and split into three doses: two were applied in the autumn (40 kg S ha⁻¹ before sowing at GS 01 and 40 kg S ha⁻¹ at GS 12), and the third dose was applied in the spring at the start of vegetation together with the first N application (GS 19). N was applied as ammonium nitrate (NH₄NO₃) and the fertilization was split into two equal doses. The second N dose was applied at the start of main growth (GS 50–53). The experimental plots were infected with P. brassicae by evenly distributing infected straw material on the plots in the autumn (GS 11).

The leaf sampling was carried out at the start of main growth (GS 50–53) at the end of April 2002. Twenty to thirty fully developed young leaves from the upper third of the crops, and leaf discs from 15–20 fully developed young leaves of a diameter of 16 mm were taken and immediately frozen in liquid nitrogen and subsequently freeze-dried.

All plots showed the early symptoms of infection with P. brassicae when the sampling was carried out: little white spots of conidia were visible on the leaves. Therefore, two kinds of leaf disc samples (15–20) from each plot were taken, one with visible symptoms of infection (white spots on the leaf material) and one without symptoms to investigate the influence of the fungal infection on S metabolites and enzyme activities.

The infection with P. brassicae takes place in late autumn and the conidia are visible in early spring as little white spots on the leaf surface and bottom; later on they build necrotic spots of up to 2 cm diameter and the leaves show deformations. Typical symptoms are also necrotic areas at the stems and bursting of the stems.

Chemical analysis

Young, fully differentiated, leaves of oilseed rape were taken at the beginning of stem elongation. For mineral analysis, the samples were dried at 60 °C in a ventilated oven until constant weight and finely ground (<0.12 mm) using a RETSCH ultra-centrifugal mill. For the determination of the total S content by X-ray fluorescence spectroscopy powdered material was prepared, mixing 1.1 g of plant material with 4.4 g of HOECHST wax C (Schnug and Haneklaus, 1999). Total N was determined by employing the Kjeldahl method.

Sulphate was determined in freeze-dried plant material according to Novozamsky et al. (1986) and organic S was calculated as the difference between total S and sulphate-S in the plant material.

In leaf disc samples, the GSH and cysteine content was determined as well as the activity of the LCD and OAS-TL. There was not enough leaf disc material to determine total N and S from this material, but differences in the mineral content were not likely.

Measurement of cysteine and glutathione

Twenty to thirty mg of fine-ground, freeze-dried plant material was diluted with 0.1 M HCl containing 4% polyvidone-25 (Hell and Bergmann, 1990). The samples were centrifuged twice for 5 min each at 14 000 g at 4 °C. Aliquots of the supernatants were neutralized with 0.08 M NaOH, reduced with 10 mM dithiothreitol, and the sulphydryl groups were derivatized with 10 mM bromobimane (Sigma-Taufkirchen, Germany) as described by Hell and Bergmann (1990). Separation, detection, and quantification of fluorescent adducts was achieved by a reversed phase column (Waters Nova-Pak C18, 4.6×250 mm) and a Hitachi HPLC System running with a gradient of 100% methanol and 0.1 M potassium acetate buffer as eluents.

Enzyme activity measurements

The activities of LCD and OAS-TL were determined in the frozen leaf disc samples as follows: the frozen plant material was ground in liquid nitrogen and the soluble proteins were extracted by adding 1 ml 20 mM TRIS/HCl, pH 8.0, to 100 mg plant material. After centrifugation the protein content of the supernatant was determined according to Bradford (1976) using bovine serum albumin as a standard.

The OAS-TL assay contained a total volume of 1 ml: 5 mM OAS, 5 mM Na₂S, 3.33 mM dithiothreitol, 100 mM TRIS/HCl, pH 7.5, and 50 μl enzyme extract (Schmidt, 1990). The solutions of OAS, Na₂S, and dithiothreitol were prepared at the start of the experiment. The reaction was initiated by the addition of Na₂S and the sample was incubated for 30 min at 37 °C before the reaction was stopped by adding 1 ml acidic ninhydrin reagent (0.8% ninhydrin (w/v) in 1:4 concentrated HCl:HOAc) in order to determine the cysteine concentration (Gaitonde, 1967). The samples were heated at 100 °C for 10 min and, finally, 2 ml EtOH was added to stabilize the colour complex. The absorption of the samples was measured at 560 nm in a micro plate reader (Fluostar Optima, BMG Labtechnologies, Offenburg). Solutions with different concentrations of L-cysteine were prepared, treated in the same way as the assay samples, and were used for the quantification of the enzymatically formed cysteine. The linearity of the product formation with respect to incubation time, and the amount of protein given to the single assay, was carefully tested.

The LCD activity was measured by the release of sulphide from cysteine in a total volume of 1 ml consisting of 2.5 mM dithiothreitol, 0.8 mM L-cysteine, 100 mM TRIS/HCl, pH 9.0, and enzyme extract. The reaction was initiated by the addition of L-cysteine. After incubation for 15 min at 37 °C the reaction was terminated by adding 100 μl of 30 mM FeCl₃ dissolved in 1.2 N HCl and 100 μl 20 mM N,N-dimethyl-p-phenylenediamine dihydrochloride dissolved in 7.2 N HCl (Siegel, 1965). The formation of methylene blue was determined in microtitre plates at 670 nm. Solutions with different concentrations of Na₂S were prepared, treated in the same way as the assay samples and were used for the quantification of enzymatically formed H₂S. For each treatment four samples were collected and in each sample the assays were repeated three times on separate days.

Statistical calculations

ANOVA was used to determine which means were significantly different from others at the 5% significance level employing the CoHort software package (Simons, 1995). The analysis of variance was conducted with three factors (S and N fertilization, and cultivars). The influence of infection with P. brassicae on the investigated parameters was calculated by a one-factorial ANOVA.

Results

In Table 1 the influence of N and S fertilization, cultivars, and infection with P. brassicae on N and S contents, on different S fractions, and enzyme activities is summarized. Both the S and N fertilization were effective as the mineral contents of both elements increased significantly with fertilization. This is a prerequisite to investigate nutritional effects on other components. The two cultivars Bristol and Lipton showed no differences with respect to the investigated components and they also showed no differences with respect to their susceptibility to the fungus P. brassicae.

Table 1.

Influence of N and S fertilization, and cultivars on L-cysteine desulphydrase and O-acetyl-L-serine(thiol)lyase activity and cysteine and glutathione content in leaf discs and total N and S, sulphate, and organic S contents in young, fully differentiated leaves of oilseed rape at the start of stem elongation (three-factorial ANOVA) and influence of infection with P. brassicae on these parameters (1 factorial ANOVA)

Factor	N (mg g−1 DW)	S (mg g−1 DW)	Sulphate (mg g−1 DW)	Organic S (mg g−1 DW)	Cysteine (μmol g−1 DW)	Glutathione (μmol g−1 DW)	L-cysteine desulphydrase (nmol H2S mg−1 protein min−1)	O-acetyl-L-serine(thiol)lyase (nmol cys mg−1 protein min −1)
S fertilization
0	46.2 a	4.78 b	0.95 b	3.83 b	0.73 b	12.1 b	15.6 a	2422 a
s.d.	5.0	0.46	0.19	0.45	0.36	3.6	4.3	1071
150	45.2 b	9.29 a	2.48 a	6.80 a	1.20 a	13.6 a	13.4 b	2084 a
s.d.	6.4	0.80	1.24	1.60	0.62	4.2	4.3	1030
N fertilization
100	40.53 b	7.00 a	1.96 a	5.03 b	0.95 a	12.1 b	12.7 b	1901 b
s.d.	2.2	2.28	1.46	1.76	0.56	4.1	3.9	857
200	50.65 a	7.07 a	1.47 b	5.60 a	0.98 a	13.7 a	16.3 a	2606 a
s.d.	3.0	2.44	0.71	2.00	0.57	3.7	4.1	1130
Cultivars
Bristol	45.7a	6.93 a	1.71 a	5.22 a	0.99 a	12.3 a	14.0 a	2190 a
s.d.	5.6	2.31	1.09	1.77	0.57	4.0	4.3	1071
Lipton	45.7 a	7.14 a	1.73 a	5.41 a	0.94 a	13.5 a	15.0 a	2316 a
s.d.	5.8	2.40	1.25	2.03	0.56	3.9	4.4	1053
Infection with: Pyrenopeziza brassicae
Yes	No data	No data	No data	No data	1.35 a	14.5 a	17.4 a	2279 a
s.d.					0.45	3.7	3.7	1102
No	No data	No data	No data	No data	0.58 b	11.2 b	11.6 b	2227 a
s.d.					0.35	3.6	2.9	1024

Tukey–Kramer test; different letters indicate statistically significant differences between means at the 5% probability level, standard deviation is given in italic letters.

Effect of sulphur fertilization on different S fractions and the activity of LCD and OAS-TL

The S fertilization had a significant effect on all investigated S fractions. The total S content increased on an average by 4.5 mg g⁻¹. The S content in the control plots are an indication of latent S deficiency with a value of 4.8 mg g⁻¹ S, therefore, the plants showed no symptoms of severe S deficiency, but the S content was below the value for a maximum yield. The fertilized plots were sufficiently supplied with S (Schnug and Haneklaus, 1998). The sulphate content in the leaves increased with S fertilization by 1.5 mg g⁻¹ and the organic S fraction by 3.3 mg S g⁻¹. The GSH and cysteine content increased significantly, too, by S fertilization, but these fractions contributed little to the increase of the organic S (0.02 mg g⁻¹ cysteine-S, 0.05 mg g⁻¹ GSH-S), which mainly resulted from increasing protein contents.

It could be demonstrated that, in plants which were fertilized with S, the LCD activity was significantly lower than without additional S fertilization. The OAS-TL on the other hand was not significantly influenced by S fertilization (Table 1) but also tended to a higher activity in unfertilized plots. Therefore, in plants which showed S deficiency, the activity of the cysteine synthesizing enzyme and the activity of the cysteine catabolizing enzyme was higher than in plants, which were supplied with sufficient S.

Effect of nitrogen fertilization on different S fractions and the activity of LCD and OAS-TL

Not only the S fertilization, but also the N fertilization, had a significant effect on the investigated S fractions; with the higher N dose the sulphate content in plant material decreased while the organic S pool increased, indicating that with higher N supply more proteins were metabolized and, therefore, more sulphate was converted into proteins. This process was also reflected by the enzyme activities: the OAS-TL activity significantly increased with N fertilization and, consequently, more cysteine was metabolized. On the other hand, the activity of the cysteine-catabolizing enzyme (LCD) also increased with N fertilization (Table 1).

Effect of infection with P. brassicae on different S fractions and the activity of LCD and OAS-TL

The influence of fungal infections on the investigated parameters is also summarized in Table 1: the cysteine content increased in the leaf disc samples 2.3-fold when a visible infection with P. brassicae was discovered, and GSH increased 1.3-fold. Both metabolites were metabolized to a higher extent in leaf areas, which were obviously damaged by the pathogen. Keeping in mind that OAS-TL is catalysing the cysteine synthesis, and cysteine being the substrate for degradation by LCD, fungal infection can be expected to influence both enzyme activities. While the LCD activity significantly increased with infection with P. brassicae, the OAS-TL activity increased only slightly and not significantly in the infected tissue.

In Fig. 1 the LCD activity was plotted against the S content of the plants and the infected plants showed a distinctly higher enzyme activity with a mean value of 17.4, whereas uninfected plants had a mean activity of only 11.6 nmol H₂S mg⁻¹ protein min⁻¹.

Fig. 1.

Relationship between total sulphur content in young, fully differentiated leaves of oilseed rape at stem elongation and L-cysteine desulphydrase (LCD) activity in relation to infection with P. brassicae. [Note: continuous lines=mean values for LCD in infected (upper) and non-infected (lower) leaf discs (see Table 1); dashed lines=LSD_5%].

Correlations between the different S fractions and the enzyme activities

Correlations between different S compounds and enzyme activities are shown in Fig. 2. Weak, but significant relationships were found between the different S fractions investigated and the LCD activity. Close relationships were only found for the total S content and the organic S and sulphate, respectively, and between cysteine and GSH.

Fig. 2.

Correlation coefficients (r) for the relationships between N and S contents, different sulphur fractions, and the enzyme activities in young, fully differentiated leaves and leaf discs of Brassica napus L. [LCD, L-cysteine desulphydrase activity (nmol H₂S mg⁻¹ protein min⁻¹); OAS-TL, O-acetyl-L-serine(thiol)lyase activity (nmol cys mg⁻¹ protein min⁻¹); Cys, cysteine, GSH, glutathione; ns, non-significant; and significant at ^*P<0.05, ^**P<0.01, and ^***P<0.001].

There was a negative relationship between total S and sulphate-S with LCD while GSH and cysteine were positively correlated with the LCD activity, revealing that higher cysteine and GSH contents were related to a higher LCD activity. The N content in the plant material showed a highly significant positive correlation with the LCD activity and a weak correlation with OAS-TL. Both enzymes were positively related to each other with a correlation coefficient of r=0.47.

Discussion

It was the aim of this work to investigate if there is a relationship between the S nutritional status of the crop, the potential of the plant to release H₂S, and fungal infections. The results clearly showed that correlations existed between all investigated S fractions and the S nutritional status. The correlations (Fig. 2) were mostly weak but highly significant. It is important to mention that these are results from a field experiment where there are always a great number of influencing factors. There are several diseases, insects, grazing animals like roe deer, and climatic conditions, which are influencing the crops. In pot experiments under controlled conditions such correlations are often much closer (De Kok et al., 1981; Schnug et al., 1995b). For a field experiment, weak but significant correlations are of high relevance to describe and understand the relationships between S fractions and metabolic processes.

With increasing S supply more free sulphate was available that could be incorporated into organic S, such as cysteine and GSH. Despite the very low proportion of cysteine and GSH from the total organic S pool, both compounds seemed to be very important in the process of SIR because the contents of cysteine and GSH significantly increased when the plant material was visibly infected with P. brassicae. An increase in the pool of GSH has been measured in response to very different environmental stress factors like chilling, heat shock, pathogen attack, active oxygen species accumulation, air pollution, or drought (Dhindsa, 1991; Kocsy et al., 1996; May et al., 1996, 1998; Nieto-Sotelo and Ho, 1986; Sen Gupta et al., 1991).

Cysteine is the first stable organic S compound which is formed in the metabolism of S, and it is the precursor for all other S-containing metabolites in the plant such as methionine, glucosinolates, and GSH (Warrilow and Hawkesford, 1998). GSH has an important role in acting as a mobile pool of reduced S in the regulation of plant growth and development, and as an antioxidant in stress responses (Lamoureux and Rusness, 1993; May et al., 1998; Noctor et al., 1998). Therefore, both are probably rapidly changing pools and act as the source for the metabolism of other S-containing compounds, which are important in SIR. Cysteine is the substrate of the LCD and, therefore, a rising S supply increased the substrate availability for the enzyme. However, while the results clearly showed an increase in LCD activity with fungal infection, S fertilization led to a decrease in the activity of LCD.

The OAS-TL activity was neither influenced by S nutrition, nor by fungal infection. Therefore OAS-TL is more likely to be regulated by the N assimilatory pathway because the OAS-TL activity significantly increased with N fertilization (Table 1) and the N content of the plant material (Fig. 2), and, consequently, more cysteine was metabolized. On the other hand, the activity of the cysteine-decomposing enzyme (LCD) also increased with N fertilization (Table 1), probably to prevent the plant from a too high and toxic cysteine pool. This is also reflected by a highly significant correlation between the N content of the plant material and LCD activity (Fig. 2). Several studies established regulatory interactions between assimilatory sulphate and nitrate reduction in plants (Brunold, 1993; Koprivova et al., 2000; Yamaguchi et al., 1999). The two assimilatory pathways are interrelated; deficiency of one nutrient represses the other pathway. OAS, the precursor of cysteine, plays an important role in the regulation of sulphate uptake and assimilation. OAS seems to be limiting for cysteine synthesis in the presence of excess sulphate (Rennenberg, 1983). By comparison, OAS accumulates during S starvation and may thus act as a signal of the sulphur status (Kim et al., 1999). OAS acts most probably as a transcriptional regulator, since higher OAS contents strongly increased mRNA levels of adenosine-5′-phosphosulphate reductase (APR), sulphite reductase, chloroplastic OAS-TL, and cytosolic serine acetyltransferase (Kopriva and Koprivova, 2003).

Both enzymes, OAS-TL and LCD showed a positive correlation (Fig. 2) in infected as well as uninfected leaf discs. Burandt et al. (2001) found an inverse relationship of both enzyme activities for different genotypes of oilseed rape. These crops received the same rate of S fertilization but differed in their susceptibility to different fungal diseases. This stresses the significance of genetic differences and putatively involved modifications of S metabolism.

The fact that the activity of OAS-TL and LCD were higher in S-deficient plants indicates that, under conditions of S deficiency, S metabolism is activated and the participating enzymes are up-regulated. There are two possible explanations: the deficient plants were more susceptible to fungal diseases and, therefore, they increased metabolic pathways which were involved in plant protection. The second explanation could be that the S-deficient plants already had a stronger fungal infestation and the mechanisms of S-induced resistance were activated. Therefore the release of H₂S can be a mechanism of protection to prevent a fungal attack, or the answer to a fungal attack, or perhaps both mechanisms work at the same time.

The data in Fig. 1 clearly reveal that the activity of LCD was more strongly influenced by the infection status of the crops than by S nutrition. The LCD activity increased by about 50% due to infection with P. brassicae. The fact that OAS-TL was not significantly up-regulated while the product of the reaction, cysteine, increased strongly, probably shows that the enzyme activity was high enough to allow a fast turnover from sulphate to cysteine. These results suggest that OAS-TL is not actively increasing the H₂S release with infection, but can only participate in an increasing H₂S release in a passive way. By contrast, LCD seemed to be an enzyme whose activity is directly induced by an infection with P. brassicae, and therefore a higher H₂S release due to a higher LCD activity is possible. This is a strong hint that the evolution of H₂S could be an important strategy of the plant to combat a fungal attack. The positive relationship between OAS-TL and LCD indicates that the activity of OAS-TL is also increasing after fungal infection, but not as a direct result of the infection, rather as a reaction to the activity of LCD which is consuming cysteine, the product of the OAS-TL reaction.

The finding that LCD activity was significantly increased in infected plant tissue is a strong indication that the release of H₂S, is correlated with the fungal infestation of the crop. Therefore, the mechanism of H₂S release as a mechanism to increase the natural resistance of the crops against fungal infestations seems to be an important part of the SIR. In addition, the cysteine content increased by more than 100% and GSH by about 30% in infected plant tissue. The influence of the fungal infection on the cysteine and GSH content and LCD activity was even stronger than that of the S nutritional status of the crop, which also had a significant positive effect on the different S fractions. These results support the concept of SIR. With a better S supply more cysteine and GSH are metabolized and, therefore, the potential of SIR is increasing.

In a further step, it will be necessary to measure the evolution of H₂S from living plants in relation to the enzyme activities in order to discover if the H₂S evolution of the crop is a strategy to combat a fungal attack.

The work was financially supported by the German Research Foundation (DFG) in the sulphur research group 383, and by funds of the chemical industry, Germany.