Light regulation of assimilatory sulphate reduction in Arabidopsis thaliana


*For correspondence (fax +41 31 332 20 59;


Adenosine 5′-phosphosulphate reductase (APR) is considered to be a key enzyme of sulphate assimilation in higher plants. We analysed the diurnal fluctuations of total APR activity and protein accumulation together with the mRNA levels of three APR isoforms of Arabidopsis thaliana. The APR activity reached maximum values 4 h after light onset in both shoots and roots; the minimum activity was detected at the beginning of the night. During prolonged light, the activity remained stable and low in shoots, but followed the normal rhythm in roots. On the other hand, the activity decreased rapidly to undetectable levels within 24 h of prolonged darkness both in shoots and roots. Subsequent re-illumination restored the activity to 50% in shoots and to 20% in roots within 8 h. The mRNA levels of all three APR isoforms showed a diurnal rhythm, with a maximum at 2 h after light onset. The variation of APR2 mRNA was more prominent compared to APR1 and APR3. 35SO42– feeding experiments showed that the incorporation of 35S into reduced sulphur compounds in vivo was significantly higher in light than in the dark. A strong increase of mRNA and protein accumulation as well as enzyme activity during the last 4 h of the dark period was observed, implying that light was not the only factor involved in APR regulation. Indeed, addition of 0.5% sucrose to the nutrient solution after 38 h of darkness led to a sevenfold increase of root APR activity over 6 h. We therefore conclude that changes in sugar concentrations are also involved in APR regulation.


Light plays an essential role in plant development, growth and reproduction. Correspondingly, many enzymes are light-inducible in etiolated seedlings or during daily dark–light transitions ( Kloppstech 1985; Kloppstech 1997; Nelson et al. 1984 ; Terzaghi & Cashmore 1995). Not only is carbon assimilation dependent on photosynthesis, but also the assimilation of nitrogen and sulphur. Therefore, it is not surprising that nitrate reductase (NR) is strongly light-regulated. Light induces transcription of the NR gene and enables the dissociation of an inhibitor and subsequent dephosphorylation of a regulatory serine. This two-step activation process is triggered by dark–light transitions, but also by other effectors, and is reversed when light is switched off ( Huber et al. 1992 ; Huber et al. 1996 ). Sucrose can replace light in inducing the transcription of the NR gene in dark-adapted or etiolated Arabidopsis thaliana plants, indicating that sugars are mediators of light responses of this gene ( Cheng et al. 1992 ). Sugars and nitrogen metabolites regulate NR expression and activity also in tobacco and Nicotiana plumbaginifolia dettached leaves ( Morcuende et al. 1998 ; Vincentz et al. 1993 ).

Sulphate assimilation is also influenced by light. ATP-sulphurylase (ATP-S) activity increased with irradiation in oat, barley and maize, and decreased when inhibitors of photosynthetic electron transport were added to the nutrient solution ( Passera et al. 1989 ). In A. thaliana, the mRNA levels of APS kinase, sulphite reductase, O-acetylserine (thiol) lyase and serine acetyltransferase were several times higher in green leaves than in etiolated tissues ( Hell et al. 1997 ). The activity of adenosine 5′-phosphosulphate (APS) reductase (APR), formerly named APS sulphotransferase ( Bick & Leustek 1998; Wray et al. 1998 ), is light-induced in Lemna minor, but feeding with O-acetyl- l-serine (OAS), the precursor of cysteine, resulted in similar induction in the dark ( Neuenschwander et al. 1991 ). A diurnal rhythm of APR with maximal activity during the light period was observed in maize ( Kocsy et al. 1997 ). It was not clearly shown, however, whether this rhythm was strictly light-dependent or if an internal clock mechanism was involved.

The purpose of this work was to analyse the light regulation of APR. This enzyme is considered to be the key enzyme of the assimilatory sulphate reduction ( Brunold 1993) because of its strategic position at the beginning of the pathway. It is highly regulated by different environmental factors such as heavy metal stress ( Rüegsegger et al. 1990 ) and chilling ( Brunner et al. 1995 ). cDNAs representing three isoforms of this enzyme have been cloned from A. thaliana ( Gutierrez-Marcos et al. 1996 ; Setya et al. 1996 ). We analysed the APR activity, mRNA levels and protein accumulation during the normal day–night regime, under constant light or dark conditions, and upon re-illumination after a prolonged dark period. To compare directly the flux through the sulphate assimilatory pathway at different times of day and illumination status, we determined the in vivo incorporation of [35S]sulphate into cysteine, γ-glutamylcysteine (γ-EC), glutathione (GSH) and proteins.


Diurnal rhythm of APR activity

Maize was shown to have strong diurnal variation in APR activity ( Kocsy et al. 1997 ). In order to examine whether this was a specific feature of C4 metabolism, we measured the APR activity in a C3 plant, Arabidopsis thaliana, every second hour during two consecutive days. Strong diurnal fluctuations in APR activity were detected in both shoots and roots ( Fig. 1a), the maximum after 4 h from light onset being approximately twice as high as the minimal activity at the beginning of night. Surprisingly, the increase in APR activity began before the light onset in shoots and during the first 4 h of the night in roots. The differences in activities between 20 and 24 h in shoots and between 10 and 14 h in roots were statistically significant at P≤ 0.05.

Figure 1.

APR activity in A. thaliana during different illumination regimes.

(a) APR activity was measured in A. thaliana shoots and roots extracts during 2 days of normal dark and light periods, indicated by the black and white bars (closed symbols) or extended light (open symbols). Mean values ± standard deviations of four to five measurements are shown. Time points 0 and 24 h correspond to 08:00 h, just before light onset.

(b) APR activity during extended darkness (closed symbols) and re-illumination after 38 h (open symbols). Time point 0 h corresponds to the beginning of the usual night period at 18:00 h; the arrow indicates the start of re-illumination.

Effect of prolonged light and darkness on APR activity

In prolonged light, APR activity in shoots remained at a constant but low level ( Fig. 1a). Interestingly, APR activity in the roots changed similarly as in a normal night. When darkness was prolonged after the end of the normal night period, APR activity quickly decreased to a hardly detectable level after an additional 24 h ( Fig. 1b). The decrease was more rapid in roots than in shoots. When plants were re-illuminated after 38 h of darkness, APR activity increased to 50% of maximal activity in shoots and to about 20% of maximal activity in roots within 8 h. In contrast, the activity remained at a very low level in control plants kept in the dark ( Fig. 1b).

To determine whether the induction of APR activity in dark-adapted plants was mediated by phytochrome, plants kept in the dark for 38 h were treated with a red (660 nm) or white light flash of 10–30 μmol m–2 sec–1 at 660 nm, as described by Mustilli & Bowler (1997). Neither red nor white light led to any detectable increase of APR activity after a further 4 or 8 h in the dark (data not shown), indicating that the phytochrome system was not involved.

Expression analysis

Northern blot analysis of RNA isolated from leaves during the day/night cycle revealed a strong diurnal fluctuation of the APR2 mRNA level, with a maximum at 2 h of the light period ( Fig. 2a, left panel). Corresponding to the results from activity measurements, the APR2 mRNA level started to increase before the onset of light, indicating an internal light-independent stimulus. The diurnal variation of APR1 mRNA level was weaker than that of APR2, but the increase in its mRNA level also started during the dark period. The level of APR3 transcript in leaves was much lower than that of APR1 and APR2, and the diurnal rhythm was not so obvious. In roots, the transcript levels of all three APR isoforms showed less pronounced changes than in shoots; however, as in shoots, the highest mRNA level was detected 2 h after light onset for all three isoforms ( Fig. 2a, right panel).

Figure 2.

Northern blot analysis of APR expression.

(a) Total RNA (10 μg) from shoots (left panel) or roots (right panel) isolated at different times during the standard 10/14 h day/night rhythm was probed with 32P-labelled cDNA fragments of APR1, APR2 and APR3 from A. thaliana. Ethidium-bromide stained RNA is included as control of RNA loading and intactness (RNA).Total RNA isolated from A. thaliana at different times during extended darkness (b), re-illumination after 38 h darkness (c), and extended light (d) was probed with 32P-labelled cDNA fragments of APR1, APR2 and APR3. Ethidium-bromide stained RNA is shown as a control of loading and RNA intactness (RNA). The upper panels present RNA from shoots, the lower ones from roots. Time point 0 h corresponds to the beginning of the corresponding illumination regime, which is indicated by the white (light) or black (dark) bars. The grey bars correspond to (b) the day and (d) the night period of the standard 10/14 h day/night rhythm.

In shoots of plants kept in extended darkness, the mRNAs of APR1 and APR2 increased at 14 h, in agreement with the standard day/night rhythm, but afterwards all three isoforms decreased rapidly to almost undetectable levels within an additional 24 h ( Fig. 2b, upper panel). Re-illumination after 38 h of darkness induced an increase of all three APR mRNA levels. APR2 and APR3 mRNA responded very quickly, reaching maximum levels after 2 h of light, whereas APR1 mRNA accumulated gradually over the entire period ( Fig. 2c). In roots, a similar sharp drop in APR mRNA levels during prolonged darkness, as well as an accumulation after re-illumination, also occurred ( Fig. 2b,c, lower panel). All three APR mRNA levels were similarly regulated in roots, reaching a maximum 2 h after re-illumination and decreasing afterwards ( Fig. 2c, lower panel).

In extended light, APR1 and APR3 mRNA levels started to accumulate in shoots after 14–20 h of illumination and remained higher than during the normal day/night period ( Fig. 2d). In roots, the mRNA levels for APR1 remained stable during the whole 32 h of extended light; APR2, however, showed a transient increase in steady-state level after 4 h of prolonged light ( Fig. 2d, lower panel), then decreased to a lower, constant level, and APR3 slowly decreased.

Western blot analysis revealed that during normal day/night rhythm, APR protein accumulation correlated with the APR activity ( Fig. 3a). In shoots and roots of plants kept in extended darkness and re-illuminated after 38 h, the APR amount correlated with activity and mRNA levels both in roots and shoots ( Fig. 3c). However, plants kept in extended light accumulated APR protein in shoots ( Fig. 3d) although the APR activity remained constant under these conditions.

Figure 3.

Western blot analysis of APR.

Crude protein extracts (10 μg) from shoots and roots of A. thaliana were resolved on 12% SDS–PAGE gels, transferred onto nitrocellulose membranes and probed with antibodies against recombinant APR2. Extracts collected during the standard 10/14 h day/night rhythm (a), extended darkness (b), re-illumination after 38 h darkness (c), and extended light (d) were analysed. The upper panel presents material from shoots, the lower one from roots. Time point 0 h corresponds to the beginning of the corresponding illumination regime, which is indicated by the white (light) or black (dark) bars. The halftone bars correspond to (b) the day and (d) the night period of the standard 10/14 h day/night rhythm.

Taken together, these results show that APR activity is regulated at the level of transcription and that no post-translational regulation was detectable.

Regulation of APR activity and expression by sucrose

The results of APR activity and expression analysis during the different light regimes strongly implied that a light-independent signal contributed to the regulation. Therefore, plants kept for 38 h in darkness were fed with 0.5% sucrose in darkness, and APR activity, mRNA and protein accumulation were measured. The sucrose feeding had no effect on APR activity in shoots (data not shown). In roots, the APR activity rapidly increased up to 6 h and slowly declined afterwards ( Fig. 4a). The rapid induction of APR mRNA levels ( Fig. 4b) and protein accumulation in roots ( Fig. 4c) were in agreement with a presumable transcriptional regulation of APR activity by sucrose.

Figure 4.

Effect of sucrose on APR in roots.

A. thaliana plants were pre-incubated in darkness for 38 h. (a) APR activity was measured in root extracts of control plants continuing the dark treatment (○) and plants fed with 0.5% sucrose (w/v) in the dark (●). Mean values ± standard deviations of four measurements are shown. Time point 0 corresponds to 38 h pretreatment in darkness. (b) Total RNA (10 μg) from roots isolated at indicated times during 8 h of feeding with 0.5% sucrose in the dark was probed with 32P-labelled cDNA fragments of APR1, APR2 and APR3 from A. thaliana. Ethidium bromide-stained RNA was included as control of RNA loading and intactness (RNA). (c) Root protein extracts (10 μg) were resolved on 12% SDS–PAGE gel, transferred onto nitrocellulose membrane and probed with antibodies against recombinant APR2.

In vivo flux through the sulphate assimilation pathway

In addition to APR activity measurements and expression analysis, it was important to determine whether sulphate assimilation in vivo was also regulated. The plants were fed with 35SO42– in the nutrient solution and the amount of radioactivity incorporated into thiols and proteins at different times of day and illumination status were measured ( Fig. 5). Feeding periods, as indicated in Fig. 5, were chosen to represent high (L1, D1: 2–6 h after the beginning of the illumination) and low APR activity (L2: 10–14 h after the beginning of illumination; D2: the first 4 h of the night period). 35S incorporation during the last 4 h of the dark period (D3) was also measured in order to find out whether the increase in APR activity at this time was reflected by a higher flux through the pathway.

Figure 5.

Incorporation of 35S into thiols and proteins during 4 h feeding with [35S]sulphate.

A. thaliana plants that were 4½-weeks-old were fed with 35SO42– in the nutrient solution for 4 h at different times of the day in light (L1, L2) or dark (D1, D2, D3) as indicated at the bottom of the figure. Labelled sulphur in SO42–, cysteine, γ-EC, GSH and protein fractions in shoots (a) and roots (b) is shown. The numbers over the bars indicate the percentage of labelled thiols relative to total thiols. Values indicated by different letters are different from each other at P≤ 0.05. APR activity at the beginning and end of the corresponding feeding period is shown. n.d., not determined.

The amounts of labelled SO42– in the shoots indicated that the transport of [35S]SO42– from the roots to the shoots was about twice as higher in the two light periods as in the dark periods ( Fig. 5a). 35S-labelled cysteine and γ-EC were only present in very low amounts compared to GSH and proteins, and did not differ significantly between the different treatments. Glutathione was labelled most during the prolonged light period (L2), and its labelling during L1 was also significantly higher than during the three dark periods. Incorporation of 35S into proteins was 5–10 times higher in the light than in the dark periods. Comparing the radioactivity incorporated into GSH and proteins during the two light periods with the corresponding APR activity showed no correlation. It should be pointed out, however, that the specific radioactivity of cysteine was about 50% higher in L2 than L1, indicating a corresponding difference in the sulphate precursor pools and an under-estimation of the flux of sulphur into GSH and proteins during L1 compared to L2 of about 50%. A corresponding reasoning for the dark labelling periods would indicate an under-estimation of the sulphur flux into GSH and proteins during D1 compared to D2 and D2 compared to D3 of about 50% and 100%, respectively.

In roots, no significant differences in contents of labelled cysteine and γ-EC were observed between the different treatments ( Fig. 5b). Similarly to shoots, the highest incorporation of 35S in GSH was found in plants labelled during the L2 period, whereas there was no significant difference between the first light period and the different dark periods. The proteins were labelled similarly during the dark (D1, D2 and D3), in the light more labelling was detected during L2 than L1. As in the shoots, the specific radioactivity of cysteine was lower in L1 than in L2, indicating a corresponding difference in the sulphate pools as in shoots.


Nitrate and sulphate assimilation pathways in plants have many common features and are also highly coordinated ( Brunold 1993). APS reductase and nitrate reductase represent the most controlled steps in these pathways and may thus be regulated, at least partially, by similar mechanisms. NR activity and mRNA levels in leaves of various plants undergo diurnal changes ( Lillo 1994). In Nicotiana tabacum leaves, NR mRNA, protein and activity decreased to very low levels within 56 h of continuous darkness, and were rapidly restored after re-illumination ( Deng et al. 1990 ). Similarly, the APR activity and mRNA levels were increased in light and decreased in darkness. The pre-dawn increase of APR activity mRNA levels indicates involvement of an endogenous clock mechanism. However, since the diurnal changes of APR activity and mRNA levels did not continue under extended darkness or light, regulation by endogenous rhythm is very probably not involved. Interestingly, the induction of APR activity by re-illumination of dark-adapted plants was not phytochrome-dependent.

The diurnal variations of the APR activity and mRNA levels detected in roots indicate that signals in addition to light trigger this enzyme regulation. On the basis of previous results with O-acetylserine ( Neuenschwander et al. 1991 ) and because NR is known to be stimulated at the transcription level by sugars ( Lillo 1994; Vincentz et al. 1993 ), amino acids or carbohydrates were considered as possible signalling substances. Indeed, sucrose induced root APR mRNA and protein synthesis as well as activity in roots of plants kept in prolonged darkness. The sucrose-induced increase of APR activity was even quicker than that measured in shoots after re-illumination of dark-adapted plants. The corresponding effect of externally applied sucrose on the activity and mRNA accumulation of APR and NR, key enzymes of assimilatory sulphate and nitrate reduction, adds an interesting facet to the coordinated regulation of these two pathways ( Brunold 1993).

It has been shown in different systems ( Brunold 1993; Schmidt & Jäger 1992) that APR activity was regulated according to the needs of plants for reduced sulphur, indicating a high metabolic flux control coefficient for this enzyme ( Fell 1997). In the results presented here, labelling of proteins and GSH from 35SO42– in the light was at comparable levels in shoots after two feeding periods (L1 and L2), even though the average APR activity was 140% higher during L1 than during L2. This result may partly be explained by the 50% higher specific radioactivity of cysteine in L2 than in L1 ( Fig. 5), which may be due to a corresponding difference in specific radioactivity of the sulphate precursor pools. Using this difference to correct the measured values leads to a calculated flux of sulphur into GSH and proteins which is 14% higher during L1 than during L2. Taking the great difference in APR activity between L1 and L2 into account, this small flux increase would indicate a low flux control coefficient of APR under the conditions applied. In roots, labelling of cysteine indicates a specific radioactivity of the sulphate precursor pools that is 50% higher during L2 than during L1. Even if this difference is taken into account to correct the flux into GSH and proteins during L1, the calculated flux is still approximately 20% smaller than that during L2, even though average APR activities were at comparable levels. Since it has been shown before under different conditions that in addition to APR the availability of OAS is of central importance for regulating the flux through assimilatory sulphate reduction ( Neuenschwander et al. 1991 ; Smith et al. 1997 ; Koprivova et al. unpublished results), it is tempting to assume that OAS was present at higher levels during L2 than L1.

The diurnal oscillations of APR mRNA and protein amounts, their decrease in extended darkness and increase after re-illumination were mostly consistent with the activity changes of the enzyme. The light effects on APR activity can thus be tracked to the regulation of mRNA expression. Only in extended light was the amount of APR protein detected by Western analysis higher than expected from activity measurements. This discrepancy can be explained by assuming that part of the APR protein was inactivated. We have recently determined that APR contains an iron–sulphur cluster as a cofactor (Kopriva et al. unpublished results), which could be sensitive to reactive oxygen species formed in extended light. A parallel loss of Fe–S clusters and enzyme activity has been demonstrated previously ( Rouault & Klausner 1996).

Interestingly, the different APR isoforms showed different time courses in their diurnal fluctuations and in the induction after re-illumination of dark-adapted shoots. APR1 and APR3 isoforms showed a similar expression pattern which was distinctly different from APR2. This observation agrees well with relationships of the cDNA sequences and the proposed evolution of APR genes ( Bick & Leustek 1998).

Taken together, we conclude that in higher plants (i) APR mRNA, APR activity and in vivo sulphate reduction change with a diurnal rhythm, (ii) sulphate assimilation also takes place during the dark period, and (iii) sucrose feeding positively affects APR mRNA and activity in roots.

Experimental procedures

Plant cultivation and treatment

Seeds of Arabidopsis thaliana, var. Columbia, were germinated in plastic pots filled with small moistened balls (2–6 mm in diameter) of burned clay (‘Blähton’, Migros, Switzerland). The pots were held in trays containing Hentschel nutrient solution ( Hentschel 1970). The plants were grown under a day/night cycle of 10 h/14 h and a light intensity of 115–160 μmol sec–1 m–2. Average temperature was 25–27°C during the day and 23°C at night. Relative humidity ranged between 40 and 60%. All experiments were performed with 4½-week-old plants. At the beginning of the last dark period before the experiments, fresh nutrient solution was applied. To measure the effect of sucrose on APR activity, plants were pre-incubated in darkness for 38 h. Afterwards the plants were transferred to a nutrient solution supplemented with 0.5% sucrose and kept in darkness for an additional 8 h.

APS reductase assay

For extractions, whole shoot or root systems of 4–8 plants were used. Shoot and root material was extracted 1:10 and 1:20 (w/v), respectively, in 50 m m NaKPO4 buffer (pH 8) supplemented with 30 m m Na2SO3, 0.5 m m 5′-AMP, and 10 m m DTE ( Imhof 1994), using a glass homogenizer. APR activity was measured in extracts as the production of [35S]sulphite, assayed as acid-volatile radioactivity formed in the presence of [35S]APS and DTE ( Brunold & Suter 1990). The protein concentrations in the extracts were determined according to Bradford (1976), with bovine serum albumin as a standard.

Isolation of total RNA and Northern blotting

Plant material was pulverized with mortar and pestle in liquid nitrogen and the RNA was isolated by phenol extraction and selective precipitation with LiCl. Electrophoresis of RNA was performed on formaldehyde–agarose gels at 120 V. RNA was transferred onto Hybond-N nylon membranes (Amersham) and hybridized with 32P-labelled total cDNA probes for APR1 and APR2 or an APR3 cDNA fragment of 600 bp from the 3′-terminus. The membranes were washed four times at different concentrations of SSC in 0.1% SDS for 20 min, the final washing step being 0.5× SSC, 0.1% SDS at 65°C, and exposed to a X-ray film (Fuji Medical RX) at –80°C for 3–8 days. These hybridization and washing conditions allowed no cross-hybridization of the APR isoforms when tested with RNA in vitro transcribed from APR cDNA clones obtained from Dr T. Leustek (Center for Agricultural Molecular Biology, Rutgers University, New Brunswick, USA). The Northern analysis was performed on two independent RNA preparations with the same results.

Western blot analysis

Protein extracts were prepared as described by Zavgorodnyaya et al. (1997) . Aliquots of 10 μg protein were subjected to SDS–PAGE and electrotransferred to nitrocellulose filter (Schleicher and Schuell, Dassel, Germany). The blots were analysed with antisera against recombinant APR2 and developed with the SuperSignal Western Blotting System (Pierce). The antisera were produced in rabbits from purified APR2 protein over-expressed in E. coli by the pET His-Tag system (Novagen). The antisera cross-reacted with the recombinant APR1 and APR3 proteins. The Western analysis was performed on two independent protein preparations with the same results.

Feeding of 35SO42– and determination of 35S in thiols and proteins

Four pots with Arabidopsis plants were fed with Hentschel nutrient solution containing 0.75 m m SO42– supplemented with 4 mCi 35SO42– for 4 h at different times of the day either in light or in dark. Shoots and roots were extracted with 0.1 m HCl containing 1 m m Na2EDTA and the extracts were centrifuged at 12 000 g for 30 min at 4°C. The thiols in the supernatant were reduced by bis-(2-mercaptoethylsulphone) (BMS) ( Bernhard et al. 1998 ) and labelled by monobromobimane as described by Newton et al. (1981) and modified by Kranner & Grill (1996). A 100 μl aliquot of each sample was separated by reversed-phase HPLC as previously described ( Rüegsegger & Brunold 1992), and fractions of 0.75 ml were collected in scintillation vials. The 35S activity was determined in a Betamatic V liquid scintillation counter (Kontron, Zürich, Switzerland). Total cysteine, γ-EC and GSH were analysed by the same HPLC system as described by Schupp & Rennenberg (1988) and modified by Rüegsegger & Brunold (1992). For measurement of 35S incorporation into proteins, these were precipitated from 200 μl of extract with 10% TCA, washed twice with 1% TCA and once with 96% ethanol, and redissolved in 400 μl 0.2 m NaOH. Radioactivity in an aliquot of the protein solution was determined using a liquid scintillation counter.

Statistical analysis

The Student–Newmann–Keuls method (SigmaStat for Windows, Version 1.0 1992–1994, Jandel Corporation) was used to determine significant differences in the enzyme activities and the contents of labelled thiols.


We thank Dr T. Leustek from the Center for Agricultural Molecular Biology, Rutgers University, New Brunswick, USA, for cDNA probes of the three APR isoforms, W. Tanner and E. Bhend for technical assistance, and H. Flückiger, Dr P. von Ballmoos and S. Jones for help in harvesting the plant material. This work was supported by Swiss National Science Foundation grant no. 3149246-96 to C.B.