Lessons from investigation of regulation of APS reductase by salt stress

Abstract

Exposure to high salinity affects plant ion homeostasis, water relations and results in oxidative stress. Therefore, various processes are induced as salt stress response including antioxidative defense systems. The tripeptide glutathione has a prominent position among the metabolites involved in such stress defense. Glutathione synthesis is dependent on supply of cysteine and thus on the assimilation of sulfate. We have investigated how the key enzyme of sulfate assimilation, adenosine 5′phosphosulfate (APS) reductase is regulated by salt stress in Arabidopsis roots. Using Arabidopsis mutants in various signaling pathways we aimed to identify the signaling cascade leading to regulation of APS reductase by NaCl. We found the enzyme to be regulated by a complex signaling network on transcriptional and post-transcriptional levels with responses of mRNA accumulation and enzyme activity largely uncoupled. Here we want to share the important lessons we have learned from this investigations.

Sulfur metabolism plays an important role in plant stress response.^1,2 Among the sulfur containing metabolites, the tripeptide glutathione (GSH) is the most important part of this response due to its function in the ascorbate glutathione cycle in detoxification of reactive oxygen species (ROS).³ As salt stress is connected with ionic stress, osmotic stress and production of ROS it is not surprising that it also triggers increase of GSH levels.^4–6 GSH synthesis is dependent on the availability of the constituent amino acids cysteine, glutamate and glycine and as such, on sufficient reduction of sulfate.⁷ The key control point of sulfate assimilation is the reduction of activated sulfate, APS, to sulfite by APS reductase.⁸ Accordingly, APS reductase is highly regulated in a demand driven manner, i.e., its activity increases when concentration of reduced sulfur compounds is low, e.g., during sulfur starvation or depletion of GSH, and decreases when excess reduced sulfur is available.^8–10 In vast majority of reports, changes in APS reductase activity correlated well with changes in mRNA and protein accumulation, indicating a simple transcriptional regulation of the corresponding genes,^8–10 with a rare occurrence of additional level of post-transcriptional redox regulation.¹¹ Whereas the physiological responses of APS reductase to different environmental stimuli are well known, very little is known about the molecular mechanisms including signaling pathways and transcription factors.

To find components of its regulatory circuit we chose salt stress as the environmental stimuli and analyzed the regulation of APS reductase in roots of Arabidopsis mutants in different signaling pathways.⁶ In that report we showed that in roots of wild-type (WT) Arabidopsis, APS reductase activity, protein accumulation, and mRNA levels were increased 3-fold after 5 hours of exposure to 150 mM NaCl. Analysis of various mutants in hormone signaling revealed that the regulation was ABA insensitive, however, otherwise the response of APS reductase activity was uncoupled from the mRNA response. Only treatment with EGTA to disrupt Ca²⁺ signaling prevented the increase in both mRNA and enzyme activity upon the salt treatment. In most of the mutants the APS reductase activity was not increased upon the salt treatment or even decreased, despite the increased mRNA accumulation (see Table 1 for summary).⁶ The conclusion that the increase in APS reductase is ABA independent was the only simple one from our study. Surprisingly, functional signaling through all major transduction pathways, including not only stress signaling hormones jasmonate, salicylate, ethylene and NO, but also cytokinins, auxin and giberellin seems to be necessary to induce APS reductase both on mRNA and activity level. The strict correlation between APS reductase activity and protein accumulation revealed a novel translational regulation of this enzyme, mechanism of which is, however, still unknown.

Table 1.

Response of APS reductase mRNA level and activity to salt stress in roots of WT Arabidopsis and mutants in signaling pathways

Signaling pathway	Mutant/treatment	mRNA	Activity
WT	Col-0, Ler	↑	↑
ABA	aba1, aba2, abi1, abi2	↑	↑
Ca2+	EGTA	=	=
jasmonate	jar1-1	↑	=
salicylate	NahG, npr1–2	↑	↓
ethylene	etr1–3, ein2-1	↑	=
NO	PTIO	↑	↓
auxin	axr1–3, tir1-1	↑	↓
cytokinin	ahk4, CKX	↑a	=
gibberelin	gai	=b	↑

^aAPR2 was not induced;

^bonly APR3 was increased.

Traditionally, stress signaling pathways were elucidated by analysis of transcript levels and promoter-reporter constructs.^12–14 However, recently it became apparent that post-transcriptional and post-translational control plays an important role as well. This has been demonstrated for several pathways of the salt signaling. For example, overexpression of SOS1 under control of constitutive promoter did not lead to accumulation of the mRNA in control plants but only under salt stress.¹⁵ Genes involved in proline biosynthesis are either under control of miRNA¹⁶ or undergo post-transcriptional regulation.¹⁷ Phosphorylation and ubiquitination are well recognized processes in salt stress signaling.^18,19 In this context the findings on APS reductase seem not so surprising anymore.

Nevertheless, transcript analysis and especially microarrays are still the methods of choice for dissecting salt stress signaling.²⁰ However, our findings⁶ and those of others^21–23 raise questions about usefulness of concentration on transcript data only for understanding biological processes. The public domain is rich in repositories of microarray experiments and programs for their mining, e.g., the GENEVESTIGATOR,²⁴ NASCArrays²⁵ or eFP browser.²⁶ However, while these data are excellent resource in pointing to direction, they are not necessarily reproducible in one's system.⁶ Even more importantly, the responses of genes to treatments are not always correlated with the responses of the encoding proteins and enzyme activities. This has been demonstrated in reports on individual genes and pathways^6,17 as well as on a much larger scale using -omics technologies.^21–23 Remarkably, several of these reports were concerned with the effects of salt stress or biotic stress. For example, very recently, Jiang et al.,²¹ undertook a proteomics analysis of Arabidopsis roots treated with NaCl, detected 215 differently abundant spots which resulted in identification of 86 proteins. Among these, many known stress-related proteins were present, as well as proteins involved in metabolism, protein synthesis and signal transduction.

Comparison of these results with transcriptomics data from the GENEVESTIGATOR database, which contains a microarray experiment under the same conditions, revealed a very poor correlation (p < 0.11) of the two sets of data.²¹ Similarly, Jones et al.,²² analysed the changes in protein abundance upon infection with bacterial pathogen. By comparing the proteome data with transcriptomics data from the same material²⁷ they found that most of the proteins were affected before any changes in transcriptome occurred.^22,27 Both reports confirm on a large scale the frequent occurrence of post-transcriptional regulation in plant response to stress. However, these observations can be taken one step further when the report of Gibon et al.,²³ is taken into account. These authors developed a platform for high throughput measurements of 23 enzyme activities. Comparison of circadian changes in these activities with corresponding variations of transcript levels again revealed a low correlation with very different amplitudes of changes and various delays in peaks of activities compared to mRNAs. Again a large contribution of posttranscriptional and post-translational regulation has been confirmed. Altogether these studies show that despite a high usefulness of the transcript data, when addressing regulation of biological processes, they have considerable limitations.

There is second highly unexpected lesson we learned from the study on regulation of APS reductase by salt. Had we analyzed the WT only, our conclusion would be a simple transcriptional regulation as in many cases before. Only after analysis of the signaling mutants it became evident that the increase of mRNA levels, protein accumulation and enzyme activity after salt stress is not so straightforward but dependent on many factors. Their involvement is not evident in the WT but only when some of these factors are not functional due to the defect in the signaling leading to its regulation. How frequent such complex regulations are, remains to be seen. There are however several parallels to our findings, e.g., the highly complex network regulating Arabidopsis response to Botrytis infection,²⁸ the JERF1 transcription factor in tomato, which is induced by NaCl, ethylene, jasmonate and ABA²⁹ or the integration of stress responses via the DELLA proteins.³⁰ It seems therefore, that such complex signaling networks might be a rule rather than exception, especially for processes regulated by stress.

Abbreviations

APS

adenosine 5′phosphosulfate

GSH

glutathione

Addendum to: Koprivova A, North KA, Kopriva S. Complex signaling network in regulation of adenosine 5′-phosphosulfate reductase by salt stress in Arabidopsis roots. Plant Physiol. 2008;146:1408–1420. doi: 10.1104/pp.107.113175.